Download Kin and social network structure in two populations of

KIN AND SOCIAL NETWORK STRUCTURE IN TWO POPULATIONS OF OCTODON DEGUS
By
Garrett Davis
Loren Hayes
Associate Professor
Biological and Environmental Sciences
(Chair)
Margaret Kovach
Professor
Biological and Environmental Sciences
(Committee Member)
Hope Klug
Assistant Professor
Biological and Environmental Sciences
(Committee Member)
1
KIN AND SOCIAL NETWORK STRUCTURE IN TWO POPULATIONS OF OCTODON DEGUS
By
Garrett Davis
A Thesis Submitted to the Faculty of the University of Tennessee at Chattanooga
in Partial Fulfillment of the Requirements of the Degree of
the Master of Science: Environmental Science
The University of Tennessee at Chattanooga
Chattanooga, Tennessee
December 2014
ii
ABSTRACT
A growing body of evidence showing that individuals of some species live in non-kin groups
suggests kin selection is not required in all species for sociality to evolve. Here I investigate two
populations of Octodon degus, a South American rodent which has been shown to form kin and non-kin
groups. I quantified genetic relatedness within social groups in two populations as well as social network
parameters (association, strength, and clustering coefficient) in order to determine if these aspects of
sociality were driven by kinship. I analyzed social network parameters relative to ecological conditions at
burrow systems used by individuals to determine if ecological characteristics could explain variation in
sociality. In both populations, genetic relatedness among individuals within social groups was not
significantly higher than randomly selected individuals from the background population, suggesting nonkin structure is common in degus. In both populations, I found significant relationships between habitat
characteristics of burrow systems and social network characteristics of individuals.
iii
1
ACKNOWLEDGEMENTS
I would first like to thank my advisor (Dr. Loren Hayes) and my other committee members, Dr.
Margaret Kovach and Dr. Hope Klug, for their knowledge and guidance. Additionally, I would like to
thank Dr. Luis Ebensperger, Dr. Rodrigo Vasquez, Dr. Elie Poulin, and the members of the Molecular
Ecology Lab at Universidad de Chile for their assistance with my research while in Chile. Lastly, I thank
my friends and family for their support and encouragement.
iv
TABLE OF CONTENTS
ABSTRACT………………………………………………………………………………………….………………………….iii
ACKNOWLEDGEMENTS………………………………………………………….………………………………………iv
LIST OF TABLES………………………………………………………….………………………………………………….vii
LIST OF FIGURES…………………………………………….…………………………………………………………….viii
LIST OF TERMS……………………………………………………………………………………………………………….ix
CHAPTER
I. INTRODUCTION………………………………………………………………………………………………………1
Sociality models………………………………………………………………………………………………….…1
Emlen’s theory of the family………………………………………………………………….………………2
Ecological constraints…………………………………………………………………………………….….3
Kin selection and inclusive fitness……………………………………………………………………..3
Reproductive skew……………………………………………………………………………………….…..4
Conflicting evidence……………………………………………………………………………………………...5
Trade-offs of sociality……………………………………………………………………………………….……6
Thesis objectives……………………………………………………………………………………………….....8
Study sites……………………………………………………………………………………………………………..9
Significance………………………………………………………………………………………………………….10
II. KIN AND SOCIAL NETWORK STRUCTURE IN TWO POPULATIONS OF
OCTODON DEGUS………………………………………………………………………………………..……….12
Introduction…………………………………………………………………………………………………………12
Intraspecific variation and sociality………………………………………………………………..14
Social networks……………………………………………………………………………………………...14
Objectives and study organism…………………………………………………………………….…15
Methods……………………………………………………………………………………………………..………17
Study populations……………………………………………………………………………………….….17
Ecological sampling…………………………………………………………………………………….….17
Social group determination………………………………………………………………….………..18
Social network analysis………………………………………………………………………….……...19
Genetic analysis…………………………………………………………………………………….……….20
Statistical analysis………………………………………………………………………………….……...21
Results………………………………………………………………………………………………………………22
Microsatellite variation…………………………………………………………………………………..22
Descriptive data………………………………………………………………………………………………23
Relatedness and social structure…………………………………………………………………….23
Ecology and network structure……………………………………………………………………….25
Discussion………………………………………………………………………………………………………….28
Social structure and kinship…………………………………………………………………………….28
Social networks and habitat conditions…………………………………………………………..32
Concluding remarks………………………………………………………………………………………..34
III. CONSERVATION IMPLICATIONS…………………………………………………………………………..35
REFERENCES………………………………………………………………………………………………………………...38
APPENDIX
A. SOCIAL GROUPS…………………………………………………………………………………………………48
B. GENETIC PROCEDURES…………………………………………………………………………………….…50
C. GENETIC DATA……………………………………………………………………………………………….…..53
D. ECOLOGICAL DATA…………………………………………………………………………………………….57
E. NETWORK DATA…………………………………………………………………………………………………61
F. STATISTICAL ANALYSIS…………………..…………………………………………………………………..66
VITA……………………………………………………………………………………………………………………………..69
LIST OF TABLES
2.1. Group size and within-group relatedness of social groups at Rinconada and Los
Molles 2007-2008………………………………………………………………………………………………………..24
2.2 Multiple regression statistics for individuals’ network parameters vs. weighted habitat
characteristics at Rinconada and Los Molles 2007-2008……………………………………………….28
1
2
3
4
LIST OF FIGURES
1.1 Conceptual framework for the drivers and outcomes of kin structure…………………………..………2
1.2 Geographic location of Los Molles (top) and Rinconada (bottom) in central Chile…………….…10
2.1 Scatterplot showing the statistically significant relationship between (a)
soil hardness and strength and (b) food biomass and clustering
coefficient for individuals at Rinconada in 2007-2008………………………………....………………26
2.2 Scatterplot showing the statistically significant relationship between (a) burrow
density and strength and (b) food biomass and clustering coefficient for
individuals at Los Molles in 2007-2008…………………………………………………...…………………..27
2.3 Social network structure of (a) a theoretical social group exhibiting kin structure
and (b) an actual social group from the Los Molles 2007 population …………...……..........29
3.1 Example of the Allee effect, demonstrating a reduced population growth rate at low
density (from Kuussaari et al. 1998) …..………………………………………………………….…………….37
viii
LIST OF TERMS
Association – social network parameter which measures the amount of spatial and temporal overlap
between two individuals, calculated from trapping and telemetry overlap
Coefficient of relatedness (r) – the proportion of genetic material (e.g. alleles) shared between two
individuals
Clustering coefficient – social network parameter which measures the level of association between an
individual’s associates, i.e. how closely one’s associates interact with each other
Direct fitness – an individual’s contribution to the gene pool of a subsequent generation based only on
the reproductive success of the individual itself
Fitness – an individual’s contribution to the gene pool of a subsequent generation
Group-size effects – the costs and benefits experienced as a result of living in close proximity with
conspecifics
Inclusive fitness – the sum of an individual’s direct and indirect fitness
Indirect fitness – benefit to the reproductive success of an individual’s relative due to help provided by
the individual
Kin selection – concept that natural selection may favor sociality when an individual maximizes its
inclusive fitness by improving the reproductive success of its relatives
Natal philopatry – offspring remain at the birthplace beyond sexual maturity
Plural breeding – breeding system in which multiple individuals breed
Singular breeding – breeding system in which one or few individuals dominate breeding
Social network analysis – the mathematical measuring of relationships between individuals within a
given area during a given time
1
2
3
ix
LIST OF TERMS
Sociality – two or more conspecifics living in close proximity
Strength – social network parameter which measures the sum of an individual’s social interactions,
based on the total number and intensity of interactions
x
CHAPTER I
INTRODUCTION
Sociality models
Animal sociality involves the cooperative interactions between conspecifics that result in social
relationships of varying degree and duration within populations of social species. Animal social systems
derive from three main components: (i) composition of social groups, (ii) social structure (interactions
between individuals), and (iii) mating system (Schradin, 2013). A major aim of behavioral ecology is to
determine the factors that cause inter- and intraspecific variation in social systems and the reproductive
consequences of this variation.
Each of these components, and in turn the social system as a whole, may be influenced by many
factors both intrinsic and extrinsic to individuals within the population. Ecological conditions, life history
traits, and the evolutionary history of the species may all influence individuals’ behavior and the type
and extent of sociality seen in a population (Figure 1.1). For example, whether offspring decide to remain
in their natal territory or disperse to a new territory may be dependent on the quality of resources (e.g.
food, mates) available in each area (Emlen, 1982). Certain life history traits, such as brain size, may
influence the capacity of individuals to maintain complex social relationships, dictating the extent to
which sociality is possible (Dunbar and Shultz, 2007). Moreover, other life history traits, such as
longevity, contribute to aspects of sociality (Arnold and Owens, 1998). For example, in long-lived species,
opportunities to establish a breeding territory may be reduced by low adult turnover rates, which may
influence rates of philopatry and dispersal. Finally, a species propensity for sociality may also be driven
1
by the characteristics of its evolutionary ancestors (Shultz et al., 2011). A comprehensive model to
explain how these factors may converge and result in animal sociality is Emlen’s (1995) ‘integrated
theory of family social dynamics.’ A major goal of my study is to test components of Emlen’s model,
specifically the ecological and social factors underlying kin structure. Herein, I describe the major
framework underlying the model and how each component contributes to the evolution of sociality.
Ecological conditions
Life history
Evolutionary history
Behavioral
mechanisms of group
formation
Kin structure
Individual differences
in inclusive fitness
Figure 1.1 Conceptual framework for the drivers and outcomes of kin structure
Emlen’s theory of the family
For decades, the study of animal sociality has largely worked under the paradigm outlined in
Emlen’s (1995) model, which posits that social groups form when offspring remain philopatric to the
2
natal nest (natal philopatry), resulting in the formation of extended family groups. Emlen’s model
integrates three major theories: ecological constraints, inclusive fitness theory, and reproductive skew.
Ecological constraints
The first component of Emlen’s model is ‘ecological constraints theory,’ which suggests that
natal philopatry originally arises in response to ecological conditions that constrain offspring dispersal
from the natal nest (Emlen, 1982). Under these conditions, it is more beneficial for offspring to remain
in the natal nest rather than disperse and attempt to breed in a new territory. For example, if an
individual is born into a natal group that maintains a high quality territory with good access to food
resources and protection from predation, and the surrounding territories are of lesser quality, the
individual stands to gain the greatest fitness by remaining philopatric, leading to the formation of social
groups composed of close relatives (Emlen, 1982). Thus, when conditions do not favor offspring
dispersal and independent breeding, we expect more and larger social groups to form. Evidence
supporting ecological constraints theory comes from research on a variety of taxa, including insects
(Field et al., 1998), fish (Bergmuller et al., 2005), birds (Komdeur et al., 1995; Arnold and Owens, 1999),
and mammals (Chapman et al., 1995; Faulkes et al., 1997; Randall et al., 2005; Lucia et al., 2008).
Kin selection and inclusive fitness
Emlen’s model also incorporates kin selection theory to explain that philopatric individuals may
further increase fitness indirectly by providing care to offspring produced by closely-related kin
(Hamilton, 1964; Maynard-Smith, 1964). Hamilton’s rule states that care should be provided by an
individual when the benefit of the care to the recipient (B), multiplied by the relatedness of the
individual to the recipient (coefficient of relatedness, r), is greater than the cost (C) to the individual for
providing the care (rB-C>0) (Hamilton, 1964). In other words, an individual may gain indirect fitness by
3
enhancing a close relative’s direct fitness through providing care to the relative’s offspring, who share
some of the same genetic material as the individual providing care. Thus, selection is predicted to favor
strategies to maximize inclusive fitness, the sum of indirect and direct fitness (Hamilton, 1964).
Although controversial (West et al., 2002; Wilson, 2005; Nowak et al., 2010), kin selection and inclusive
fitness are foundations of animal social theory (Emlen, 1995; Abbot et al., 2011). Decades of research
have supported the role of kin selection as a driver of sociality in numerous species, including
invertebrates (Trivers and Hare, 1976), birds (Brown, 1987), and mammals (Solomon, 2003; Kappeler,
2008). Based on this research, natal philopatry and kin structure are generally accepted as defining
characteristics of social groups in most social species (Lacey and Sherman, 2007).
Reproductive skew
The final component of Emlen’s model, reproductive skew theory, builds upon aspects of both
ecological constraints and kin selection. Reproductive skew describes how much direct reproduction is
shared among individuals in group. High skew is when one or a few individuals breed, whereas low
skew is when reproduction is shared more evenly among several group members. The central idea
behind reproductive skew theory is that, following the formation of social groups due to ecological
constraints and natal philopatry, dominant breeders in the group may share reproduction with
subordinates under certain conditions. In particular, dominant individuals may allow subordinates to
breed if it induces subordinates to remain in the group, and if the presence of subordinates increases
the dominant’s inclusive fitness above what it would be if they were to disperse (Emlen, 1995). More
specific aspects of the theory, such as decreasing ecological constraints leading to increasing shared
reproduction, have been supported in multiple species (Curry, 1988; Emlen and Wrege, 1991).
Regardless, group stability may vary considerably in relation to competition for resources or mates.
4
Such competition will influence the amount of reproductive skew and the potential costs and benefits
to group members, which in turn influences group stability and whether groups disband or remain
together (Emlen, 1995).
Conflicting evidence
Emlen’s model is not universally accepted, and a growing body of evidence suggests that natal
philopatry is not the only mechanism underlying the formation of social groups. Some invertebrates
(Queller et al., 2000; Seppa et al., 2008), fish (Avise and Shapiro, 1986), birds (Griesser et al., 2008), and
mammals (Faulkes and Bennett, 2001; Kappeler, 2008; Ebensperger et al., 2009) form groups when
adults move into existing social groups or establish new groups with unrelated conspecifics. For
example, adult nutria (Myocastor coypus) migrate between social groups (Guichon et al., 2003), with
subsequent research showing groups that lack kin structure (Tunez et al., 2009). In such cases,
individuals cannot benefit from indirect fitness gains and thus, their inclusive fitness derives solely
from direct fitness.
Further, recent research has also called into question the extent to which inclusive fitness
adequately explains group-living even amongst related individuals (Nowak et al., 2010). Although the
benefits of helping kin are intrinsically obvious, empirically determining the extent and result of such help
is difficult (West and Griffin, 2002). For example, helping affects fitness in numerous ways, including
indirectly (Hamilton, 1964) or from group augmentation (see group size effects below), which in turn may
influence survival and reproductive success. Helping may also influence the future direct fitness of group
members if it contributes to improved breeding position within the group. Further, it is likely that indirect
benefits of helping kin have largely been overestimated. For example, estimates of inclusive fitness have
often included the effects of helping on direct descendants as well as those of kin, resulting in artificially
5
high estimates of the indirect benefits (Clutton-Brock, 2002). Conversely, the direct fitness benefits of
group living have likely been underestimated, with some research demonstrating that helpers may
increase their direct fitness overall by improving their likelihood for survival and successful reproduction
or increasing their likelihood of successful dispersal in the future (Grinnell et al., 1995; Komdeur, 1996;
Clutton-Brock, 2002). The inherent trade-offs that arise from group living have rarely been compared to
the indirect benefits gained by group members, a comparison that is necessary to fully understand the
costs and benefits of such a system and ultimately explain its evolutionary significance (West et al., 2001).
Trade-offs of sociality
Regardless of kinship, sociality may result in increased direct fitness of individuals through a
number of mechanisms. These fitness benefits are largely tied to group-size effects i.e., costs and
benefits derived simply from living in close proximity to other individuals (Krause and Ruxton, 2002). For
example, individuals may benefit from a reduced need for vigilance and reduced predation risk simply
due to the collective vigilance of the group as a whole (‘many eyes hypothesis’). This improved vigilance
as a product of group living has been demonstrated in both vertebrate and invertebrate species
(Kenward, 1978; Treherne and Foster, 1980). Further, reduced predation risk may stem simply from
mathematically reducing an individual’s likelihood of being depredated relative to other individuals
around it, known as the dilution effect (Williams, 1966). For example, a solitary individual that
encounters a predator has a certain likelihood of being depredated (N), whereas in a group of several
others that individual’s likelihood of being depredated is reduced to 1/N simply due to probability. For
example, Calvert et al. (1979) demonstrated that, in winter aggregations of monarch butterflies,
predation rates per individual decreased with increasing colony size, even though larger colonies
attracted more predators.
6
Sociality may also benefit individuals by providing better access to resources through several
mechanisms. If individuals in a group share information amongst themselves, the group as a whole may
benefit from an increased ability to find high quality resources, such as food. In such cases, groups serve
as a sort of information center and group members benefit from the information gathered by other
group members (Ward and Zahavi, 1973). Further, in predatory species, sociality may result in improved
rates of prey capture. For example, in groups of black-headed gulls (Chroicocephalus ridibundus), an
individual’s hunting success on prey fish increases with the number of predators in the group (Gotmark
et al., 1986). Similar effects can be seen in pack-hunting animals, such as spotted hyena (Crocuta
crocuta) groups assembled to hunt larger prey (Kruuk, 1972).
At the same time, sociality may be costly to individuals under some environmental conditions.
Most obviously, individuals surrounded by other conspecifics are still in competition with each other for
resources, including food and mates. However, there are also less obvious potential detriments to living
socially. Being in close proximity to several individuals may increase the likelihood of disease
transmission and parasitism within a group (Côté and Poulin, 1995). For example, Brown and Brown
(1986) showed that the number of parasites on nestling cliff swallows (Hirundo pyrrhonota) increased
with increased colony size and that subsequently nestling body mass decreased as a result of increased
parasitism. Further, the behavioral traits necessary to maintain sociality may in fact result in a trade-off
to an individual’s success in other aspects of life. For example, Magurran and Seghers (1991)
demonstrated that guppies (Poecilia reticulata) prone to shoaling were also less aggressive when in
competition for food. Additionally, group living may result in an increased risk of cannibalism (Thiel
2011) and infanticide (Van Schaik and Kappeler, 1997). Lastly, living with close relatives incurs the
inherent risk of inbreeding, though this may be reduced or avoided through sex-biased dispersal and/or
reproductive suppression (Wolff, 1992).
7
While many of these potential trade-offs to sociality may seem logical, from an evolutionary
perspective they become particularly interesting in cases in which sociality results in reduced direct
fitness of group members and indirect fitness is unlikely. In cases where social groups are not composed
of close relatives (i.e. non-kin groups), individuals will not benefit from indirect fitness gains, and thus
may actually experience net fitness costs under conditions in which sociality reduces their direct fitness.
While not common, the presence of non-kin groups has been demonstrated in several taxa (Davies and
Lundberg, 1984; Túnez et al., 2009). In particular, analysis across bird taxa has demonstrated that
approximately 15% of cooperatively-breeding bird species primarily nest with non-related individuals
(Riehl, 2013). This seems to present an evolutionary paradox, as an individual’s inclusive fitness may be
reduced by group-living and selection should not favor the trait. Thus, it is crucial to fully investigate
such situations in order to understand the interactions between a given social system and the costs and
benefits that arise from it in order to ultimately understand how they may evolve.
Thesis objectives
The goal of this research is to describe the social structure in two populations of degus
(Octodon degus), a social rodent endemic to central Chile. Degus are plural breeders that live socially
in underground burrows, with female group members providing care to both related and unrelated
offspring. Previous research on one degu population indicates that social groups consist of unrelated
individuals (Quirici et al., 2011) and that some individuals in large groups experience direct fitness
costs (Hayes et al., 2009). This is particularly interesting given that degus are short-lived (lifespan
generally less than 1 year in the wild) and typically only reproduce once during a lifetime; individuals
who lose opportunities to maximize fitness are unlikely to benefit during a future reproductive cycle.
However, degus live in ecologically distinct populations throughout their geographic range, with little
8
known about the genetic structure of social associations in other populations experiencing different
ecological conditions. Further, no research has been conducted investigating the relationship between
within-site and between-site ecological conditions and sociality at the scale of individual associations
(as opposed to social groups as a whole). Thus, the specific aims of this study were two-fold: 1)
determine the genetic composition of social groups in a second population in a different habitat, and 2)
use social network analysis (Sih et al., 2009) to investigate individual and group-level relationships
between genetic relatedness, ecological conditions, and social structure in two populations. In doing so, I
can test the working hypotheses that kin structure and local habitat conditions influence social
interactions among individuals, and therefore the social network structure of the population.
Study sites
This study was conducted on two degu populations in geographically distinct sites (400 km
apart) in north-central Chile, Rinconada de Maipú (33°23′S, 70°31′W, altitude 495 m) and Bocatoma
Los Molles (30°45′S, 70°15′W, altitude 2,600 m; see Figure 1.2). The sites are characterized by
differences in several ecological traits, with Rinconada having harder soil, greater food abundance,
greater distance from burrows to overhead cover, and lower density of burrow openings than Los
Molles. Previous research has also demonstrated differences in sociality between the two populations,
with slightly larger social groups at Rinconada than Los Molles (Ebensperger et al., 2012a). Prior to this
study, there had been no genetic analyses to determine kin structure within the Los Molles population.
Further, there had been no examination of the relationships between local ecological conditions, kin
structure, and social structure at the individual level for either population.
9
Figure 1.2 Geographic location of Los Molles (top) and Rinconada (bottom) in central Chile
Significance
Although rare, the existence
ce of non-kin groups challenges the established paraadigm of kin
selection as the driving force of socciality in these species. While there are potential dire
ect fitness
benefits to sociality regardless of kiinship (e.g., decreased predation risk, increased foraaging efficiency),
indirect benefits are minimal or non
n-existent in groups of distantly related or unrelated
ated individuals.
Moreover, the relationship between sociality and fitness varies considerably across species,
species and group
10
size indices are often unlikely to capture all variation within a social system (Krause and Ruxton, 2002;
Ebensperger et al., 2012b). For example, there may be direct fitness costs to individuals living in large
groups (Rasa, 1989; Lacey, 2004). By investigating social relationships on the scale of individuals, a
major goal of this study, it may be possible to deduce links between local ecological conditions, genetic
relatedness, and social structure that were not apparent when using other indices of sociality (e.g.
group size). This may reveal relationships that were previously unseen, and thus lead to a more
comprehensive understanding of the evolution of sociality.
11
CHAPTER II
KINSHIP AND SOCIAL NETWORK STRUCTURE IN TWO POPULATIONS OF OCTODON DEGUS
Introduction
Social structure summarizes the nature and extent to which animals interact with others within
a population (Whitehead, 2008; Schradin, 2013). In social species, the cooperative interactions among
individuals in a population are the background upon which foraging, mating, and reproductive
interactions take place (Wolf et al., 2007). Thus, determining the factors that influence these
interactions is crucial to developing a comprehensive understanding of the evolution of sociality. One
well-established model explaining potential conditions leading to and favoring sociality is Emlen’s (1995)
“integrated theory of family social dynamics”. This model incorporates two important theories ecological constraints (Emlen, 1982) and kin selection (Hamilton, 1964) - to explain the evolution of
animal sociality. The model posits that extended family groups (kin groups) form when juveniles remain
philopatric to the natal group under conditions that limit direct reproduction (ecological constraints;
Emlen, 1982). Under these conditions, kin selection theory predicts that breeders benefit when
philopatric individuals assist with offspring care (alloparental care) and philopatric individuals benefit
indirectly by providing care to non-descendent offspring produced by closely related kin. Thus, parental
care directed towards closely related kin is predicted to increase an individual’s inclusive fitness
(Hamilton, 1964; Maynard-Smith, 1964).
The two main thrusts of Emlen’s model – ecological constraints and kin selection – have been
the subject of considerable theoretical and empirical work for decades. Thus, the impacts of ecological
12
constraints on animal sociality have been demonstrated in invertebrates (Emlen, 1982; Rehan and
Schwarz, 2011) and vertebrates (Komdeur, 1992; Travis et al., 1995; Lucia et al., 2008; Schoepf and
Schradin, 2012). Consequently, ecological constraints are often viewed as a primary driver for social
group formation. Regarding the influence of kin selection, decades of research have validated that
groups typically consist of extended families as has been observed in invertebrates (Trivers and Hare,
1976), birds (Brown, 1987; Stacey and Ligon, 1991), and mammals (Solomon, 2003; Kappeler, 2008).
Taken together, these observations suggest that natal philopatry and inclusive fitness benefits are the
defining characteristics of groups in most social species (Emlen, 1995; Lacey and Sherman, 2007; but
see: Griffin and West, 2002; Nowak et al., 2010).
However, a growing body of evidence suggests that Emlen’s model is not universal for social
animals and that natal philopatry is not the only mechanism underlying the formation of social groups.
In some invertebrates (Queller et al., 2000; Seppa et al., 2008), fish (Avise and Shapiro, 1986), birds
(Griesser et al., 2008), and mammals (Faulkes et al., 2001; Guichon et al., 2003; Ebensperger et al.,
2009), adults move into existing social groups or establish new groups with unrelated conspecifics
(Ebensperger and Hayes, 2008). Consequently, some of these species may live in non-kin groups.
Mammals with low mean levels of relatedness within social groups include rodents (Tunez et al., 2009;
Quirici et al., 2011), white-tailed deer (Odocoileus virginianus), and river otters (Lontra Canadensis)
(Smith, 2014).
In non-kin groups, the potential costs of group living (e.g. competition for resources) and
cooperation (e.g. cheating) are not outweighed by indirect fitness benefits of cooperating with kin.
However, cooperation may exist irrespective of the genetic relatedness of individuals (West and Griffin,
2002). For example, it has been shown that the amount of parental care provided by subordinates in
meerkat (Suricata suricatta) social groups is not driven by the relatedness of the subordinate to the
13
offspring that is receiving care (Clutton-Brock et al., 2001). Thus, it may be necessary to investigate
other avenues of fitness gains outside of kinship in cases where indirect fitness is unlikely to explain the
evolution of sociality.
Intraspecific variation and sociality
Historically, research on animal social systems has focused on single populations, assuming that
the social system is fixed by natural selection and consistent across species. However, intraspecific
variation in social systems has been observed in reptiles (Shine and Fitzgerald, 1995), birds (Komdeur,
1992) and mammals (Travis and Slobodchikoff, 1993; Brashares and Arcese, 2002; Ophir et al., 2007).
Such variation is expected if the associated costs and benefits of sociality depend on local ecological
conditions and result in differential selection on the behaviors influencing group formation (Emlen and
Oring, 1977; Lott, 1991). At the proximate level, intraspecific variation in social structure may arise due
to genetic variation and/or varying levels of phenotypic plasticity between populations (Schradin, 2013).
Since non-kin groups provide a challenge to the fundamental model of group formation, it is critical to
investigate the site-specific social systems and environmental conditions characteristics of multiple
populations in species that exhibit non-kin social structure. This approach will allow researchers to
determine if non-kin groups are common or the rare product of local conditions acting upon individual
populations.
Social networks
The most common metric of sociality – group size – provides only one dimension of an animal’s
social system. A challenge to developing a comprehensive model for sociality is to determine social
relationships at varying scales and dimensions that collectively make up the social structure of a
14
population. A quantifiable method of analysis for such questions is social network analysis (Whitehead,
2008; Wey et al., 2008; Sih et al., 2009). Social networks model the ways in which individuals interact
with other individuals in the population, allowing researchers to quantify the strength and extent of
relationships in ways that traditional methods of social determination cannot (Sih et al., 2009). By
analyzing the network dynamics at both the individual and social group level (calculated as means from
the values of each group member), it is possible to answer questions about an individual’s social
connections as well as questions about how the individual associations interact at varying levels to form
the social structure of the population as a whole (Wey et al., 2008). For example, social network analysis
has provided insights into complex patterns of sociality, including quantifying distinct structural layers
within a population’s social system (Wolf & Trillmich, 2008) and determining how social associations
predict patterns of cooperation (Croft et al., 2006). Based on theory, we expect stronger social
interactions among kin than non-kin, a prediction that can be tested by comparing within-group
relatedness (calculated as a mean from the pairwise relatedness values of group members) with grouplevel social network parameters (association, strength) which quantify the proportion of time individuals
spend in close proximity with each other. Further, ecological characteristics may influence social
network parameters by dictating how individuals move through their environment, both spatially and
temporally, ultimately affecting the extent to which they interact with other individuals in a given area.
Thus, habitat use may influence a given social network parameter (Croft et al., 2004).
Objectives and study organism
The degu (Octodon degus) is a group-living caviomorph rodent endemic to central Chile (Hayes
et al., 2011). Degus are widespread, occurring in ecologically distinct habitats throughout their
geographic range (Meserve et al., 1984; Ebensperger et al., 2012a), making them a good model
15
organism for examining how local conditions influence the formation and composition of groups and
social associations at the population level. In one population (Rinconada de Maipú, Chile; 33°23′S,
70°31′W), the immigration and emigration of adults into and out of groups is a more important driver of
group formation than natal philopatry by offspring (Ebensperger et al., 2009). Consequently, genetic
relatedness (R) of individuals within groups is similar to that of the background population, indicating an
absence of kin structure (Quirici et al., 2011). A recent study comparing social groups in Rinconada and a
second population (Bocatoma Los Molles) revealed that groups differ in size between these populations
(Ebensperger et al., 2012a). This observation suggests some degree of intraspecific variation in degu
social organization. To date, no one has investigated if these differences in social organization and local
ecological conditions are linked to differences in kin and social network structure. The objectives of this
study were to determine if kin structure differed between two degu populations and to use social
network analysis to investigate possible links between ecological conditions, kinship, and social
associations at both individual and group-level scales within each population.
While non-kin groups are prevalent at Rinconada, the kin structure of social groups at Los Molles
was previously unknown. Ebensperger et al. (2012a) also showed that the sites differ in predation risk
and distribution of food resources. Based on Emlen’s model, we then expect differences in sociality
between the two populations. In particular, since resources are more patchily distributed at Los Molles
than Rinconada, I expect greater natal philopatry at Los Molles and predict that a greater percentage of
groups are kin-based in this population. Additionally, if kinship is driving social interactions among group
members, I predict a positive relationship between within-group relatedness and group-level strength
and association in both populations. Lastly, individuals inhabiting high quality habitats may experience
stronger and/or more social interactions. If social network structure at the individual level is driven by
within-site ecological conditions (e.g. food availability near burrows), I predict a positive relationship
16
between the habitat quality of burrows used by an individual and that individual’s social network
parameters (strength and clustering coefficient).
Methods
Study populations
I determined the social network and kin structure of social groups from two degu populations in
central-north Chile, Estación Experimental Rinconada de Maipú (33˚23′S, 70˚31′W, altitude 495 m)
(hereafter Rinconada) and Bocatoma Los Molles (30˚45′S, 70˚15′W, altitude 2,600 m) (hereafter Los
Molles). The environmental conditions at these sites differ in several ways, with Rinconada having
harder soil, greater food abundance, greater distance from burrows to overhead cover, and lower
density of burrow openings than Los Molles (Ebensperger et al., 2012a). Predator sightings are more
frequent at Rinconada than Los Molles (Ebensperger et al., 2012a). The fieldwork was conducted in 2007
and 2008, during the time when females were in late pregnancy or lactating (i.e. September–October at
Rinconada and November–December at Los Molles).
Ecological sampling
I used data on ecological conditions at burrow systems (food availability, burrow density,
and soil hardness; Ebensperger et al., 2012a) to determine if local ecological variation predicted
social structure within each site. To quantify food availability, a 250 x 250-mm quadrat was placed
at 3 meters and 9 meters from the center of each burrow system in one of the cardinal directions
(randomly selected for each distance at each burrow system), and all above-ground green herbs
were removed, dried, and weighed for biomass. Burrow density (openings per square meter) was
quantified by counting the number of burrow openings within a 9-m radius from the center of each
17
burrow system. Soil hardness was sampled similarly to food availability, with a soil penetrability
measurement taken at 3 and 9-m from the center of each burrow system in one of the cardinal
directions. Distance to overhead cover was measured from the center of each burrow system using
a 100-m measuring tape (Ebensperger et al., 2012a).
Social group determination
Degus are diurnal and remain in underground burrows with conspecifics overnight (Ebensperger
et al., 2004). Thus, the main criterion used to assign degus to social groups was the sharing of burrow
systems during the nighttime (Ebensperger et al., 2004). To determine social group membership, I used
a combination of night-time telemetry and early morning burrow trapping. During burrow trapping, a
burrow system was defined as a group of burrows surrounding a central location where individuals were
repeatedly found during telemetry (Hayes et al., 2007). Burrow systems were trapped an average of
31.4′±′1.2 (mean ± standard error) days in 2007 and 45.3′±′1.6 days in 2008 at Rinconada, and for 30
days in 2007 and 21 days in 2008 at Los Molles. Tomahawk live-traps (model 201, Tomahawk Live Trap
Company, Tomahawk, Wisconsin, USA) were set prior to the emergence of adults during the early
morning hours (0700-0730) and were checked and closed after 1.5 hours. The identity, sex, body mass,
and reproductive condition of all individuals were determined at first capture. Additionally, a small
tissue sample was taken from each individual’s ear the first time it was captured and stored in 99%
ethanol at 0 °C. Adults weighing more than 170 g were fitted with 8 g (BR radio-collars, AVM Instrument
Co., Colfax, California, USA) or 7–9 g radio-transmitters (RI-2D, Holohil Systems Limited, Carp, Ontario,
Canada) with unique frequencies.
During night-time telemetry, females were radio-tracked to burrow systems. Previous studies at
Rinconada have demonstrated that telemetry locations represent sites where degus remain
18
underground throughout the night (Ebensperger et al., 2004). Locations were determined once per
night approximately 1 hour after sunset using an LA 12-Q receiver (for radio collars tuned to 150.000–
151.999 MHz frequency; AVM Instrument Co., USA) and a hand held, 3-element Yagi antenna (AVM
instrument Co.).
To determine social group membership, I created a similarity matrix of pairwise associations of
the burrow locations of all adult degus during trapping and telemetry (Whitehead, 2009). Associations
were determined using the “simple ratio” association index (Ginsberg and Young, 1992), i.e. the number
of times that two individuals are captured or tracked via telemetry at the same burrow system on the
same day divided by the total number of times each is captured/tracked on the same day regardless of
burrow system. Only associations with a value greater than 0.1 (i.e. 10% overlap of trapping/telemetry
locations) were included. Since social network parameters are calculated based on trapping data, only
degus that were trapped on at least five days were included in analyses to exclude poorly sampled
individuals (Wey et al., 2013). Thus, some individuals and social groups previously used in Ebensperger
et al. 2012a were excluded from our study to avoid biasing the network data. For example, Ebensperger
et al. 2012a reported four social groups at Los Molles in 2007; however, one of these groups consisted
of a solitary individual, and several individuals within two other groups had poor trapping data. Thus, I
use only one social group from this year in my study. Social groups were determined using a hierarchical
cluster analysis in SOCPROG 2.0 software (Whitehead, 2009).
Social network analysis
Social network analysis was used to look for patterns of sociality at the individual level, including
pairwise relationships between group members and non-group members. For each individual, I
calculated the strength – the sum of associations (Whitehead, 2008) - calculated from the pairwise
19
association networks. High strength indicates a high total amount of spatial and temporal overlap with
other individuals, resulting from strong associations, many associations, or a combination of both. For
each individual, I also calculated the clustering coefficient, a measure of how connected an individual’s
associates are to each other (e.g. an individual with a high clustering coefficient has close associations
with individuals who also associate closely with each other, forming a “cluster”). For each social group, I
calculated the mean association (from each pair of group members, based on the “simple ratio”
explained above) and the mean strength, based on the individual values for each group member.
Network parameters were calculated from pairwise similarity matrices in SOCPROG 2.0.
Genetic analysis
Genetic analyses to determine relatedness (R) were conducted in the Molecular Ecology lab at
the Universidad de Chile in Santiago, Chile. Analyses were conducted on tissue samples collected from
n=14 and n=26 individuals at Los Molles and n=21 and n=29 individuals at Rinconada in 2007 and 2008,
respectively. Genomic DNA was extracted from tissue using a standard salt extraction protocol.
Amplification of DNA was achieved using polymerase chain reaction of 100 ng of DNA from each
individual using the conditions recommended by Quan et al (2009). Amplification was confirmed with
agarose gel electrophoresis. Individuals were genotyped using 5 degu microsatellite loci (OCDE3, OCDE6,
OCDE11, OCDE12, OCDE13; Quan et al. 2009). These loci were used because they were polymorphic and
showed no linkage disequilibrium during previous studies (Quan et al., 2009; Quirici et al., 2011). Allele
quantification and testing for linkage disequilibrium were performed in GENEPOP 4.2 (Raymond and
Russet, 1995). Microsatellite sequencing was performed by Macrogen, Inc. (Seoul, South Korea). Allele
sizes were determined and genotypes assigned using PeakScanner 2.4 software. Deviations from Hardy-
20
Weinberg equilibrium and the pairwise coefficient of relatedness (R) among individuals were calculated
using the ML-Relate software (Kalinowski et al., 2006).
Statistical analysis
To determine if social groups consisted of closely-related kin, mean pairwise relatedness across
group members was compared to the relatedness of the background population consisting of all
individuals for which there was genetic data. To determine the relatedness of the background
population, bootstrapping analysis (n=1000 permutations, with replacement) was performed on
randomly selected pairs of individuals irrespective of social group, with sample sizes dependent on the
number of individuals in each social group (e.g. 3 randomly selected pairs for group size = 3) using R
version 3.1.1 statistical software. Groups with mean pairwise relatedness that fell outside of the 95%
confidence interval for the randomly-selected background population were considered statistically
different from the background population.
To examine how population, group size, and genetic relatedness influence group-level social
network parameters (mean association and mean strength), I first used Akaike Information Criterion
(Akaike, 1974) to determine the best fit model for both mean association and mean strength. Each
possible combination of factors and interactions was tested and the model with the lowest AIC value
was selected. The best fit model for each network parameter was then used to test for significant main
effects and interactions. For mean strength, the best fit model included population, group size, and
relatedness as factors (R2 = 0.90, AIC = 2.59). Thus, an ANCOVA was performed with each of those three
fixed factors as well as the population*relatedness interaction. For mean association, the best fit model
included only population as a fixed factor (R2 = 0.85, AIC = -61.60), and thus group size and relatedness
were removed as factors influencing mean association. With only one fixed factor, an independent
21
samples t-test was run to test for differences in mean association between the two populations.
Additionally, simple linear regressions were run on each population to test for relationships between
group size and relatedness. All group-level analyses were conducted using SAS 9.3 (SAS Institute, Inc.,
Cary, NC).
To evaluate the relationship between an individual’s social network parameters (strength and
clustering coefficient) and the ecology (food biomass, burrow density, soil hardness) of its burrow
systems within both populations, I conducted multiple regressions with all weighted (based on
proportion of captures at burrow systems) ecological characteristics as independent variables and each
network parameter as the dependent variable. To test the assumption that the model was linear, I
visually inspected a plot of standardized residuals. To test for autocorrelation, the Durbin–Watson
statistic (d) had to range between 1.5 and 2.5. To test for homoscedasticity, I visually inspected the data
point spread showing the regression standardized residual vs. the regression standardized predicted
value. Variables were considered collinear if the variability inflation factor (VIF) was greater than 4.0.
Multiple regressions were conducted with SPSS Statistics 22.2 (IBM, Inc., Chicago, IL). For all analyses, I
set the alpha level to P=0.05. Throughout, I report means with standard errors (SE).
Results
Microsatellite variation
There was no evidence of linkage disequilibrium across all 5 loci screened (P > 0.05 for each
loci). The number of alleles per locus ranged from 5-12. The observed heterozygosity of loci ranged from
0.36-0.79 for Rinconada and 0.48-0.91 for Los Molles. In both populations, deviations from HardyWeinberg equilibrium were detected for two loci (OCDE6 and OCDE12, P < 0.01). Therefore, estimations
of pairwise relatedness were adjusted to account for potential null alleles using the ML-Relate software.
22
Descriptive data
The pairwise relatedness between individuals ranged from 0.00-0.52 and 0.00-0.58 for
Rinconada and Los Molles, respectively. The mean group-level relatedness ranged from 0.07 to 0.21 at
Rinconada and from 0.09 to 0.25 at Los Molles.
Individual network strength ranged from 0.07-4.13 (mean±SE=1.92±0.81) and 1.01-8.00
(4.55±1.44) for Rinconada and Los Molles, respectively. The clustering coefficient for individuals ranged
from 0.01-0.94 (0.34±0.17) for Rinconanda and 0.41-1.00 (0.81±0.37) for Los Molles. At the group level,
mean strength ranged from 0.09-3.48 (2.23±0.67) at Rinconada and 1.00-6.79 (3.36±1.40) at Los Molles.
Mean association ranged from 0.17-0.96 (0.44±0.51) at Rinconada and 0.45-1.00 (0.88±0.46) at Los
Molles.
Relatedness and social structure
Mean group-level relatedness (R) ranged from 0.07-0.25 across all social groups examined.
Bootstrapping analysis indicated that social group members were not significantly more related to each
other compared to randomly selected individuals from the background population, with mean pairwise
relatedness of all groups falling within the 95% confidence intervals of the background population (Table
2.1). Additionally, there was not a statistically significant relationship between group size and
relatedness at either Los Molles (β = -0.79, R2 = 0.63, P = 0.06) or Rinconada (β = -0.27, R2 = 0.07, P =
0.30).
23
Table 2.1 Group size and within-group relatedness of social groups at Rinconada and Los Molles 20072008
(i) Rinconada
Year
Social Group
2007
1
2007
2
2007
3
2007
4
2007
5
2007
6
2007
7
2008
8
2008
9
2008
10
2008
11
2008
12
2008
13
2008
14
2008
15
2008
16
2008
17
(ii) Los Molles
Year
Social Group
1
2007
2
2008
3
2008
4
2008
5
2008
6
2008
Group Size
3
7
6
4
3
3
6
5
2
2
3
3
3
2
2
3
4
Genetic Relatedness
0.19
0.08
0.15
0.17
0.14
0.11
0.09
0.11
0.07
0.21
0.16
0.09
0.11
0.12
0.14
0.09
0.07
95% CI
0-0.31
0-0.21
0-0.22
0-0.25
0-0.31
0-0.31
0-0.22
0-0.24
0-0.41
0-0.41
0-0.31
0-0.31
0-0.31
0-0.41
0-0.41
0-0.31
0-0.25
Group Size
4
6
2
3
6
7
Genetic Relatedness
0.09
0.09
0.21
0.25
0.09
0.10
95% CI
0-0.23
0-0.23
0-0.46
0-0.33
0-0.23
0-0.23
The ANCOVA revealed that group size (F1, 18 = 4.65, P = 0.05) and the population*relatedness
interaction (F1, 18 = 7.44, P = 0.01) were statistically significant predictors of mean strength. Post-hoc
simple linear regressions of relatedness and group size on mean strength for each population separately
showed a statistically significant positive relationship between group size and strength at Rinconada (β =
24
0.56, R2 = 0.31, P = 0.02) but not at Los Molles (β = 0.74, R2 = 0.55, P = 0.09). Relatedness alone was not a
statistically significant predictor of mean strength at Rinconada (β = -0.08, R2 = 0.01, P = 0.76) but
showed a statistically significant negative relationship at Los Molles (β = -0.85, R2 = 0.72, P = 0.03). The
independent samples t-test for mean association revealed a statistically significant difference (P < 0.01)
between the two populations, with Los Molles (mean = 0.88±0.09) social groups having greater mean
association than groups at Rinconada (mean = 0.44±0.07).
Ecology and network structure
Multiple regression analyses reasonably met the regression model assumptions. For the
strength analysis, model-level significance was detected at both Los Molles (F3,29 = 47.47, R2 = 0.85, P <
0.01) and Rinconada (F3,79 = 6.45, R2 = 0.20, P < 0.01). Similarly, the model for clustering coefficient was
significant at both Los Molles (F3,27 = 11.64, R2 = 0.59, P <0.01) and Rinconanda (F3,75 = 15.10, R2 = 0.39, P
< 0.01). At both sites, analyses revealed statistically significant relationships between the ecological
characteristics of burrow systems used by individuals and the individuals’ network parameters. At
Rinconada, there was a statistically significant negative relationship between strength and soil hardness
and a statistically significant positive relationship between clustering coefficient and food biomass
(Table 2.2; Figure 2.1a-b). In other words, at Rinconada, as soil hardness increased, individuals’ network
strength decreased, whereas when food availability increased, individuals’ clustering coefficient also
increased. At Los Molles, there was a statistically significant positive relationship between network
strength and soil hardness, food biomass, and burrow density, and a statistically significant negative
relationship between clustering coefficient and both food biomass and burrow density (Table 2.2; Figure
2.2a-b). In other words, individuals’ network strength increased with increasing food availability,
25
increasing soil hardness, and increasing burrow density. However, individuals’ clustering coefficient
decreased with increasing food availability and increasing burr
burrow density.
(a)
(b)
Figure 2.1 Scatterplot showing the statistically significant relationship between (a) soil hardness and
strength and (b) food biomass and clustering coefficient for indivi
individuals
duals at Rinconada in 20072007
2008
26
(a)
(b)
Figure 2.2 Scatterplot showing the statistically significant relationship between (a) burrow density and
strength and (b) food biomass and clustering coefficient for individ
individuals
uals at Los Molles in 20072007
2008
27
Table 2.2 Multiple regression statistics for individuals’ network parameters vs. weighted habitat
characteristics at Rinconada and Los Molles 2007-2008
Rinconada
Beta
t-value
p-value
-0.41
-0.16
-0.41
-0.13
-3.97
-1.38
< 0.01
0.17
Burrow density
Strength
Clustering coefficient
0.07
-0.19
0.07
-0.16
0.65
-1.6
Food biomass
Strength
Clustering coefficient
0.18
0.54
0.17
0.53
1.6
5.49
Predictor Variable
Soil hardness
Strength
Clustering coefficient
Partial r
Los Molles
Beta
t-value
p-value
0.77
-0.33
0.64
-0.36
6.11
-1.69
< 0.01
0.10
0.52
0.11
0.90
-0.44
1.27
-0.60
10.69
-2.39
< 0.01
0.03
0.11
< 0.01
0.43
-0.74
0.24
-1.04
2.44
-5.40
0.02
< 0.01
Partial r
Discussion
This study confirms previous evidence for non-kin structure in degu social groups, challenging
the importance of kin selection on this species (see Figure 2.3). My study did not reveal the underlying
social network and ecological drivers of variation in kin structure in degus. However, this study yielded
important insights about the social structure of degus and the impact of habitat quality on degu social
networks. Thus, my study does provide insight into the evolution of non-kin sociality.
Social structure and kinship
In my study, mean pairwise relatedness within social groups was not significantly greater than
would be expected from random pairwise comparisons of individuals selected from the background
population in both Rinconada and Los Molles. These observations, and those made previously in
Rinconada (Quirici et al., 2011), suggest that non-kin group structure is typical of degu sociality and not
just a characteristic of one population. Additionally, my observations that group size is not a significant
28
predictor of group relatedness for either population are consistent with previous findings regarding the
mechanisms of group formation in degus at Rinconada. Although natal philopatry plays a role in the
formation of degu groups, non-sex
sex bias
biased
ed dispersal and the movement of adults between groups are
important drivers (Ebensperger
Ebensperger et al.
al., 2009; Quirici et al., 2011).
). Under these conditions, a negative
relationship between group size and relatedness is not expected, as the composition of groups varies
based on the relative influence of each mechanism on group formation. Further analysis is needed
need to
determine the extent to which each mechanism influences group composition, particularly at Los
Molles.
Figure 2.3. Social network structure of (a) a theoretical social group exhibiting kin structure and (b) an
actual social group from the Los Molles 2007 population
While my study revealed significant population
population-level
level differences in the relationship between
various
us social and genetic factors an
and group-level
level network characteristics, these results should be
29
interpreted with some caution due to the influence of inherent population differences, as well as the
characteristics of the network parameters themselves. For example, while mean association was
significantly higher at Los Molles than Rinconada, this may have been influenced by differences in degu
abundance between populations. Abundance was considerably lower at Los Molles than at Rinconada
across both years. It is likely that higher association values at Los Molles were the result of individuals
having fewer potential associates and thus, being captured in the same burrow systems with individuals
proportionally more often than would occur in populations with more individuals. Similarly, while there
was a significant relationship between group size and strength at Rinconada but not Los Molles, the R2
of the regression was considerably higher at Los Molles than Rinconada (0.55 vs. 0.31), suggesting that
the lack of significance may in part be linked to a small sample size (n = 6 groups at Los Molles, n = 17
groups at Rinconada).
Theory predicts that kin groups form as a result of natal philopatry, usually by one sex (Emlen,
1995). However, a growing body of evidence calls into question the validity of kin selection as the
ultimate driver of sociality across taxa (West and Griffin, 2002; Wilson, 2005). Although natal philopatry
(and the resultant kin groups) remains a common mechanism of group formation in many species, other
mechanisms of group formation, including the immigration and emigration of adults, influence group
structure in some invertebrates (Trivers and Hare, 1976), birds (Brown, 1987), and mammals (Solomon,
2003; Kappeler, 2008, Ebensperger and Hayes, 2008). At Rinconada, the dispersal of degu offspring is not
sex-biased, with both sexes dispersing at roughly the same rate (Ebensperger et al., 2009; Quirici et al.,
2011). Further, the primary determinant of group formation and composition in Rinconada is the
disappearance of adults and the movement of adults between social groups. As a result, annual turnover
of adults comprising social groups is typically high (Ebensperger et al., 2009), likely explaining low kin
structure in this population. Although I did not monitor these behaviors at Los Molles, I expect similar
30
mechanisms to have evolved to maintain non-kin structure in this population. A test of this hypothesis
would require a multi-year study to track individuals and their social affiliations between and within
seasons (Ebensperger et al., 2009).
Theory also predicts that relatedness between individuals will facilitate the evolution of
cooperation (Emlen, 1982). Closely related individuals are thought to benefit from direct benefits and
indirect benefits associated with cooperation. However, in species in which groups form social groups in
the absence of kin selection, including degus, unrelated individuals cooperate with little or no chance for
indirect fitness benefits. Several studies on cooperative breeders in birds and mammals have shown that
helpers may care for unrelated young and that unrelated helpers often invest as heavily in offspring care as
close relatives (Dunn et al., 1995; Clutton-Brock, 2000). One possible explanation for this involves potential
future direct fitness gains, where an unrelated helper may forfeit direct fitness during a given breeding
cycle in order to improve potential future direct fitness (Kokko et al., 2002). For example, Reyer (1984)
found that in pied kingfisher (Ceryle rudis) social groups, unrelated ‘secondary’ helpers increased their
future direct fitness when opportunities to mate became available in subsequent breeding seasons.
Life history may explain the evolution of non-kin groups in some cases where kin selection does
not provide an adequate explanation for group-living. For example, kin structure is expected in long-lived
species in which social groups experience low turnover rates. Evidence for this hypothesis comes from
studies showing kin structure in long-lived species such as African elephants (Loxodonta Africana) (Archie
et al., 2006), coypus (Myocastor coypus) (Tunez et al., 2009), and multiple primate species (Silk, 2002). In
contrast, due to high turnover rates, social structure in species with short lifespans often lacks kin
structure, as has been seen in woodrats (Neotoma macrotis) (Matocq and Lacey, 2004) and black grouse
(Tetrao tetrix) (Legibre et al., 2008). Similarly, the breeding strategies of animals can have significant
effects on the genetic structure of social groups. Singular breeders, species with high levels of
31
reproductive skew (Brown, 1987; Hayes, 2000; Silk, 2007) are expected to show high levels of group
relatedness since most or all offspring come from one dominant breeder. In contrast, plural breeders,
species with low skew in groups, are expected to have lower levels of relatedness between group
members (Ross, 2001).
Evidence suggests that a combination of these factors could explain non-kin structure in degus. In
terms of life history, degus have low survival (Ebensperger et al., 2009; Ebensperger et al., 2013) and high
turnover rates from year to year (Ebensperger et al., 2009). In terms of breeding system, degus are plural
breeders with communal care, with most females within social groups showing physical signs of pregnancy
and lactation (Hayes et al., 2009). Interestingly, a recent long-term study suggests that degu sociality may
have evolved as a strategy to deal with harsh mean environmental conditions (Ebensperger et al., 2014),
possibly explaining why group-living persists despite low kin structure.
Social networks and habitat conditions
Contrary to previous work in which local ecological conditions had little predictive power for
group sizes (Hayes et al., 2009; Ebensperger et al., 2012a), I observed that social network structure was
influenced by local ecological conditions in both populations (see Figure 2.1 and 2.2, Table 2.2). At
Rinconada, the negative relationship between strength and soil hardness suggests that individuals
inhabiting burrow systems with softer soil experience stronger and/or more social associations. Previous
studies found that the energetic cost of digging in hard soil is greater than digging in soft soils
(Ebensperger and Bozinovic, 2000a) and that degus digging in groups remove more soil per capita than
solitary individuals (Ebensperger and Bozinovic, 2000b). Thus, softer soil may provide better habitat and
result in a greater degree of sociality. Similarly, the positive relationship between food biomass and
clustering coefficient suggests that individuals may be clustering together around burrows where food
32
resources are abundant. My observation is in agreement with previous studies on invertebrates (Tanner
et al., 2011) and vertebrates (Foster et al., 2012) showing that food availability influences a population’s
social network structure.
At Los Molles, the positive relationships between network strength and both food biomass and
burrow density suggest a similar trend that individuals inhabiting high quality habitats having stronger
and/or more social associations. In contrast to the observed trend at Rinconada, the relationship
between network strength and soil hardness at Los Molles was positive. This difference may be
explained by site-level differences in ecological conditions (Ebensperger et al., 2012a). Overall, the soil
at Los Molles is softer than Rinconada. Since some level of soil hardness is necessary to maintain the
structure of burrows, it is possible that harder soil provides better habitat quality at Los Molles, as softer
soil may not maintain the burrow structures. Other relationships between ecological conditions and
social network structure observed at Los Molles, but not at Rinconada, are more difficult to interpret.
The negative relationships between clustering coefficient and both food biomass and burrow density
suggest that individuals are not clustering more strongly in areas with abundant food and burrows. It is
possible that differences in predation risk (Ebensperger et al., 2012a) influence the distribution of degus
and thus, social network structure at Los Molles. Examinations of the relationship between spatial and
temporal variation in predator abundance and social network structure are needed to test this
hypothesis. Alternatively, it is possible that these unexpected trends were driven by the effect of small
sample sizes (N=29 individuals) on multiple regression analysis.
Regardless, my results suggest that local ecological conditions influence social interactions and
help shape social structure in degu populations. Intraspecific social variation in response to local
ecological conditions has also been demonstrated in numerous taxa (Lott, 1991; Schradin, 2013),
including reptiles (Shine and Fitzgerald, 1995), birds (Davies and Lundberg, 1984) and mammals
33
(MacDonald, 1979; Roberts et al., 1998). Regarding social networks, Henzi et al. (2009) found that
female associations of chacma baboons (Papio hamydra ursinus) varied cyclically in relation to temporal
variation in food abundance. In this sense, degu social structure seems to fit within a common theme,
that local ecological conditions are a significant driver of social variation across species. Future work
should aim to determine if the processes (e.g. phenotypic plasticity) underlying intraspecfic variation in
social structure (Schradin, 2013) differ between sites. Such work could have important implications for
fully explaining the drivers of social variation.
Concluding remarks
The major take-home point of this study is that degu kin structure is consistently non-kin based
and largely insensitive to intraspecific variation in ecological and social network structure. Thus, the results
of this and previous studies on degus (Ebensperger et al., 2009; Quirici et al., 2011) suggest a social system
that does not conform to principles of long-standing paradigms for animal sociality (Hamilton, 1964;
Emlen, 1982). However, my findings also demonstrate that degu social network structure is influenced by
local ecological conditions, and that these influences may result in population-specific social structure at
the individual level. To fully understand these relationships, future work should investigate how degu
social networks vary in relation to temporal changes in ecological conditions across populations. At the
broader scale, researchers need to further examine the complex relationships between life history,
ecological conditions, and social/kin structure. To accomplish this, future research should make use of
large comparative databases (e.g. PanTHERIA, Jones et al., 2009; Lukas and Clutton-Brock, 2012) to
determine if the relationships between these factors are consistent across taxa.
34
CHAPTER III
CONSERVATION IMPLICATIONS
Understanding the variables that shape a population’s social structure has wide-ranging
implications for the conservation and management of social species. Population viability is closely tied
to effective population sizes, which in part may be determined by aspects of sociality, such as
reproductive skew and genetic variation (Anthony and Blumstein, 2000). Thus, social species present
different challenges regarding conservation than solitary species, as in many cases populations must be
considered not solely by the total number of individuals but by the number of social units (e.g. social
groups).
Conservation biologists need to consider the extent to which animals are social, including
whether a species is ‘socially obligate’ or ‘socially flexible.’ In socially obligate species (i.e. species in
which individuals cannot survive under solitary conditions), the effective population size is significantly
reduced, as each individual can only survive within a social group with a minimum number of other
individuals. This presents a significant constraint on populations of such species, as below a certain
population density threshold it may be impossible to maintain sufficient group sizes, and the population
will experience the consequences of an Allee effect (i.e. negative growth rate at low densities, Allee et
al., 1949; Courchamp et al., 1999; Figure 3.1). In response to Allee effects, greater aggregations of
individuals may also change fundamental aspects of the population’s ecology, such as its mating system
and susceptibility to disease outbreaks (Stephens and Sutherland, 1999).
35
Such constraints imposed by sociality have been documented across diverse taxa. White-winged
choughs (Corcorax melanorhamphos) are incapable of successfully breeding in groups of less than four
(Heinsohn, 1992). Cant (1998) found that banded mongoose (Mungos mungo) groups with less than six
individuals failed to successfully raise any pups despite several breeding attempts over multiple years.
Further, it has been suggested that high rates of group extinction within such species are a direct result
of this minimum threshold related to sociality (Clutton-Brock et al., 1999). Thus, adequately
understanding the social dynamics of species may be critical to determining the population densities
necessary to prevent local extinctions.
In addition to determining the conditions under which a species may survive, social structure
can also affect the potential impacts of anthropogenic changes on population viability. Behavioral traits
in social species may be affected by human-induced habitat changes. For example, habitat alteration or
fragmentation may lead to changes in a population’s mating system and rates of migration and
dispersal, which in turn may affect the actual population size, the levels of reproductive skew, or the
population growth rate (Anthony and Blumstein, 2000), all of which may impact the likelihood of the
population to grow or diminish.
Finally, wildlife managers may be able to use the social structure of a species or population to
potentially improve conservation efforts. In particular, since this study and others have demonstrated
how local ecological conditions may affect social structure, managers may use this knowledge to create
conditions in which survival and reproduction are maximized, potentially helping to conserve imperiled
species. For example, if it is known that individuals are attracted to areas with other conspecifics, it may
be possible to induce migrations to higher quality habitat by translocating individuals and/or using
decoys to attract larger sects of the population (Stephens and Sutherland, 1999). Understanding the
underlying causes of reproductive skew and suppression may also help managers create conditions
36
under which reproductive success is maximized. For example, if it is known that a social structure is
typically affected by dominant individuals suppressing the reproduction of subordinates in large groups,
altering resource distributions in a manner that creates several smaller groups rather than a few large
groups may result in fewer individuals being reproductively suppressed (Anthony and Blumstein, 2000).
Thus, this and similar work provides the opportunity to both further investigate the evolutionary
significance of sociality and also gain practical knowledge toward improving conservation efforts in
social species.
Figure 3.1 Example of the Allee effect, demonstrating a reduced population growth rate at low density
(from Kuussaari et al. 1998)
37
REFERENCES
Abbot, P., Abe, J., Alcock, J., Alizon, S., Alpedrinha, J. A., Andersson, M., Andre, J.P., Van Balalen, M.,
Balloux, F., et al. 2011. Inclusive fitness theory and eusociality. Nature. 472: E1-E4.
Akaike, H. 1974. A new look at the statistical model identification. Automatic Control, IEEE Transactions.
19: 716-723.
Allee, W. C., Park, O., Emerson, A. E., Park, T., and Schmidt, K. P. 1949. Principles of Animal Ecology. WB
Saundere Co. Ltd.
Anthony, L. L., and Blumstein, D. T. 2000. Integrating behaviour into wildlife conservation: the multiple
ways that behaviour can reduce Ne. Biological Conservation. 95: 303-315.
Archie, E. A., Moss, C. J., and Alberts, S. C. 2006. The ties that bind: genetic relatedness predicts the
fission and fusion of social groups in wild African elephants. Proceedings of the Royal Society B:
Biological Sciences. 273: 513-522.
Arnold, K. E., and Owens, I. P. 1998. Cooperative breeding in birds: a comparative test of the life history
hypothesis. Proceedings of the Royal Society of London. Series B: Biological Sciences. 265: 739745.
Arnold, K. E., and Owens, I. P. 1999. Cooperative breeding in birds: the role of ecology. Behavioral
Ecology. 10: 465-471.
Avise, J. C., and Shapiro, D. Y. 1986. Evaluating kinship of newly settled juveniles within social groups of
the coral reef fish, Anthias squamipinnis. Evolution. 40: 1051-1059.
Bergmüller, R., Heg, D., and Taborsky, M. 2005. Helpers in a cooperatively breeding cichlid stay and pay
or disperse and breed, depending on ecological constraints. Proceedings of the Royal Society B:
Biological Sciences. 272: 325-331.
Brashares, J. S., and Arcese, P. 2002. Role of forage, habitat and predation in the behavioural plasticity of
a small African antelope. Journal of Animal Ecology. 71: 626-638.
Brown, C. R., and Brown, M. B. 1986. Ectoparasitism as a cost of coloniality in cliff swallows (Hirundo
pyrrhonota). Ecology. 67: 1206-1218.
Brown, J. L. 1987. Helping and communal breeding in birds. Princeton University Press.
38
Calvert, W. H., Hedrick, L. E., and Brower, L. P. 1979. Mortality of the monarch butterfly (Danaus
plexippus): avian predation at five overwintering sites in Mexico. Science. 204: 847-851.
Cant, M. A. 1998. A model for the evolution of reproductive skew without reproductive suppression.
Animal Behaviour. 55: 163-169.
Chapman, C. A., Chapman, L. J., and Wrangham, R. W. 1995. Ecological constraints on group size: an
analysis of spider monkey and chimpanzee subgroups. Behavioral Ecology and Sociobiology. 36:
59-70.
Chaverri, G. 2010. Comparative social network analysis in a leaf-roosting bat. Behavioral Ecology and
Sociobiology. 64: 1619-1630.
Clutton-Brock, T.H. 2002. Breeding together: kin selection and mutualism in cooperative
vertebrates. Science. 296: 69-72.
Clutton-Brock, T. H., Brotherton, P. N. M., O'riain, M. J., Griffin, A. S., Gaynor, D., Kansky, R., Sharpe, L.
and McIlrath, G. M. 2001. Contributions to cooperative rearing in meerkats. Animal Behaviour.
61: 705-710.
Clutton-Brock, T. H., Brotherton, P. N. M., O'Riain, M. J., Griffin, A. S., Gaynor, D., Sharpe, L., Kansky, R.,
Manser, M.B., and McIlrath, G. M. 2000. Individual contributions to babysitting in a cooperative
mongoose, Suricata suricatta. Proceedings of the Royal Society of London. Series B: Biological
Sciences. 267: 301-305.
Clutton-Brock, T. H., Gaynor, D., McIlrath, G. M., Maccoll, A. D. C., Kansky, R., Chadwick, P., Manser, M.,
Skinner, J.D., and Brotherton, P.N.M. 1999. Predation, group size and mortality in a cooperative
mongoose, Suricata suricatta. Journal of Animal Ecology. 68: 672-683.
Côté, I. M., and Poulin, R. 1995. Parasitism and group size in social animals: a meta-analysis. Behavioral
Ecology. 6: 159-165.
Courchamp, F., Clutton-Brock, T.H., and Grenfell, B. 1999. Inverse density dependence and the Allee
effect. Trends in Ecology & Evolution. 14: 405-410.
Croft, D. P., and Krause, M. S. B. J. 2004. Is sexual segregation in the guppy, Poecilia reticulata,
consistent with the predation risk hypothesis? Environmental Biology of Fishes. 71: 127-133.
Croft, D. P., James, R., Thomas, P. O. R., Hathaway, C., Mawdsley, D., Laland, K. N., and Krause, J. 2006.
Social structure and co-operative interactions in a wild population of guppies (Poecilia
reticulata). Behavioral Ecology and Sociobiology. 59: 644-650.
Curry, R. L. 1988. Group structure, within-group conflict and reproductive tactics in cooperatively
breeding Galápagos mockingbirds, Nesomimus parvulus. Animal Behaviour. 36: 1708-1728.
Davies, N. B., and Lundberg, A. 1984. Food distribution and a variable mating system in the dunnock,
Prunella modularis. The Journal of Animal Ecology. 53: 895-912.
39
Dunbar, R. I., and Shultz, S. 2007. Evolution in the social brain. Science. 317: 1344-1347.
Dunn, P. O., Cockburn, A., and Mulder, R. A. 1995. Fairy-wren helpers often care for young to which they
are unrelated. Proceedings of the Royal Society of London. Series B: Biological Sciences. 259:
339-343.
Ebensperger, L. A., and Bozinovic, F. 2000a. Communal burrowing in the hystricognath rodent, Octodon
degus: a benefit of sociality? Behavioral Ecology and Sociobiology. 47: 365-369.
Ebensperger, L. A., and Bozinovic, F. 2000b. Energetics and burrowing behaviour in the semifossorial
degu Octodon degus (Rodentia: Octodontidae). Journal of Zoology. 252: 179-186.
Ebensperger, L. A., and Hayes, L. D. 2008. On the dynamics of rodent social groups. Behavioural
Processes. 79: 85-92.
Ebensperger, L. A., Chesh, A. S., Castro, R. A., Tolhuysen, L. O., Quirici, V., Burger, J. R., and Hayes, L. D.
2009. Instability rules social groups in the communal breeder rodent Octodon degus. Ethology.
115: 540-554.
Ebensperger, L. A., Chesh, A. S., Castro, R. A., Tolhuysen, L. O., Quirici, V., Burger, J. R., Sobrero, R., and
Hayes, L. D. 2011. Burrow limitations and group living in the communally rearing rodent,
Octodon degus. Journal of Mammalogy. 92: 21-30.
Ebensperger, L. A., Hurtado, M. J., Soto-Gamboa, M., Lacey, E. A., and Chang, A. T. 2004. Communal
nesting and kinship in degus (Octodon degus). Naturwissenschaften. 91: 391-395.
Ebensperger, L. A., Tapia, D., Ramírez-Estrada, J., León, C., Soto-Gamboa, M., and Hayes, L. D. 2013.
Fecal cortisol levels predict breeding but not survival of females in the short-lived rodent,
Octodon degus. General and Comparative Endocrinology. 186: 164-171.
Ebensperger, L. A., Villegas, Á., Abades, S., and Hayes, L. D. 2014. Mean ecological conditions modulate
the effects of group living and communal rearing on offspring production and
survival. Behavioral Ecology. 25: 862-870.
Ebensperger, L. A., Sobrero, R., Quirici, V., Castro, R. A., Tolhuysen, L. O., Vargas, F., Burger, J.R., Quispe,
R., Villavicencio, C.P., Vasquez, R.A., and Hayes, L.D. 2012a. Ecological drivers of group living in
two populations of the communally rearing rodent, Octodon degus. Behavioral Ecology and
Sociobiology. 66: 261-274.
Ebensperger, L. A., Rivera, D. S., and Hayes, L. D. 2012b. Direct fitness of group living mammals varies
with breeding strategy, climate and fitness estimates. Journal of Animal Ecology. 81: 1013-1023.
Edenbrow, M., Darden, S. K., Ramnarine, I. W., Evans, J. P., James, R., and Croft, D. P. 2011.
Environmental effects on social interaction networks and male reproductive behaviour in
guppies, Poecilia reticulata. Animal Behaviour. 81: 551-558.
40
Emlen, S. T. 1982. The evolution of helping. I. An ecological constraints model. American Naturalist. 119:
29-39.
Emlen, S. T. 1995. An evolutionary theory of the family. Proceedings of the National Academy of
Sciences. 92: 8092-8099.
Emlen, S. T., and Oring, L. W. 1977. Ecology, sexual selection, and the evolution of mating
systems. Science. 197: 215-223.
Emlen, S. T., and Wrege, P. H. 1991. Breeding biology of white-fronted bee-eaters at Nakuru: the
influence of helpers on breeder fitness. The Journal of Animal Ecology. 60: 309-326.
Faulkes, C. G., and Bennett, N. C. 2001. Family values: group dynamics and social control of reproduction
in African mole-rats. Trends in Ecology & Evolution. 16: 184-190.
Faulkes, C. G., Bennett, N. C., Bruford, M. W., O'brien, H. P., Aguilar, G. H., and Jarvis, J. U. M. 1997.
Ecological constraints drive social evolution in the African mole–rats. Proceedings of the Royal
Society of London. Series B: Biological Sciences. 264: 1619-1627.
Field, J., Foster, W., Shreeves, G., and Sumner, S. 1998. Ecological constraints on independent nesting in
facultatively eusocial hover wasps. Proceedings of the Royal Society of London. Series B:
Biological Sciences. 265: 973-977.
Foster, E. A., Franks, D. W., Morrell, L. J., Balcomb, K. C., Parsons, K. M., van Ginneken, A., and Croft, D.
P. 2012. Social network correlates of food availability in an endangered population of killer
whales, Orcinus orca. Animal Behaviour. 83: 731-736.
Ginsberg, J. R., and Young, T. P. 1992. Measuring association between individuals or groups in
behavioural studies. Animal Behaviour. 44: 377-379.
Götmark, F., Winkler, D. W., and Andersson, M. 1986. Flock-feeding on fish schools increases individual
success in gulls. Nature. 319: 589-591.
Griesser, M., Nystrand, M., Eggers, S., and Ekman, J. 2008. Social constraints limit dispersal and
settlement decisions in a group-living bird species. Behavioral Ecology. 19: 317-324.
Grinnell, J., Packer, C., and Pusey, A. E. 1995. Cooperation in male lions: kinship, reciprocity or
mutualism? Animal Behaviour. 49: 95-105.
Guichón, M. L., Borgnia, M., Righi, C. F., Cassini, G. H., and Cassini, M. H. 2003. Social behavior and group
formation in the coypu (Myocastor coypus) in the Argentinean Pampas. Journal of Mammalogy.
84: 254-262.
Hamilton, W. D. 1964. The genetical evolution of social behaviour I. Journal of Theoretical Biology. 7: 116.
41
Hayes, L. D. 2000. To nest communally or not to nest communally: a review of rodent communal nesting
and nursing. Animal Behaviour. 59: 677-688.
Hayes, L. D., Burger, J. R., Soto-Gamboa, M., Sobrero, R., and Ebensperger, L. A. 2011. Towards an
integrative model of sociality in caviomorph rodents. Journal of Mammalogy. 92: 65-77.
Hayes, L. D., Chesh, A. S., and Ebensperger, L. A. 2007. Ecological predictors of range areas and use of
burrow systems in the diurnal rodent, Octodon degus. Ethology. 113: 155-165.
Hayes, L. D., Chesh, A. S., Castro, R. A., Tolhuysen, L. O., Burger, J. R., Bhattacharjee, J., and Ebensperger,
L. A. 2009. Fitness consequences of group living in the degu, Octodon degus, a plural breeder
rodent with communal care. Animal Behaviour. 78: 131-139.
Heinsohn, R. G. 1992. Cooperative enhancement of reproductive success in white-winged choughs.
Evolutionary Ecology. 6: 97-114.
Henzi, S. P., Lusseau, D., Weingrill, T., Van Schaik, C. P., and Barrett, L. 2009. Cyclicity in the structure of
female baboon social networks. Behavioral Ecology and Sociobiology. 63: 1015-1021.
Jones, K. E., Bielby, J., Cardillo, M., Fritz, S. A., O'Dell, J., Orme, C. D. L., Safi, K., Sechrest, W., Boakes, E.H.
et al. 2009. PanTHERIA: a species-level database of life history, ecology, and geography of extant
and recently extinct mammals. Ecology. 90: 2648-2648.
Kalinowski, S. T., Wagner, A. P., and Taper, M. L. 2006. ML-RELATE: a computer program for maximum
likelihood estimation of relatedness and relationship. Molecular Ecology Notes. 6: 576-579.
Kappeler, P.M. 2008. Genetic and ecological determinants of primate social systems. In Ecology and
Social Evolution. 225-243. Springer Berlin Heidelberg.
Kenward, R. E. 1978. Hawks and doves: factors affecting success and selection in goshawk attacks on
woodpigeons. The Journal of Animal Ecology. 47: 449-460.
Kokko, H., Johnstone, R. A., and Wright, J. 2002. The evolution of parental and alloparental effort in
cooperatively breeding groups: when should helpers pay to stay? Behavioral Ecology. 13: 291300.
Komdeur, J. 1992. Importance of habitat saturation and territory quality for evolution of cooperative
breeding in the Seychelles warbler. Nature. 358: 493-495.
Komdeur, J. 1996. Influence of age on reproductive performance in the Seychelles warbler. Behavioral
Ecology. 7: 417-425.
Komdeur, J., Huffstadt, A., Prast, W., Castle, G., Mileto, R., and Wattel, J. 1995. Transfer experiments of
Seychelles warblers to new islands: changes in dispersal and helping behaviour. Animal
Behaviour. 49: 695-708.
Krause, J., and Ruxton, G. D. 2002. Living in groups. Oxford University Press.
42
Kruuk, H. 1972. The spotted hyena: a study of predation and social behavior. University of Chicago Press.
Kuussaari, M., Saccheri, I., Camara, M., and Hanski, I. 1998. Allee effect and population dynamics in the
Glanville fritillary butterfly. Oikos. 82: 384-392.
Lacey, E. A. 2004. Sociality reduces individual direct fitness in a communally breeding rodent, the
colonial tuco-tuco (Ctenomys sociabilis). Behavioral Ecology and Sociobiology. 56: 449-457.
Lacey, E.A., and Sherman, P.W. 2007. The ecology of sociality in rodents. In Rodent societies: an
ecological and evolutionary perspective. 243-254. University of Chicago Press.
Lebigre, C., Alatalo, R. V., Forss, H. E., and Siitari, H. 2008. Low levels of relatedness on black grouse leks
despite male philopatry. Molecular Ecology. 17: 4512-4521.
Lott, D. F. 1991. Intraspecific variation in the social systems of wild vertebrates. Cambridge University
Press.
Lucia, K. E., Keane, B., Hayes, L. D., Lin, Y. K., Schaefer, R. L., and Solomon, N. G. 2008. Philopatry in
prairie voles: an evaluation of the habitat saturation hypothesis. Behavioral Ecology. 19: 774783.
Lukas, D., and Clutton-Brock, T.H. 2012. Life histories and the evolution of cooperative breeding in
mammals. Proceedings of the Royal Society B: Biological Sciences. 279: 4065-4070.
Macdonald, D. W. 1979. The flexible social system of the golden jackal, Canis aureus. Behavioral Ecology
and Sociobiology. 5: 17-38.
Magurran, A. E., and Seghers, B. H. 1991. Variation in schooling and aggression amongst guppy (Poecilia
reticulata) populations in Trinidad. Behaviour. 118: 214-234.
Matocq, M. D., and Lacey, E. A. 2004. Philopatry, kin clusters, and genetic relatedness in a population of
woodrats (Neotoma macrotis). Behavioral Ecology. 15: 647-653.
Maynard-Smith, J. 1964. Group selection and kin selection. Nature. 201: 1145-1147.
Meserve, P. L., Martin, R. E., and Rodriguez, J. A. I. M. E. 1984. Comparative ecology of the caviomorph
rodent Octodon degus in two Chilean mediterranean-type communities. Revista Chilena de
Historia Natural. 57: 79-89.
Nowak, M. A., Tarnita, C. E., and Wilson, E. O. 2010. The evolution of eusociality. Nature. 466: 10571062.
Ophir, A. G., Phelps, S. M., Sorin, A. B., and Wolff, J. O. 2007. Morphological, genetic, and behavioral
comparisons of two prairie vole populations in the field and laboratory. Journal of Mammalogy.
88: 989-999.
43
Quan, Y. F., MacManes, M. D., Ebensperger, L. A., Lacey, E. A., and Hayes, L. D. 2009. Isolation and
characterization of polymorphic microsatellite loci from Octodon degus. Molecular Ecology
Resources. 9: 999-1001.
Queller, D. C. 2000. Relatedness and the fraternal major transitions. Philosophical Transactions of the
Royal Society of London. Series B: Biological Sciences. 355: 1647-1655.
Quirici, V., Faugeron, S., Hayes, L. D., and Ebensperger, L. A. 2011. Absence of kin structure in a
population of the group-living rodent Octodon degus. Behavioral Ecology. 22: 248-254.
Quirici, V., Faugeron, S., Hayes, L. D., and Ebensperger, L. A. 2011. The influence of group size on natal
dispersal in the communally rearing and semifossorial rodent, Octodon degus. Behavioral
Ecology and Sociobiology. 65: 787-798.
Quispe, R., Villavicencio, C.P., Vasquez, R.A., and Hayes, L. D. 2012. Ecological drivers of group living in
two populations of the communally rearing rodent, Octodon degus. Behavioral Ecology and
Sociobiology. 66: 261-274.
Randall, J. A., Rogovin, K., Parker, P. G., and Eimes, J. A. 2005. Flexible social structure of a desert rodent,
Rhombomys opimus: philopatry, kinship, and ecological constraints. Behavioral Ecology. 16: 961973.
Rasa, O. A. E. 1989. The costs and effectiveness of vigilance behaviour in the dwarf mongoose:
implications for fitness and optimal group size. Ethology Ecology & Evolution. 1: 265-282.
Raymond, M., and Rousset, F. 1995. GENEPOP (version 1.2): population genetics software for exact tests
and ecumenicism. Journal of Heredity. 86: 248-249.
Rehan, S. M., Schwarz, M. P., and Richards, M. H. 2011. Fitness consequences of ecological constraints
and implications for the evolution of sociality in an incipiently social bee. Biological Journal of
the Linnean Society. 103: 57-67.
Reyer, H. U. 1984. Investment and relatedness: A cost/benefit analysis of breeding and helping in the
pied kingfisher (Ceryle rudis). Animal Behaviour. 32: 1163-1178.
Riehl, C. 2013. Evolutionary routes to non-kin cooperative breeding in birds. Proceedings of the Royal
Society B: Biological Sciences. 280: 1772.
Roberts, R. L., Williams, J. R., Wang, A. K., and Carter, C. 1998. Cooperative breeding and monogamy in
prairie voles: influence of the sire and geographical variation. Animal Behaviour. 55: 1131-1140.
Ross, K. G. 2001. Molecular ecology of social behaviour: analyses of breeding systems and genetic
structure. Molecular Ecology. 10: 265-284.
Schoepf, I., and Schradin, C. 2012. Better off alone! Reproductive competition and ecological constraints
determine sociality in the African striped mouse (Rhabdomys pumilio). Journal of Animal
Ecology. 81: 649-656.
44
Schradin, C. 2013. Intraspecific variation in social organization by genetic variation, developmental
plasticity, social flexibility or entirely extrinsic factors. Philosophical Transactions of the Royal
Society B: Biological Sciences. 368: 1618-1629.
Seppä, P., Fernández-Escudero, I., Gyllenstrand, N., and Pamilo, P. 2008. Colony fission affects kinship in
a social insect. Behavioral Ecology and Sociobiology. 62: 589-597.
Shine, R., and Fitzgerald, M. 1995. Variation in mating systems and sexual size dimorphism between
populations of the Australian python Morelia spilota (Serpentes: Pythonidae). Oecologia. 103:
490-498.
Shultz, S., Opie, C., and Atkinson, Q. D. 2011. Stepwise evolution of stable sociality in primates. Nature.
479: 219-222.
Sih, A., Hanser, S. F., and McHugh, K. A. 2009. Social network theory: new insights and issues for
behavioral ecologists. Behavioral Ecology and Sociobiology. 63: 975-988.
Silk, J. B. 2002. Kin selection in primate groups. International Journal of Primatology. 23: 849-875.
Silk, J. B. 2007. The adaptive value of sociality in mammalian groups. Philosophical Transactions of the
Royal Society B: Biological Sciences. 362: 539-559.
Smith, J. E. 2014. Hamilton's legacy: kinship, cooperation and social tolerance in mammalian
groups. Animal Behaviour. 92: 291-304.
Solomon, N. G. 2003. A reexamination of factors influencing philopatry in rodents. Journal of
Mammalogy. 84: 1182-1197.
Stacey, P. B., and Ligon, J. D. 1991. The benefits-of-philopatry hypothesis for the evolution of
cooperative breeding: variation in territory quality and group size effects. American Naturalist.
137: 831-846.
Stephens, P. A., and Sutherland, W. J. 1999. Consequences of the Allee effect for behaviour, ecology and
conservation. Trends in Ecology & Evolution. 14: 401-405.
Tanner, C. J., Salali, G. D., and Jackson, A. L. 2011. Feeding and non-feeding aggression can be induced in
invasive shore crabs by altering food distribution. Behavioral Ecology and Sociobiology. 65: 249256.
Thiel, M. 2011. The evolution of sociality: peracarid crustaceans as model organisms. New Frontiers in
Crustacean Biology. 285-297.
Travis, S. E., and Slobodchikoff, C. N. 1993. Effects of food resource distribution on the social system of
Gunnison's prairie dog (Cynomys gunnisoni). Canadian Journal of Zoology. 71: 1186-1192.
45
Travis, S. E., Slobodchikoff, C. N., and Keim, P. 1995. Ecological and demographic effects on intraspecific
variation in the social system of prairie dogs. Ecology. 76: 1794-1803.
Treherne, J. E., and Foster, W. A. 1980. The effects of group size on predator avoidance in a marine
insect. Animal Behaviour. 28: 1119-1122.
Trivers, R. L., and Hare, H. 1976. Haploidploidy and the evolution of the social insect. Science, 191: 249263.
Túnez, J. I., Guichón, M. L., Centrón, D., Henderson, A. P., Callahan, C., and Cassini, M. H. 2009.
Relatedness and social organization of coypus in the Argentinean pampas. Molecular Ecology.
18: 147-155.
Van Schaik, C. P., and Kappeler, P. M. 1997. Infanticide risk and the evolution of male–female association
in primates. Proceedings of the Royal Society of London. Series B: Biological Sciences. 264: 16871694.
Ward, P., and Zahavi, A. 1973. The importance of certain assemblages of birds as “information-centres”
for food-finding. Ibis. 115: 517-534.
West, S. A., Murray, M. G., Machado, C. A., Griffin, A. S., and Herre, E. A. 2001. Testing Hamilton's rule
with competition between relatives. Nature. 409: 510-513.
West, S. A., Pen, I., and Griffin, A. S. 2002. Cooperation and competition between relatives. Science. 296:
72-75.
Wey, T. W., Burger, J. R., Ebensperger, L. A., and Hayes, L. D. 2013. Reproductive correlates of social
network variation in plurally breeding degus (Octodon degus). Animal Behavior. 85: 1407-1414.
Wey, T., Blumstein, D. T., Shen, W., and Jordán, F. 2008. Social network analysis of animal behaviour: a
promising tool for the study of sociality. Animal Behaviour. 75: 333-344.
Whitehead, H. 2008. Analyzing animal societies: quantitative methods for vertebrate social analysis.
University of Chicago Press.
Whitehead, H. 2009. SOCPROG programs: analysing animal social structures. Behavioral Ecology and
Sociobiology. 63: 765-778.
Williams, G.C. 1966. Adaptation and natural selection. Princeton University Press.
Wilson, E. O. 2005. Kin selection as the key to altruism: its rise and fall. Social Research: An International
Quarterly. 72: 1-8.
Wolf, J. B., and Trillmich, F. 2008. Kin in space: social viscosity in a spatially and genetically substructured
network. Proceedings of the Royal Society B: Biological Sciences. 275: 2063-2069.
46
Wolf, J. B., Mawdsley, D., Trillmich, F., and James, R. 2007. Social structure in a colonial mammal:
unravelling hidden structural layers and their foundations by network analysis. Animal
Behaviour. 74: 1293-1302.
Wolff, J. O. 1992. Parents suppress reproduction and stimulate dispersal in opposite-sex juvenile whitefooted mice. Nature. 359: 409-410.
47
APPENDIX A
SOCIAL GROUPS
48
ID = individual’s ear tag number, individuals in the same column represent social groups
Rinconada 2007:
ID
ID
ID
ID
ID
ID
ID
1012
4052
2434
2223
1155
4002
2202
2454
4012
143
4441
1255
1111
134
2352
4010
3323
1312
3325
2453
2341
155
1253
1314
2053
4443
1041
1445
154
1354
3011
141
Rinconada 2008:
ID
ID
ID
315
4325 1021
1023
3400 410
240
251
100
Los Molles 2007:
ID
4243
1041
3252
ID
4144
1311
4322
ID
1015
243
54
ID
1
2200
1200
54
Los Molles 2008:
ID
3062
3037
3061
ID
3066
3049
ID
3068
2200
3036
3050
3069
810
ID
2489
3046
3041
3042
2488
3040
400
ID
2418
3100
3063
3045
3070
3071
49
ID
3355
124
ID
412
252
ID
4422
4423
4455
ID
3221
255
3005
214
APPENDIX B
GENETIC PROCEDURES
50
DNA Extraction Protocol:
1) Cut tissue sample into small pieces
2) Heat sample at 50 degrees C for 30 minutes
3) Add 440ml extraction buffer
4) Add 44ul SDS 20%
5) Add 12ul Proteinase K
6) Vortex for ~30 seconds
7) Heat at 60 degrees C for 60 minutes
8) Add 300ul 6M saline solution
9) Vortex for at least 60 seconds
10) Centrifuge for 30 minutes at 10,000 RPM and 4 degrees C
11) Pipette 750ul into new tubes
12) Add 750 ul isopropanol
13) Invert tubes to mix
14) Incubate at -20 degrees C for 30 minutes
15) Centrifuge for 20 minutes at 13,000 RPM and 4 degrees C
16) Pour out excess solution, careful to keep DNA precipitate in tube
17) Dry inside of tube with paper towel
18) Add 200ul 70% ethanol
19) Centrifuge for 10 minutes at 13,000 RPM and 4 degrees C
20) Remove excess liquid again with paper towel
21) Dry sample for 1-2 hours at 60 degrees C
22 Dissolve DNA in 100-300ul of water
PCR Primer Mixture Volumes and Annealing Temperatures:
Locus
OCDE3
OCDE6
OCDE11
OCDE12
Mg
2.0
1.75
1.75
1.75
dTNP
0.8
0.4
0.4
0.4
Primer
0.8
0.53
0.53
0.53
Taq
0.75
0.5
0.5
0.5
DNA
30
30
30
50
Temp
62
58
58.5
63
OCDE13
1.5
0.2
0.53
0.5
20
58
Gel Electrophoresis:
1) To make 1.5% agarose gel, mix 1.5g agarose and 100ml TAE 1x in flask and heat until completely
dissolved
2) Add 2.5ul ethidium bromide to flask
3) Pour mixture into gel box, allow to solidify for ~45 minutes
4) Fill electrophoresis chamber with TAE 1x, add gel
5) Put drops of 2ul buffer on parafilm (1 for each sample)
6) Add 4ul PCR product for each sample to drops, then load in gel (leaving first slot for molecular ladder)
51
7) Add 2 ul ladder to 2ul buffer and load in first slot
8) Run electrophoresis for 45 minutes at 110 volts
52
APPENDIX C
GENETIC DATA
53
Mean Relatedness of Social Groups:
2007 Rinconada:
Grp
1
2
3
4
5
6
7
8
9
Relatedness
0.186
0.082
0.148
0.166
0.139
0.114
0.088
0.112
0.069
2008 Rinconada:
Grp
1
2
3
4
5
6
7
8
9
10
Relatedness
0.114
0.071
0.212
0.155
0.088
0.11
0.122
0.136
0.094
0.073
2007 Los Molles:
Grp
1
Relatedness
0.094
54
2008 Los Molles:
1
2
3
4
5
0.091
0.213
0.246
0.89
0.102
Background Relatedness Confidence Intervals:
GS = group size, 95 CI = 95% confidence interval
Rinconanda:
GS
3
7
6
4
3
3
6
5
2
2
3
3
3
2
2
3
4
95 CI
0-0.31
0-0.21
0-0.22
0-0.25
0-0.31
0-0.31
0-0.22
0-0.24
0-0.41
0-0.41
0-0.31
0-0.31
0-0.31
0-0.41
0-0.41
0-0.31
0-0.25
Los Molles:
GS
4
6
2
3
6
95 CI
0-0.23
0-0.23
0-0.46
0-0.33
0-0.23
55
7
0-0.23
Primer data:
Ho and He are the observed and expected levels of heterozygosity.
Rinconada:
Primer
OCDE 3
OCDE 6
OCDE 11
OCDE 12
OCDE 13
#
alleles
8
5
8
7
12
Ho
0.71
0.45
0.59
0.47
0.79
He
0.84
0.74
0.63
0.72
0.84
P value
0.38
<0.01
0.44
<0.01
0.29
#
alleles
7
6
9
6
10
Ho
0.54
0.48
0.67
0.51
0.91
He
0.66
0.79
0.79
0.68
0.81
P value
0.32
<0.01
0.28
<0.01
0.52
Los Molles:
Primer
OCDE 3
OCDE 6
OCDE 11
OCDE 12
OCDE 13
56
APPENDIX D
ECOLOGICAL DATA
57
Rinconada
Biomass Burrows
(g/m2)
(#/m2)
79.27
0.16
81.51
0.15
106.83
0.07
108.42
0.14
93.10
0.12
87.89
0.17
111.73
0.06
107.73
0.06
100.73
0.07
124.68
0.18
146.16
0.08
133.36
0.19
121.38
0.16
131.52
0.18
111.23
0.13
99.87
0.14
131.16
0.16
130.49
0.16
87.44
0.18
83.14
0.17
110.96
0.12
91.84
0.15
84.55
0.14
79.05
0.17
79.31
0.11
89.17
0.16
99.07
0.06
99.03
0.15
62.84
0.17
114.53
0.15
107.64
0.08
Soil
(kPa)
2971.90
3052.01
2976.81
3059.76
3067.14
2988.10
2977.22
3001.27
3044.40
3086.28
2932.89
3055.96
3094.33
3079.69
3112.47
3095.54
3067.14
3068.93
2956.95
2963.87
3057.11
3057.11
3062.32
3051.46
3130.77
3047.84
3070.28
3049.44
3064.29
3116.28
3056.33
82.49
133.45
126.09
3037.87
3055.74
3053.63
0.16
0.19
0.19
58
82.56
91.81
65.94
118.78
99.67
104.71
116.08
126.78
132.93
58.33
105.26
74.60
92.80
148.48
101.73
69.26
281.60
242.24
117.60
92.80
117.60
133.68
281.60
136.34
54.46
196.10
130.83
92.80
61.94
0.17
0.14
0.17
0.17
0.16
0.15
0.14
0.17
0.17
0.18
0.07
0.14
0.07
0.09
0.13
0.13
0.09
0.10
0.10
0.07
0.10
0.09
0.09
0.10
0.17
0.13
0.08
0.07
0.15
3015.70
3096.29
3088.38
3062.85
3087.36
3052.50
3042.05
3084.64
3046.01
3104.00
3008.44
3121.56
3089.10
3038.29
3133.80
3116.19
3060.87
3050.98
2919.72
3089.10
2919.72
3052.05
3060.87
2935.85
3111.54
3106.15
2950.81
3089.10
3126.74
117.60
57.29
139.09
164.20
0.10
0.09
0.14
0.13
2919.72
3126.33
3139.20
3112.62
281.60
281.60
84.75
87.81
111.20
0.09
0.09
0.09
0.13
0.09
3060.87
3060.87
3137.26
2940.05
3131.97
59
126.52
63.60
154.49
91.89
96.36
82.61
85.34
241.45
0.14
0.10
0.16
0.10
0.10
0.14
0.11
0.09
3137.54
3123.97
3145.56
3139.73
3130.59
3130.21
3130.13
3016.62
Los Molles
Biomass Burrows
(g/m2)
(#/m2)
0.00
0.15
0.00
0.15
0.00
0.15
0.00
0.15
509.71
0.16
39.20
0.20
39.20
0.20
0.00
0.27
296.08
0.16
601.60
0.16
39.20
0.20
8.80
0.20
376.48
0.16
568.18
0.16
340.30
0.16
0.00
0.27
0.00
0.27
280.00
0.16
134.40
0.11
39.20
0.20
8.80
0.20
8.80
0.20
0.00
0.27
134.40
0.11
39.20
0.20
39.20
0.20
0.00
0.27
Soil
(kPa)
2428.65
2428.65
2428.65
2428.65
1585.38
2180.09
2180.09
1186.40
2180.47
1329.43
2180.09
1111.12
1956.51
1341.56
2057.29
1186.40
1186.40
2225.26
1743.47
2180.09
1111.12
1111.12
1186.40
1743.47
2180.09
2180.09
1186.40
60
APPENDIX E
NETWORK DATA
61
Rinconada:
ID (sex)
0124 (F)
0134 (F)
0141 (F)
0143 (F)
0151 (F)
0153 (F)
0154 (F)
0155 (F)
0432 (M)
1012 (M)
1014 (F)
1033 (F)
1041 (F)
1042 (F)
1103 (F)
1111 (F)
1115 (M)
1155 (F)
1251 (F)
1252 (M)
1253 (F)
1255 (F)
1311 (F)
1312 (F)
1314 (F)
1354 (F)
1412 (M)
1445 (M)
2053 (F)
2054 (F)
2202 (F)
2223 (F)
2341 (F)
2352 (F)
Strength
1.25
2.84
2.34
1.43
1.62
1.76
1.38
2.57
2.41
2.62
1.83
1.61
1.53
3.57
2.9
1.79
2.78
2.37
0.87
0.41
2.78
2.35
1.89
2.79
2.26
2.98
2.99
1.92
2.73
2.03
3
2.36
2.1
2.4
Eig Centrality
0
0.01
0
0
0
0
0
0
0
0
0.1
0.01
0.21
0.51
0.02
0.03
0.02
0
0
0.02
0.38
0.03
0.04
0
0
0.44
0.02
0.01
0.02
0
0.02
0
0.01
0
Reach
2.85
6.85
4.96
2.23
3.32
3.76
3.24
5.3
5.36
5.51
4.91
2.64
4.12
9.49
6.11
4.27
5.82
5.54
1.21
0.66
7.17
4.91
4.12
5.72
4.65
8.22
8.34
4.11
7.22
5.23
6.94
6.15
5.29
5.2
2434 (F)
2452 (F)
2453 (F)
2454 (F)
3011 (F)
1.61
1.85
3.05
2.02
2.04
0.91
0
0
0.06
0
0.01
0.04
2.85
4.14
5.87
4.48
4.77
2.54
62
Clustering
Coeff
0.26
0.22
0.23
0.18
0.21
0.16
0.34
0.19
0.26
0.18
0.14
0.2
0.31
0.36
0.38
0.41
0.37
0.24
0.07
0
0.39
0.24
0.35
0.21
0.29
0.43
0.19
0.15
0.22
0.2
0.21
0.2
0.25
0.23
Affinity
2.28
2.41
2.12
1.56
2.05
2.14
2.35
2.06
2.22
2.1
2.68
1.64
2.69
2.66
2.11
2.39
2.09
2.34
1.4
1.62
2.58
2.09
2.18
2.05
2.06
2.76
2.79
2.14
2.65
2.57
2.31
2.6
2.52
2.17
0.16
0.2
0.19
0.22
0.24
0
1.77
2.23
1.92
2.22
2.34
2.79
3(34)15
3102 (F)
3114 (F)
3210 (F)
3251 (F)
3323 (M)
3325 (M)
4002 (M)
4010 (F)
4012 (F)
4043 (F)
4052 (M)
4114 (M)
4201 (M)
4311 (M)
4441 (F)
4443 (F)
1.75
1.98
1.89
1.33
1.98
2.27
1.59
2.24
2.59
1
2.54
1.23
1.68
2.3
3.36
1.53
ID (sex)
0001(F)
0043(M)
0054(M)
0100(M)
0124(F)
0130(F)
0203(F)
0214(F)
0215(F)
0220(F)
0240(F)
Strength
1.13
0
2
1.3
0.1
0
1.07
2.88
1.39
1
3.31
Eig
Centrality
0
0
0
0.2
0
0
0
0.1
0
0.17
0.47
Reach
1.26
0
4.42
4.44
0.11
0
1.19
8.45
1.76
4.13
10.73
Clustering
Coeff
0.27
NaN
0.92
0.87
NaN
NaN
0.24
0.88
0.1
NaN
0.61
Affinity
1.12
NaN
2.21
3.42
1.08
NaN
1.11
2.93
1.27
4.13
3.24
0243(F)
0251(F)
0252(F)
0255(F)
0315(F)
0410(F)
0411(F)
0412(F)
948(M)
1015(F)
1.99
3.32
2.07
3
4.13
1.35
0.89
1.92
0.5
2.42
0
0.46
0.1
0.09
0.49
0
0
0.13
0
0
4.38
10.31
4.02
8.57
10.35
1.93
0.59
4.61
1.21
4.09
0.87
0.57
0.33
0.88
0.33
0.24
0.01
0.4
NaN
0.49
2.2
3.11
1.94
2.86
2.5
1.43
0.67
2.4
2.42
1.69
0.01
0
0.02
0
0
0
0.02
0.32
0.31
0
0.37
0.01
0.01
0
0.01
0
4.9
4.13
5.18
2.85
4.4
5.8
3.33
6.31
6.27
1.52
7.09
2.07
2.47
5.55
7.77
3.58
63
0.81
0.28
0.59
0.09
0.16
0.35
0.21
0.41
0.27
0.5
0.45
0.31
0.17
0.31
0.21
0.2
2.8
2.09
2.74
2.15
2.22
2.55
2.09
2.82
2.42
1.52
2.79
1.69
1.47
2.41
2.31
2.34
1021(M)
1023(F)
1041(F)
1311(F)
1445(M)
2040(F)
3005(F)
3221(F)
3252(M)
3323(M)
3355(F)
3400(F)
4144(F)
4243(F)
4322(F)
4325(F)
4422(F)
4423(F)
4455(F)
022(01)(F)
13(15)1(F)
1.48
3.14
0.56
0.41
0.7
1.56
2.83
2.85
1.48
2
0.07
0.67
0.66
0.53
0.51
1.31
1.22
1.08
0.64
1.95
1.95
0
0.45
0
0
0
0.06
0.09
0.09
0
0.01
0
0
0
0
0
0
0
0
0
0.02
0.03
1.98
10.02
0.64
0.31
0.62
3.13
8.26
8.11
1.32
3.9
0.13
0.87
0.49
0.61
0.33
1.38
1.27
1.11
0.81
3.86
3.97
0.19
0.63
0.33
0.17
NaN
0.71
0.94
0.81
0.03
0.73
NaN
NaN
0.07
0.36
0.1
0.11
0.23
0.2
0.15
0.66
0.65
1.34
3.2
1.14
0.77
0.89
2.02
2.92
2.84
0.89
1.95
1.95
1.31
0.74
1.14
0.65
1.05
1.04
1.03
1.27
1.98
2.04
Los Molles
ID (sex)
0001(F)
0011(F)
0024(M)
0054(F)
0055(F)
0100(F)
0300(F)
1000(F)
1200(F)
1300(M)
2200(F)
2400(F)
3100(F)
3300(F)
4100(F)
Strength
4.5
3
4
4.5
1.33
1.17
4
4
5
3
4.67
1.83
2
1.33
5
Eig Centrality
0.4
0
0.36
0.4
0
0
0
0
0.43
0
0.41
0
0
0
0.43
ID (sex)
Strength
Eig Centrality
Reach
21.08
11
19.11
21.08
1.83
1.81
12
12
22.67
11
21.67
2.36
8
2.28
22.67
Reach
64
Clustering
Coeff
0.95
1
0.99
0.95
0.4
0.33
0.67
0.67
0.88
1
0.93
0.22
1
0.33
0.88
Clustering
Coeff
Affinity
4.69
3.67
4.78
4.69
1.38
1.55
3
3
4.53
3.67
4.64
1.29
4
1.71
4.53
Affinity
0400(F)
2200(F)
2418(F)
2488(F)
2489(F)
3036(F)
3037(F)
3040(M)
3041(F)
3042(F)
3043(F)
3044(F)
3045(M)
3046(F)
3.54
6
7.75
2.88
4.14
5
2
4.32
4.15
3.46
2
6.92
8
3.08
0
0
0.36
0
0
0
0
0
0
0
0
0.3
0.37
0
13.69
34
51.6
10.25
16.61
30
4
15.52
16.79
12.38
4
43.92
53.83
10.82
0.49
0.93
0.77
0.69
0.69
1
1
0.41
0.7
0.56
1
0.66
0.76
0.66
3.87
5.67
6.66
3.56
4.01
6
2
3.6
4.05
3.58
2
6.35
6.73
3.51
3048(F)
3049(F)
3050(M)
3061(M)
3062(F)
3063(F)
3066(F)
3068(F)
3069(F)
3070(F)
3071(M)
3072(M)
3074(F)
3099(F)
3100(F)
4400(F)
801(F)
806(F)
810(F)
3100d(F)
2
1
6
2
2
7
1
6
6
7.67
7
7
2
4.33
3
5
4.5
6
6
4.03
0
0
0
0
0
0.34
0
0
0
0.36
0.35
0.35
0
0
0.12
0
0.19
0.31
0
0
4
1
34
4
4
49.13
1
34
34
51.11
50.33
50.33
4
17.79
19.13
30
29.12
44.42
34
16.43
1
NaN
0.93
1
1
0.87
NaN
0.93
0.93
0.77
0.9
0.9
1
0.55
0.76
1
0.69
1
0.93
0.74
2
1
5.67
2
2
7.02
1
5.67
5.67
6.67
7.19
7.19
2
4.11
6.38
6
6.47
7.4
5.67
4.07
65
APPENDIX F
STATISTICAL ANALYSIS
66
Bootstrapping code for calculating confidence intervals of background relatedness, using R v. 3.1:
> bstrap <- c()
> for (i in 1:1000){
+ # take the sample
+ bsample <- sample(x,7,replace=T) (x previously defined as “relatedness” from column in spreadsheet)
+ # calculate the bootstrap (here n=7 for GS=7)
+ bestimate <- mean(bsample)
+ bstrap <- c(bstrap,bestimate)}
> #lower bound for 95%:
> quantile(bstrap,.025)
> #upper bound for 95%
> quantile(bstrap,.975)
AIC Code for SAS v. 9.3:
Data (GS= group size):
DATA groupr;
INPUT Population GS Relatedness Association Strength;
DATALINES;
1 3 0.19 0.57 1.51
1 7 0.08 0.30 2.12
1 6 0.15 0.28 2.44
1 4 0.17 0.23 2.69
1 3 0.14 0.06 2.33
1 3 0.11 0.39 2.14
1 6 0.09 0.27 2.44
1 5 0.11 0.91 3.48
1 2 0.07 0.50 0.99
1 2 0.21 0.25 1.42
1 3 0.16 0.31 0.86
1 3 0.09 0.32 0.53
1 3 0.11 0.96 2.14
1 2 0.12 0.83 2.00
1 2 0.14 0.35 1.10
1 3 0.09 0.20 0.98
1 4 0.07 0.82 2.89
2 6 0.09 1.00 6.79
2 2 0.21 1.00 1.00
2 3 0.25 1.00 2.00
2 6 0.09 0.92 5.83
2 7 0.10 0.44 3.65
2 4 0.09 0.94 4.67
;
Strength Analysis:
67
proc reg data=groupr outest=est;
model Strength = population GS relatedness/ selection=adjrsq sse aic ;
output out=out p=p r=r; run; quit;
proc reg data=groupr outest=est0;
model Strength = population GS relatedness/ noint selection=adjrsq sse aic ;
output out=out0 p=p r=r; run; quit;
data estout;
set est est0; run;
proc sort data=estout; by _aic_;
proc print data=estout(obs=23); run;
Association Analysis:
proc reg data=groupr outest=est;
model Association = population GS relatedness/ selection=adjrsq sse aic ;
output out=out p=p r=r; run; quit;
proc reg data=groupr outest=est0;
model Association = population GS relatedness/ noint selection=adjrsq sse aic ;
output out=out0 p=p r=r; run; quit;
data estout;
set est est0; run;
proc sort data=estout; by _aic_;
proc print data=estout(obs=23); run;
68
VITA
Garrett Davis was born in Springville, New York to the parents of Margaret Ronan and Dennis
Davis. He attended Pioneer High School in Yorkshire, New York and, after graduation, obtained a
Bachelor of Science degree in Wildlife Science from the SUNY College of Environmental Science &
Forestry in Syracuse, New York. Garrett conducted fieldwork in Alabama, California, and Costa Rica
before entering the Environmental Sciences graduate program at the University of Tennessee at
Chattanooga in January 2012. He will graduate with his Master of Science degree in Environmental
Science in December 2014.
69