THE ROLE OF CUTICULAR HYDROCARBONS AS
SOCIAL RECOGNITION CUES IN
THE PAVEMENT ANT
(TETRA MOR1UM CAESPITUM)
B.S., University of Colorado at Denver, 2003
A thesis submitted to the
University of Colorado at Denver and Health Sciences Center
in partial fulfillment of the
requirements for the degree of
Master of Integrated Science
This thesis for the Master of Integrated Science
has been approved
Dr. David S. Albeck. Ph.D.
Sano, Kazuhiro (Master, Integrated Science)
The role of Cuticular Hydrocarbons as Social Recognition Cues in the Pavement Ant
Thesis directed by Assistant Professor Michael J. Greene
The primary function of cuticular hydrocarbons is to provide protection against water
loss. However, ants utilize cues present in cuticular hydrocarbon profiles to recognize
and discriminate nestmates from non-nestmates. In this study, the role of cuticular-
hydrocarbon-based recognition cues in regulating the social recognition of the ant
Tetramorium caespitum was investigated. Nestmate recognition, the recognition of
conspecifics that belong to a colony, using cues in cuticular hydrocarbons was
investigated along with the ants ability to use hydrocarbon-based cues in species
recognition. The effects of social isolation, temperature and humidity on the
expression of recognition cues were examined. Furthermore, the anti-desiccation
property of cuticular hydrocarbon was also measured. I demonstrated that nestmate
recognition and species recognition of T. caespitum are mediated by a cue present in
cuticular hydrocarbon profiles. The proportion of aggressive ants towards
hydrocarbon stimuli from non-nestmates was significantly higher than towards
hydrocarbon stimuli of nestmates. When focal ants were introduced to the non-
nestmate or heterospecific hydrocarbon stimuli, they responded with significantly
higher levels of aggression compared to the hydrocarbon stimuli from nestmates.
Four weeks of social isolation altered the nestmate recognition cues of T. caespitum.
There were no significant differences in the affects of temperature and humidity
levels. I also demonstrated the anti-desiccation property of the cuticular
hydrocarbons. When the hydrocarbon extract was applied to the artificial membrane,
the amount of water loss via evaporative water loss was significantly decreased by
54%. I conclude that the cues used in nestmate and species recognition responses of
T. caespitum are present in cuticular hydrocarbons and that social isolation, and
possibly environmental factors, affects the expression of social recognition cues in
This abstract accurately represents the content of the candidates thesis. I recommend
My thanks to my advisor, Dr. Michael J. Greene for his patience with me for
courses of graduate work, all the training and advice for my thesis projects.
2 Methods and Materials.............................................15
2.1 Field Sites.......................................................15
2.2 Collecting Pavement Ants T. caespitum.............................15
2.3 Care of Captive Ants..............................................15
2.4 Extraction and Purification of the Cuticular Hydrocarbons.........16
2.5 Preparation of Experimental Stimuli...............................16
2.6 Behavioral Bioassay for Nestmate and Species Recognition..........16
2.7 Ethogram Score....................................................17
2.8 Gas Spectrometry/Mass Spectometry.................................19
2.9 Statistical Analysis..............................................19
2.10 Experiment for Aim 1.............................................19
2.11 Experiment for Aim 2.............................................20
2.12 Experiment for Aim 3.............................................22
3.1 Results for Aim 1.................................................24
3.2 Results for Aim 2.................................................34
3.3 Results for Aim 3.................................................38
4.2 Recognition Behavior and Hydrocarbon..............................42
4.3 Significance of Social Interaction................................45
4.4 Physiological Function of Hydrocarbon.............................46
LIST OF FIGURES
1.1 The pavement ant T. caespitum..........................................14
2.2 Screw-top vial setup for water loss experiment.........................23
3.1.1 The proportion of individuals who responded with aggression
between nestmate and non-nestmate hydrocarbon.................27
3.1.2 The degree of aggression towards hydrocarbon stimuli of nestmate
3.1.3 The degree of aggression towards each experimental group...............32
3.2.1 Decrease in the mean ethogram scores of focal ants towards isolated
3.2.2 The average temperatures in C and average relative humidity in %
among 4 treatment group.................................................37
3.3.1 The amount (g) of water loss between two experimental groups at
3.3.2 The average water loss (g) between the snake skins treated with HC
extract from T. caespitum and snake skins without HC treatment..........41
LIST OF TABLES
2.1 The description and sample images of T. caespitum behavior
to define ethogram scores...............................................18
3.1.1 The number of ants contacted HC, number of ants responded
with aggression, and their proportions..................................28
3.1.2 Descriptove statistics for comparison of mean ethogram
scores for nestmate recognition.........................................30
3.1.3 Comparisons of mean ethogram scores to the HC profile
from different colonies.................................................31
3.1.4 Descriptive statistics for nestmate and heterospecific recognition.....33
3.2.1 The change in the level of aggression toward isolated
nestmate with duration of time..........................................36
3.3.2 Means and raw scores for how much water lost in gram...................40
The aim of this research was to study the mechanisms of social recognition in the
Pavement Ant (Tetramorium caespitum), an invasive pest distributed across Eurasia
and North America. Specifically, it was my goal to study how T. caespitum
recognizes non-nestmate conspecific ants and heterospecific ants using cues present
in cuticular hydrocarbons, and how social isolation and abiotic environmental
variables affect the expression and perception of these cues. Further, it was my aim
to measure the anti-desiccation properties of the cuticular hydrocarbon, the primary
function of these chemicals.
The specific research aims are as follows:
1. To determine whether T. caespitum recognizes and discriminates nestmates from
non-nestmate conspecifics and from heterospecific ants using cues present in
2. To measure the effect of social isolation and abiotic factors on the expression of
recognition cues present in cuticular hydrocarbon profile.
3. To investigate the anti-desiccation properties of cuticular hydrocarbons.
The ability of the basic units that comprise living systems to recognize the group
membership of other units is ubiquitous. For example, the Old World monkeys of the
genus Cercopithecoidea are known for their complex social relations, including
dominance hierarchies and non-kin alliances, and their social behavior reflects the
ability to identify and recognize both kin and non-kin individuals (Cheney et al.,
1986). On a much smaller scale, our bodies are able to recognize and discriminate
cells and tissues that are foreign through the function of immune systems (Golsby et
al., 2003). From the single cell level to the society level, the ability to recognize the
group membership of other units is one of the fundamental organization principles to
maintain cohesive social or organized structures.
Recognition systems involve three components: 1) the expression component, 2) the
perception component, and 3) the action component. Recognition systems require a
minimum of two participants called the cue-bearer' and the evaluator (Sherman &
Holmes, 1985, Fletcher, 1987, Reeve, 1989, Sherman et al., 1997, Liebert & Starks,
2004, Mateo, 2004, Tsutsui, 2004). When a cue-bearer expresses cues to an
evaluator, cues are perceived by evaluators through sensory mechanisms. Then, an
evaluator processes and evaluates cues by comparing them with an internal template.
An appropriate response by the evaluator, i.e., the action component, results from
matching between the information in the cue and the template (Liebert & Starks,
2004, Mateo, 2004, Tsutsui, 2004). Templates are often thought to be part of
memory in the central nervous system. Both cues and template can be learned,
genetically determined, or both (Liebert & Starks, 2004, Mateo, 2004, Tsutsui, 2004).
Identifying a cue bearer as member of a group or as a foreigner depends on if the
cue matches the template or not (Liebert & Starks, 2004, Mateo, 2004, Tsutsui,
Recognition systems allow the basic units of living systems to form cohesive
organization structures by allowing them to respond appropriately to changes in their
environment. Holldobler and Wilson (1994) argued that, social organization has
been one of the most consistently successful strategies in all of evolutionary history.
The organization of social insect societies is a well known example of how units of a
system interact to form a larger living system that exhibits a cohesive social structure
(Holldobler and Wilson, 1990, 1994, Wilson, 1975). Social insects include the
termites (order Is op ter a), bees (superfamily Apoidae), wasps (order Hymenoptera)
and ants (family Formicidae).
Most of the over 12,000 species of Formicidae, all of the about 2,000 species of
termite (Isoptera), and the approximately 800 species in the wasp family Vespidae
exhibit eusociality. The Apidae, honey bees, bumble bees, and stingless bees, also
exhibit eusociality (Holldobler & Wilson, 1994). Eusocial animals exhibit the
following traits: 1) cooperation in caring of young, 2) reproductive division of labor,
and 3) the overlap of more than two generations of life stages contributing to the
colony labor (Wilson, 1975). Eusociality is thought to have led to the tremendous
success of social insects. Social insects can be found on most terrestrial habitats, and
all the social insects together make up 80% of the insect biomass (Holldobler &
Social insect colonies are structured with either a single queen (monogyny) or
multiple queens (polygyny) and up to hundreds of thousands of workers that are
sisters who are highly genetically related (Wilson, 1975). Within a colony, the queen
is specialized for reproduction, producing workers and new queens. Queens typically
mate after leaving their mother colony, using stored sperm to produce workers for the
rest of their lives (Holldobler & Wilson, 1990). Workers within a colony perform the
jobs necessary to ensure the survival and reproductive capacity of the colony (Greene
& Gordon, 2003, Robinson, 2001, Seeley, 2001, Wagner et al., 2001, & Wilson,
1975). Tasks of workers can be categorized into three larger categories: 1) foraging
for food, 2) nest maintenance, and 3) taking care of eggs and broods (Greene &
Gordon, 2003, Robinson, 2001, Seeley, 2001, Wagner et al., 2001, & Wilson, 1975).
Workers and queens are closely related and a colony works as a unit often referred as
the superorganism (Wilson, 1975). The term superorganism was devised by
Wheeler (1928) to address the similarity between the organization of eusocial
colonies and the physiological processes of individual organisms. The superorganism
theory helps to explain the evolutionary stability of eusocial structure by focusing on
the reproductive units (i.e., colonies) as individuals and sterile individuals as
independent parts of the larger system, much like cells a body (Holldobler & Wilson,
1990). This theory explains eusociality from queens (reproductive individuals)
perspective while kin selection and inclusive fitness explains social behavior of
workers. The members within the colony is essential for the eusociality.
Kin selection refers to the selection of genetic alleles across generations due to
preferred treatment among related individuals (Wilson, 1975). Kin selection is based
on the idea of inclusive fitness (Hamilton, 1964). Inclusive fitness explains altruistic
behavior as a way to maximize fitness by taking common genes passing onto next
generation into the account of fitness. In other words, the likelihood of genetic alleles
passing on to the next generation is included in the outcome of ones fitness rather
than counting only the number of viable offspring surviving to reproduce in the next
generation as a fitness (Hamilton, 1964). It can be used to explain the presumably
altruistic behavior by sterile workers towards the reproductively active queen.
Eusocial colonies exhibit high levels of reproductive skew (Holldobler &Wilson,
1990, Cuviller-Hot et al., 2004, Faulkes & Bennett, 2001,). Reproductive skew refers
to a social structure in which reproduction is partitioned unequally among members
of a group or there is a reproductive division of labor present in a group (Faulkes &
Bennett, 2001). Social insect colonies exhibit a high degree of reproductive skew
because the many workers are normally sterile and the single queen or few queens of
a colony reproduce (Holldobler & Wilson, 1990). The workers are considered non-
reproductive castes among social insects (Wilson, 1975). The dominance over
reproduction is often present in eusocial colonies and reproductive individuals control
reproduction of subordinate with various way: infanticide of offspring of
subordinates, interference by dominants with subordinate mating attempts or
suppression of subordinate reproductive physiology (Faulkes and Abbott, 1997).
Dominants among social insects tend to control reproduction of subordinates by
suppressing their reproductive physiology and dominants control reproductive
physiology of subordinates chemically or behaviorally to reduce the risk to the
dominants (Winston and Slessoe, 2001, Cuvillier-Hot et al., 2004, Howard and
Blomquist, 2005). In most social insect species, behavioral dominance is not
necessary because workers are sterile.
Haplodiploidy is a mode of sex determination used by social insects in which
unfertilized eggs give rise to haploid males while females rise from fertilized eggs as
diploids. Haplodiploidy maintains the close relatedness among group members as
well as reproductive skew. This mode of sex determination is known to give as high
as 75% of relatedness among sisters. In contrast, daughters and mothers share up to
50% of their genetic information (Holldobler & Wilson, 1990). In this system of sex
determination, new queens are also sisters to the workers of the colony. Thus,
workers and new queens are up to 75% related to each other. If workers were able to
produce their own offspring, the mothers and daughter would only be up to 50%
related. Therefore, sterile workers are thought to help their mothers to produce sister
queens (i.e., sisters of the workers) because they are contributing 75% of their genetic
information to the next generation instead of up to 50% of their genetic information if
they produced their own offspring. Workers thus have a direct fitness value of zero,
but a rather high indirect fitness costs by helping their mother produce more sisters.
Reproductive skew, kin selection, and haplodiploidy contribute to social insect
societies being closed societies from which unrelated intruders are prevented from
entering. It is important for sterile workers to keep intruders out of their society in
order to protect their inclusive fitness levels. The exploitation of resources by
outsiders directly alters the survival of the queen. By excluding intruders, workers
protect the reproductive output in the form of new queens and their sister queens that
will contribute genetic information to future generations.
Cuticular Hydrocarbon Profiles
The hydrocarbons coating the cuticles of social insects are referred to as cuticular
hydrocarbon profiles. Social insect cuticular hydrocarbons contain both saturated and
unsaturated n-alkanes and n-alkenes and methyl-branched alkanes, and insect
hydrocarbons range in chain length from 11 -43 carbons (Howard and Blomquist
2005, Hackman, 1984). The primary function of the hydrocarbons is to protect insects
from water loss (Lockey, 1988, Hadley, 1984). Hydrocarbons are produced internally
in abdominal oenocytes (Lockey, 1988), and those internally produced hydrocarbons
are secreted through exocrine glands (Gobin et al., 2003). The hydrocarbon
compositions are species-specific, sex-specific, and colony- and caste- specific
(Greene & Gordon, 2003, Sledge et al., 2001, Wagner et al. 2001 Singer, 1998,
Bonavita-cougourdan et al., 1991), and recognition systems of ants are highly
dependent on those molecules.
The cuticular hydrocarbon profile of ants consists of three principal classes of
hydrocarbons that are n-alkanes, branched alkanes, and n-alkenes. It has been
speculated that the strong involvement of branched alkanes and alkenes in recognition
behavior (Lucas et al., 2005) with primary use of n-alkenes (Holldobler, 1995)
because of the chemo-physical properties of those classes of hydrocarbon (Gibbs,
2002). The unsaturated hydrocarbons (alkenes) can have more variation with the
positioning of the double bond and same for branched alkanes with the different
positioning of the branches while straight chained saturated hydrocarbons (alkane)
can only vary so much.
Nestmate recognition refers to the ability of social insects to recognize nestmates
from non-nestmate conspecifics. The ability to discriminate nestmates from non-
nestmate is crucial for maintaining the integrity of colony, including its ability to
survive and reproduce. This excludes unrelated individuals from the colony, thus
protecting the colony and its resources from exploitation.
The action component of nestmate recognition in social insects typically involves
agonistic behavior such as stinging, chasing, and biting. Bees respond to non-
nestmate intruders with chasing, biting, and grappling (Buchwald & Breed, 2005,
Breed et al. 2004). Wasps respond to intruders with grappling, crashing, nibbling,
chasing, pecking, and biting (Venkataraman, et al., 1990). Agonistic behavior of
termites includes avoidance, defecation, and biting (Adams, 1991). For ants,
aggressive behavior towards intruders includes flaring of mandibles, stinging, or
biting (Holldobler & Wilson, 1990). This can be seen in ant wars. For example,
when two neighboring colonies of Tetramorium caespitum encounter each other, a
war between two colonies can result in which thousands of workers fight along the
boundary between the two colonies (Holldobler & Wilson, 1990).
The expression component of social insect nestmate recognition cues is mainly
chemical. The nestmate recognition cues of both honeybees (Apis mellifera) and the
stingless bee (Trigona fulviventris) are dominated by fatty acids (Buchwald & Breed,
2005, Breed et al. 2004). However, T. fulviventris also utilize alkane hydrocarbons
and floral oils for recognition of nestmates (Buchwald & Breed, 2005). The cuticular
lipids of another species of the stingless bee (Scaptrigona bipunctata) are composed
of a mixture of hydrocarbons (saturated and unsaturated alkanes and alkenes) within
the range of C23-C29 (Jungnickel et al., 2004). Jungnickel et al. (2004) also concluded
that 7 compounds of hydrocarbons were involved in the nestmate recognition of S.
bipunctata by comparing the relative abundance of cuticular hydrocarbons between
two different colonies that were 10 m apart (Jungnickel et al. 2004). However, their
study did not include the behavioral assay using the extraction of cuticular
hydrocarbons. Several species of Apidae, in particular the honeybee (A. mellifera),
utilize alarm pheromones to label intruders (Breed et al. 2004). The alarm
pheromones mark the intruder and also recruit more workers from the colony to
participate in the aggressive behavior towards marked intruders in order to defend
their colony (Breed et al. 2004). Therefore, those bees can not only recognize
intruders, but also mark them as their enemies.
Termites also recognize nestmates via cuticular hydrocarbons. Clement and Bagneres
(1998) analyzed and compared cuticular hydrocarbons among colonies of
Reticulitemes species and showed the colony specificity of the cuticular hydrocarbon
profile. Haverty et al. (1999) and Haverty & Throne (1989) furthered this analysis
and correlated differences in colonial phenotypes of hydrocarbon composition with
agonism towards non-nestmates, such as immediate aggression and mortility in
species of Reticulitemes and Zootermopsis. Kaib et al. (2004) also reported that the
level of aggression among colonies increases as the differences in composition of
cuticular hydrocarbons increase in the termite Macrotermes subhyalines. They also
speculated that unsaturated hydrocarbons are the key compounds in the nestmate
recognition from multiple regression analysis. However, Kirchner and Minkley
(2003) demonstrated that nestmate recognition of the harvester termite Hodotermes
mossambicus depends on the colony-specific composition of intestinal bacterial
symbionts. They demonstrated this simply by treating H. mossambicus with
antibiotics that caused suppression of intestinal bacterial symbionts, and paired non-
nestmates treated with antibiotics did not exhibit agonistic behavior towards each
Several species of social wasps are thought to utilize cuticular hydrocarbons because
of their colony-specific profiles (Singer et al., 1992, Espelie et al., 1994). Ruther et al.
(2002) demonstrated strong involvement of cuticular hydrocarbon in nestmate
recognition of the Europian hornet, Vespa crabro by testing recognition responses of
V. crabro towards dead nestmates treated with synthetic hydrocarbons, dead non-
nestmates, and dead nestmates. Vespa crabro responded with signitficantly higher
levels of aggreesion towards dead nestmates treated with synthetic hydrocarbons and
dead non-nestmate than towards untreated dead nestmates.
Ants rely on nestmate recognition cues present in cuticular hydrocarbons (Lucas et
al., 2005, Wagner et al., 2000, Lahav et al., 1999). The hydrocarbon molecules can be
found in the outermost layer of the insect body, the epicuticle that is composed of a
mixture of lipids (Lahv et al., 1999, Singer, 1998). Lahav et al. (1999) demonstrated
that the nestmate recognition cue in Cataglyphys niger is present in the cuticular
hydrocarbons by using chemical compound collected from the postpharyngeal gland
(PPG) with the assumption that the PPG is the hydrocarbon-producing gland.
However a recent study by Soroker et al. (1994) demonstrated that C. niger retains
their cuticular hydrocarbon profiles after PPGs were removed, and the hydrocarbons
seem to be produced in abdominal oenocyte (Lockey 1988). The internally produced
hydrocarbons are secreted through exocrine glands (Gobin et al., 2003), of which
about 100 types of exocrine glands are known in social insects (Billen, 2002). PPG is
a one of the secretly gland and contains hydrocarbons that are similar to cuticular
hydrocarbons in their contens (Soroker et al., 1995). Wagner et al (2000)
demonstrated that hydrocarbons are the nestmate recognition cues in Pogonomyrmex
barbatus by isolating the hydrocarbon component of the cuticular lipid by passing it
through a silica gel column. In their experiment, hydrocarbons themselves were
sufficient enough to elicit nestmate recognition.
However, it is still not clear if any of single classes of hydrocarbons are more
important than others or whether the hydrocarbon profile work as a whole and
structural complexity of profile is important. Lucas et al. (2005) tested the responses
of Pachycondyla subversa towards papers that were coated with either cuticular
hydrocarbons or other components of cuticular lipids without hydrocarbons of non-
nestmates. In their experiment, the movement of individual ants were restricted by
using immobilizing device. The device, according to them, helped to eliminate
unpredictable behavior and made the experiment more consistent, but this method
also eliminated the ability for individuals to conduct in biting behavior. Therefore,
the behavioral responses were measured only with antennal retraction and mandibular
flaring. In the experiment, P. subversa responded with aggression towards
hydrocarbon fractions of cuticular lipids from non-nestmate while they did not
respond aggressively towards non-hydrocarbon fraction of cuticular lipids of non-
nestmates. However the differences in their method from other recognition studies,
they demonstrated that the cue used in the nestmate recognition was hydrocarbons
and not other lipids.
Lucas et al. (2005) also tested behavioral response of P. subversa to each class of
hydrocarbons from two closely related species P. villosa and P. inversa (Lucas et al.,
2002) using the same immobilizing device described above. In their experiment, only
the branched alkanes elicited the recognition behavior of P. subversa. However, the
recent study of Greene and Gordon (2007) suggested the involvement of alkanes in
the recognition mechanisms of ants and the importance of the structural complexity of
the hydrocarbon profiles. They conducted an experiment comparing the behavioral
responses of the Argentine ant Linepithema humile towards four types of stimuli:
synthetic M-alkane standard alone, extracted cuticular hydrocarbons of nestmates, 12-
alkane standard + extracted cuticular hydrocarbons of nestmates, and blank control.
The 22-alkane standard alone, extracted cuticular hydrocarbons of nestmates, or blank
control did not elicit recognition response while standard + extracted cuticular
hydrocarbons of nestmates elicited an aggressive response. They also conducted
similar experiment with Aphaenogaster cockerelli using cuticular hydrocarbons from
Pogonmyrmex barbatus. While any hydrocarbons introduced in single structural
classes did not elicit aggressive responses, mixtures of any of two classes or all of
three elicited aggressive responses of A. cockerelli. Therefore, it is still unclear that if
one class of hydrocarbon is more important than others in the recognition behavior of
Despite the well-studied mechanisms of action component and expression component
of nestmate recognition systems in social insects, the perception component of
recognition system was not understood clearly until Ozaki et al. (2005) suggested the
sensory adaptation mechanism. They demonstrated that the sensillums of the
carpenter ant Camponotus japonicus respond only to non-nestmate cuticular
hydrocarbon blends and not to the nestmate cuticular hydrocarbon blend. When the
electrophysiological response of the receptor neuron in chemosemsilla of C.
japonicus to the nestmate cuticular hydrocarbon and non-nestmate cuticular
hydrocarbon were compared, the sensillum exhibited almost all or none sensitivity to
the introduced stimuli in which neuron fired action potentials only in response to the
non-nestmate hydrocarbon stimuli (Ozaki et al, 2005).
Species recognition or interspecific recognition plays a significant role in increasing
fitness. For social insects, species recognition and nestmate recognition are both
necessary for maintaining the cohesion and survival of a colony (Holldobler and
Wilson, 1990). Failure to recognize heterospecific species can be critical to the fitness
of the colony. For example, species of Formica are often targeted for slave workers
by the slave-maker ant, Polyergus berviceps (Johanson et al., 2001). Once queens of
P. berviceps kill the queens of targeted colonies, they are able to mimic colony- and
specie- specific hydrocarbon profiles of targeted colonies to make workers of the
targeted colonies to become their slave workers (Johanson et al., 2001).
P. villosa and P. inversa respond differently to the P. subversa whereas P. subversa
does not respond differently to P. villosa and P. inversa (Lucas, 2005). All three
species show higher levels of aggression toward heterospecifics than toward their
conspecifics (Lucas 2005). Marier et al. (2004) also showed differences in the
agonistic behavior of Crematogaster scutellaris towards non-nestmate conspecifics
and towards heterospecific species. Crematogaster scutellaris respond with classical
aggression such as gripping to conspecific non-nestmates, while they often engage in
the secretion of toxic venom to hetrospecific intruders.
Because cuticular hydrocarbons are also species-specific (Howard, 1993),
hydrocarbon profiles are thought to contain the species recognition cues for many
social insects. However, there are not enough studies done to provide a generalization
between hydrocarbons and species recognition (Howard and Blomquist, 2005).
The division of labor is an important feature of social insect colonies. Along with
recognition of nestmates and of other species, many social insects are capable of
recognizing the tasks individuals perform (Greene & Gordon, 2003, Sledge et al,
2001, Singer, 1998. Bonavita-cougourdan et al., 1991, Wagner et al., 2001).
The tasks of workers can be categorized into three larger categories: foraging for food
(patrolling and foraging), nest maintenance (inside nest maintenance and midden
work), and taking care of eggs and broods (Wilson, 1975, Wagner et al., 2001,
Greene & Gordon 2003, Robinson, 2001, Seeley, 2001). Through information
exchange during interactions among workers via antennal contacts, the activity levels
of each task are monitored and the numbers of the individuals participating in each
task are adjusted if necessary (Sendova-Franks et al., 2002, Wagner et al., 2001,
Robinson, 2001, Seeley, 2001, Gordon 1989, 1987, Wilson, 1975).
Harvester ants use cues in cuticular hydrocarbons to recognize the task workers are
performing (Greene & Gordon, 2007, 2003, Wagner et al., 2001, Wagner et al.,
1998). Wagner et al. (1998) analyzed and compared the cuticular hydrocarbon
composition of Pogonomyrmex barbatus among three tasks: patrolling, foraging, and
nest maintenance using gas chromatography-mass spectrometry. They found
differences in the relative proportions of classes of hydrocarbon compounds with
foragers and patrollers having higher proportion of straight chain alkanes relative to
monomethylalkanes, dimethylalkanes, and alkenes compare to nest maintenance.
Relating to the finding from this study, Wagner et al. (2001) investigated how those
differences in relative proportions of classes of hydrocarbon occurred. They designed
three experiments all using nest maintenance workers of P. barbatus that spend most
of their time working inside their nest. In the first experiment, they subjected workers
to one of two treatments: 1) 50 minute periods of exposure to out-door conditions a
day for 20 days, or 2) no exposure to out-door conditions for 20 days. The exposure
to outside conditions led to the significant increase in the relative abundance of n-
alkane. Second, they subjected workers to one of two treatment: 1) high ultraviolet
exposure for 50 minutes a day for 11 days, or 2) low ultraviolet exposure for 50
minutes a day for 11 days. The exposure to high or low ultraviolet did not affect the
cuticular hydrocarbons quantitatively or qualitatively. In third experiment, they
subjected workers to one of four treatment: 1) cool and moist conditions, 2) cool and
dry conditions, 3) warm and moist conditions, or 4) warm and dry conditions for 11
days. Temperature was defined as 35 to 40C for warm condition and 25 to 30C for
cool condition. Humidity was defined as 95 to 100% for high condition and 10 to
30% for low condition. Warm, dry conditions led to the significant increases in the
relative abundance of -alkane. Wagner et al. (2001) argued that warm, dry
conditions resembled to the environment outside nests and concluded that the
difference in the relative abundance of hydrocarbon profiles among the task is due to
the amount of exposure to the outside of their colony that depends on the tasks.
Greene and Gordon (2003, 2007) further investigated the significance of the
difference in hydrocarbon composition among the tasks on the behavior of colony as
a whole. They mimicked the flow of returning patrollers using glass beads coated
with cuticular hydrocarbon of patrollers, whole cuticular lipids of patrollers,
hydrocarbons of nest maintenance, or blank controls. The patroller hydrocarbons
themselves were sufficient to recruit foragers come out of nest entrance while blank
and nest maintenance hydrocarbons were not (Greene and Gordon, 2003), and they
also found the optimal interaction rate between foragers and patrollers for the
maximum execution of recruitment of foragers as 1 encountering per 10 seconds
(Greene and Gordon, 2007).
Maintenance of Nestmate Recognition Cues
The hydrocarbon profiles that can provide cues for nestmate recognition are colony
specific and often referred as Gestalt colony odor (Soroker et al., 1994). However,
the profiles of hydrocarbons are altered constantly in response to the environment,
due to the activities individual insects participate in or according to the time (Vander
Meer & Morel, 1998). For example, the cuticular hydrocarbon of Argentine ant L.
humile can be contaminated with long chain hydrocarbons by contacting a particular
prey item, the brown-banded cockroach, Supella longipalpa, and can elicit aggression
from their nestmate (Liang et al, 2001). The type of diet they take also affects
colony- specific profiles of hydrocarbons in L. humile (Liang & Silverman, 2000) and
in the Formosan subterranean termite Coptotermes formosanus (Darlington, 1994).
The seasonal variation in the colony specific profiles of cuticular hydrocarbons
among a formicine ant Formica japonica is also reported (Liu et al., 2001). As
mentioned earlier, the task-specific of the cuticular hydrocarbons within a colony of
harvester ants Pogonomyrmex barbatus is due to the differences in task related
environment (Wagner et al., 2001). Therefore the templates of nestmate recognition
system also have to be updated as well (Vander Meer & Morel, 1998).
Social isolation of mature workers also alters nestmate recognition in the ant
Camponotus fellah (Boulay & Lenoir, 2001). Twenty days of isolation of workers
from their nestmates resulted isolated workers to reduce aggression towards non-
nestmate conspecific as low as towards their nestmate while the level of aggression
towards heterospecific individuals remained same. Another study by Boulay et al.
(2000) showed that 20-40 days of isolation resulted in progressive divergence of
cuticular hydrocarbon profiles in their relative abundance. They also showed that 20-
40 days of isolation led behavioral modification in which the resident nestmate
responded to isolated nestmate with aggression (Boulay, 2000). However, their
chemical analysis also showed quantitative difference in the cuticular hydrocarbons in
the nonisolated workers after 40 days. This suggested that the templates for nestmate
recognition also have to be updated to recognize their nestmates (Boulay, 2000).
Also, the harvester ant Pogonomyrex barbatus shows the alteration of their cuticular
hydrocarbon compositions in both qualitatively and quantitatively after they are
housed in the laboratory (Tissot et al., 2001). Qualitatively, the laboratory housed
workers had higher amount of the methyl-branched alkanes and straight chained
alkanes. Quantitatively, laboratory colony had more than two fold greater amounts of
cuticular hydrocarbons than the field colony did on average.
The colonial signatures or the Gestalt systems of cuticular hydrocarbons are
maintained through trophallaxis and allo-grooming (Lucas et al., 2004, Boulay et al.,
2000, Soroker et al., 1995). The hydrocarbons are produced internally among each
individual (Gobin et al., 2003), and the quality and quantity highly reflect
environment and activities of individuals (Liang et al, 2001, Liu et al, 2001, Wagner
et al., 2001, Boulay, 2000, Liang & Silverman, 2000). Therefore, the colonial
specificity of hydrocarbon profiles have to be integrated through physical contacts
such as trophallaxis and allo-grooming. It is also possible to argue that individuals
are updating their templates for nestmate recognition by desensitizing their sensillum
while participating in those physical contact activities to maintain colonial odours.
Katazav-Gozansky et al. (2004) investigated what type of interaction was necessary
for the integration of nestmate recognition of Camponotus fellah by isolating workers
with different degrees of isolation: 1) limited physical contact permited, 2) exposed
only to nest volatiles, but no physical contact, and 3) complete isolation from the nest.
They demonstrated that completely isolated workers were attacked by their original
nestmates after 2 weeks of isolation while it took over 2 months for workers exposed
to nest volatiles to be attached by their former nestmates. Workers that were
permitted for limited physical contact were still treated as nestmate even after 2
months of isolation. They also showed divergence in both PPG and cuticular
hydrocarbon profiles between non-isolated workers and completely isolated workers
with workers with workers with limited physical contact allowed and workes only
expressed to the nest volatiles expressing somewhat intermediate hydrocarbon
profiles with some overlap. This suggests the alteration of colonial hydrocarbon
signiture and updating their templates as perception component of nestmate
recognition are rather gradual and they are consistently altered and updated.
Physiological function of cuticular hydrocarbons
The primary function of cuticular hydrocarbons is thought to be the prevention of
water loss (Hadley, 1984). Hadley reported that amounts of epicuticular wax in some
species change in response to seasonal changes in the environment, or insect species
found in hot, dry environment tended to have greater amount of epicuticular wax than
closely related species found in moderate environments (Hadley, 1984). If this is the
case, recognition using hydrocarbons as cues can be thought of as secondary in the
The Pavement Ant T. caespitum belongs to the family Formicidae, subfamily
Myrmicinae. Pavement Ants are small in body size, ranging from 1/10 to 1/16 inch in
length, with dark or dark brown body, pale-colored legs and antennae, and a series of
grooves on the head and thorax. The posterior/dorsal thorax has two spines
projecting upward, and the petiol have two nodes. Their antennae consist of 12
segments, including a 3-segmented club (Figure 1.1). T. caespitum is native to
Europe and was introduced to United States during the 1700s or 1800s (McGlynn,
1999). They form polygyne colonies in which colonies might have more them one
queen and each colony can have up to tens of thousands of workers (Steiner, et al.,
2003, Brian, et al., 1967). They feed on variety of food including grease, meat, insects,
seeds, fruit, and sweet. They tend to make their nest around foundations, under rocks,
and in cracks of sidewalks and driveways (Holldobler & Willson, 1990, Cranshaw,
2006). As well as nestmate recognition behavior and the war between neighboring
colonies, they are also suspected to exhibit certain degrees of unicoloniality
(Holldobler & Willson, 1990, Steiner et al., 2003). This species was selected for this
study because of high abundance in the city of Denver.
Grooves 0ne Pa'r sP'nes
Picture taken from http://www.ipm.ucdavis.edu/TOOLS/ANTKEY
Figure 1.1. The pavement ant Tetramorium caespitum.
Methods and Materials
2.1 Field Sites
Ants were collected from a total of 22 sites. Twelve sites were on the Downtown
Denver Campus of UCDHSC and 10 sites were in the residential areas of downtown
and uptown Denver. Sixteen of those sites were also used as focal colonies in the
behavioral assay of nestmate and species recognition (see 2.8). The pavement ant is
highly abundant and easily located in the area, making it a convenient study animal.
For the heterospecific recognition, western harvester ants Pogonomyrmex
occidentalis were collected at weedy field in northwest Denver near downtown and
Carpenter ants (genus Camponotus) were collected at residential area in uptown
Denver. Both ants interact with T. caespitum (personal observation).
2.2 Collecting Pavement Ants T. caespitum
Ants were collected into collecting tubes using an aspirator. Ants were collected
from foraging trails leading from nests or by using honey baits to lure forgagers out
of the nest. Ants used for hydrocarbon extraction were killed by freezing at -20C.
Samples were stored at -20C until the extraction of hydrocarbons. Live ants that
were used in the isolation study (see 2.11) were transferred into 30x18 cm2 plastic
containers. All ants used in this study were collected between April and July of 2005,
and between April and August of 2006.
2.3 Care of Captive Ants
Live ants collected from colonies in the field were housed in the laboratory in the
30x18 cm2 plastic containers at room temperature (~20C) and relative humidity of
approximately 25%. The inside walls of containers were coated with teflon paint
(INSECT-a-SLIP, BioQuip Products, Inc > to prevent ants from escaping. Water
was provided ad libidum using glass tubes filled with water and sealed with cotton
balls. Ants were able to drink from the water-saturated cotton. Ants were fed with
the Bhatker diet (Holldobler & Wilson, 1994), consisting of albumin protein from
eggs, sugar, water and vitamins mixed in an agar solution. Pieces of the set agar were
placed in the ant containers and replaced every 2 to 3 days. Ants were exposed to
approximately 12 hour light/12 hour dark photocycles. About 1000-3000 ants were
added to each container. The accumulation of dead bodies was checked daily in order
to keep the colony clean and assess the rate at which ants were dying.
2.4 Extraction and Purification of the Cuticular Hydrocarbons
In order to extract surface lipids, thawed ant samples were soaked with 1.0 ml of
pentane for 10 min with mild agitation every minute. This extraction method allows
for extraction of surface lipids without extraction of lipids from inside the ants
(Nelson and Blomquist, 1995). The lipid extracts were then passed through a
chromatography column to isolate the hydrocarbon components from other cuticular
lipids. The packed phase of the column was silica gel (70-230 mesh, average pore
diameter: 60A. SIGMA). Silica gel was used as column to remove polar component
of cuticular lipids, which bind to packing material when a non-polar eluent such as
pentane is passed through the column (Nelson and Blomquist, 1995, Wagner et al.,
2000). To elute hydrocarbons, 2 to 3 ml of pentane was passed through the column
into a screw top vial.
2.5 Preparation of Experimental Stimuli
Stimuli were prepared by placing 5mm diameter glass beads in the vials containing
isolated hydrocarbons in 100% pentane. The glass beads were left in the solvent
overnight at room temperature to allow pentane to evaporate. In this manner, the
cuticular hydrocarbons from the ants transferred to the surface of glass beads. The
glass beads were stored at -20C until the day of experiment, about 0 to 2 days after
extraction. The glass beads were used to mimic the size of the ants T. caespitum
interacts with in the field while also removing other sensory information, besides
chemical information from hydrocarbons.
For nestmate recognition bioassay (see 2.10), 50 pavement workers worth of
hydrocarbon were added to each glass bead. For the heterospecific species
recognition bioassay (2.10), five workers of P. occidentalis worth of hydrocarbon
were applied to each glass bead and five Camponotus workers worth of hydrocarbon
were added to each glass bead.
2.6 Behavioral Bioassay for Nestmate and Species Recognition
The aggressive behavior exhibited by focal ants was measured towards glass beads
coated with hydrocarbons. The glass beads were introduced near the entrance of
focal colonies. Behavioral responses of focal ants were observed for 10 minutes. The
data were collected every 30-second interval during the 10-minute observation
period. All behavioral tests were conducted as blind tests. As well as using ethogram
(see 2.7) scores I also counted the number of ants showing either biting behavior or
mandibular flaring as aggressive ants and divided the number of aggressive ants by
total number of ants contacted the hydrocarbon stimuli to compare proportions. All
the behavioral tests at focal colonies were performed between 5:00PM and 9:00PM
when ants were starting to come out of nest for foraging. The behavioral assays in
the field were conducted between April and July of 2005, and between April and
September of 2006. Four focal colonies of T.caespitum were tested for their
responses in July of 2005, and fourteen focal colonies of T.caespitum were tested in
August and September of 2006.
2.7 Ethogram Score
To measure aggressive level of workers, I gave the score based on scale with 3 levels
(Table 2.1). Antenation received score of 0, openings of the mandible received 1, and
biting received 2. Antennation occurs when an ant touches the stimuli with its
antennae. Aggressive behavior was observed in the 10 minutes interval. Scores were
calculated using aggression index formula (modified from Hefetz et al., 1996) to
allow the comparison in the fraction.
A.I.= __________~ (lOmin interval)
A.I. means the aggression index. A.I./' is a factor applied to each act regarding its
degree of aggressiveness. A.I./ = 0 for antennation, A.I./ = 1 for mandibler flaring,
and A.I./' = 2 for biting. The number of total ants contacted with the stimuli presented
divided the sum of the A.I./' for the final ethogram score for the analysis.
Table 2.1. The description of T. caespitum behavior to define ethogram scores.
Ethogram Score Short description of behavior
0 Ants come in contact with the stimuli with antennae.
1 Ants exhibit mandibular flaring towards the stimuli.
2 Ants show biting behavior towards the stimuli. i@
Pictures taken from http://www.mvrmecos.net/anttaxa.html
2.8 Gas Chromatography/Mass Spectrometry
The chemical structures of hydrocarbons are currently being analyzed by using gas
chromatography/mass spectrometry (Wagner, Tissot, Cuevas, & Gordon, 2000) with
the collaboration of Clare Chen, Dr. Lisa Lanning, and Dr. Marc Donsky from the
UCDHSC Chemistry Department. Although these data will appear in the peer-
reviewed manuscript describing these data, they are nor reported here because the
data are not complete.
2.9 Statistical Analysis
All ANOVA calculations were performed in Sigma Stat (Version 2.0). Linear
regression models, Pearsons correlation and paired t-tests were performed using
SPSS version 12.0 for Windows.
2.10 Experiment for Aim 1: To determine if T. caespitum recognizes and
discriminates their nestmate ants from non-nestmate conspecific ants and
heterospecific ants using cues present in cuticular hydrocarbons.
Experiment 1: Do T. caespitum use cuticular hydrocarbon profiles as nestmate
recognition cues? Comparison of levels of aggression using the proportion of
aggressive ants to stimuli
To investigate relation between hydrocarbons and nestmate recognition of T.
caespitum, the behavioral assays were conducted at 4 different focal colonies on the
campus of UCDHSC in July of 2005. Cuticular hydrocarbon stimuli were prepared
from ants collected from 8 different colonies of T. caespitum including 4 the focal
colonies. As a control, blank beads coated with solvent used in the hydrocarbon
extraction process (pentane) were tested with the experimental stimuli. Behavioral
responses of focal ants were observed for 10 minutes. The data were collected with
intervals of every 30 seconds (bioassay described in 2.6). Each focal colony was
presented with the blank control, nestmate hydrocarbon stimuli, non-nestmate
hydrocarbon stimuli from 2 different colonies, and stimuli from heterospecific ants
(see 2.1 & 2.5). Stimuli were presented in random order. All behavioral tests were
conducted as blind tests.
The behavioral responses were recoded in ethogram scores (see 2.7). The presence or
absence of aggressive behavior (either biting behavior or mandibular flaring) were
counted and divided by total number of focal ants came in contact with stimuli
instead using ethogram score in order to compare the proportion of ants were
aggressive to the introduced stimuli. Ethogram scores were also compared in order to
measure the differences in the level of aggression towards nestmate and non-nestmate
to analyze the data in continuous scaling as well as binomial scaling.
To analyze data statistically, repeated-measures ANOVA and LSD post hoc test were
used for both comparisons of ethogram scores and proportions of aggressive ants for
all the experiment run for the Aim 1.
Experiment 2: Does T.caespitum recognize heterospecific species as well as non-
nestmate of conspecifics using cues present in cuticular hydrocarbon profiles?
In this experiment, the following stimuli were tested at 14 focal colonies of T.
caespitum: 1) nestmate hydrocarbon profiles, 2) two randomly assigned non-nestmate
hydrocarbon profiles, 3) two different heterospecific hydrocarbon profiles (see 2.1
and 2.5), and 4) a blank control.
The glass beads were introduced near the entrance of focal colonies, and behavioral
responses of focal ants were observed for 10 minutes. The data were collected every
30 seconds. Stimuli were presented in random order. Every 30 seconds, the behavior
displayed by the ants in contact with stimuli was recorded according to the ethogram
scale (Table 2.1). The aggressive behavior in this experiment was defined and scored
as 2 = biting, 1 = mandibular flaring, and 0 = antenation (see Table 2.1). This
experiment was conducted in August and September of 2006.
2.11 Experiment for Aim 2: To measure the effect of social isolation and
abiotic factors on profile of individual cuticular hydrocarbon
composition and nestmate recognition.
The colonial signature of cuticular hydrocarbon is thought to be maintained through
interaction among individuals using antennation and trophallaxis.
To investigate the impact of social isolation and environmental conditions on the
expression and perception of nestmate recognition cues, I separated workers from
original colony into four different experimental conditions:
1. Treatment group 1 was conditioned under high temperature (~25C) and high
relative humidity (-80%).
2. Treatment group 2 was housed under high temperature (~25C) and low
relative humidity (-40%).
3. Treatment group 3 was housed under low temperature (~20C) and high
relative humidity (-80%).
4. Treatment group 4 was housed under low temperature (~20C) and low
relative humidity (-40%).
To maintain higher temperatures, ants in treatment group 1 and treatment group 2
were housed inside an incubator at 25C. Treatment group 3 and treatment group 4
were housed inside a cabinet where temperature was at room temperature,
approximately 20C. To maintain the relatively high humidity in treatment 1 and 3,1
sprayed the water twice a day and 50ml beakers containing the wet cotton balls were
also placed in the containers. For treatment group 2 and 4, the 50ml beaker
containing Dri-rite (W.A. Hammound DRIERITE Company, LTD) was placed in
their containers. The DRIERITE was replaced every 48 hours. HOBO data logger
devices were placed in each container to monitor the temperature and relative
humidity for a period of 48 hours. Ants were fed with the Bhatker diet every 4 days
during the experiment in progress. Water was also provided ad libidum using glass
tubes filled with water and sealed with cotton balls.
Each of the 4 different colonies was split into four treatment groups to run 4 sets of
replicates in the experiment. However, only 2 sets of replicate successfully ended the
experimental period. For first set of this experiment, each treatment group received
400 workers. For second set of this experiment, each treatment group received 300
workers. The experiment was run for a total of 28 days (4 weeks). Every 7 days, 10
to 15 workers were collected from a colony fragment from each treatment group and
frozen to archive the cuticular hydrocarbon profile at the time in the experimental
On the 28th day of the experiment, I conducted behavioral bioassays to measure
nestmate recognition responses from the ants collected at sampling points during the
experiment. Frozen bodies from each sampling point, week 1, week 2, and week 3
were introduced to live ants of week 4 in a random sequential order. Dead ants were
introduced one at a time to investigate if the original population recognizes their
former nestmates as foreign. The behavioral responses of focal ants were observed
for 10 minutes. The data (i.e., the ethogram scores (see 2.7)) were collected at 30
seconds intervals. I analyzed data One way-ANOVA using week (time) and
environmental conditions as factors in the analysis.
I predicted that if social isolation affected the expression of nestmate recognition
cues, then nestmate hydrocarbon profiles would change with time in all treatment
groups. Thus, ants in all colony fragments in week four would be expected to be
aggressive towards odors from nestmates collected previously during week 1. In
other ants, high temperature and low humidity cause changes to hydrocarbon profiles
(Wagner et al., 2001), thus I hypothesized that the high temperature/low humidity
treatment condition would yield aggression in response to former nestmates and other
conditions would not make this change.
To investigate consistency/plasticity of hydrocarbon profile within same colony, I
analyzed the cuticular hydrocarbon from samples collected every 7 days for 28 days
and component of each hydrocarbon will be compared using GC/MS to investigate
over time change in the component of cuticular hydrocarbons. Because of technical
problems these data will be reported in the peer-reviewed manuscript based on this
thesis, but are not reported here.
2.12 Experiment for Aim 3: To demonstrate anti-desiccation properties
of the cuticular hydrocarbons.
To demonstrate the water loss protection property of cuticular hydrocarbons, I used
the evaporative water loss measurement methods outlined by Roberts and Lillywhite
(1983). To investigate the primary function of the cuticular hydrocarbon, in other
words the ability to prevent water loss across the cuticle, an experiment was
performed that simulated to mimic the physiological function of epicuticle. However,
because I was not able to make measurements using pieces of cuticle using the
methods I had available, I instead used snake skins to mimic the insect cuticle.
Because snakes shed their skin in single pieces, I was able to create biologically
relevant barriers to water loss using them. Snake skins were collected from the cages
of Bull Snakes (Pituophis catenifer) and stored in a freezer at -20C until used in the
experiment. I removed all lipids from snake skins by soaking them in hexane
overnight (to remove non-polar lipids) and then in a 50:50 mixture of methanol and
chloroform (to remove polar lipid) overnight. The experimental group received
isolated hydrocarbon from 200 T. caespitum workers while the control was untreated.
The snake skins were cut into 8mm in their diameter circles and placed in a vial cap
with a 5mm hole between two plastic septa with 8mm holes cut in them (Figure 2.1).
Vials containing deionized water were capped so that the snake skin pieces formed a
barrier between the water in the tube and the air outside (Figure 2.1). They were
hung upside down so that water was in direct contact with skins, and kept in the
incubator at 25 C for 72 hours. Each vial was weighed every 24 hours for 72 hours.
Independent t-test was used to compare the difference in mass of water lost between
two treatment groups at 72 hours.
Figure 2.1. Screw-top vial setup for water loss experiment.
3.1 Results for Aim 1: To determine if the ant recognizes non-nestmate
conspecific ants and heterospecific ants using cues present in
3.1.1 Analysis 1: Does T. caespitum use cuticular hydrocarbon profiles as nestmate
recognition cues? Comparison of levels of aggression using the proportion of
aggressive ants to stimuli.
These data show that T. caespitum is able to discriminate non-nestmate from nestmate
ants using a cue present in their cuticular hydrocarbons. For this analysis, the
proportion of aggressive ants was calculated by dividing the total number of
aggressive ants towards the stimuli by the total number of ants in contact with each
stimuli at time points during the 10 minutes observation period. Ants were considered
aggressive if they flared their mandibles or bit the stimuli, but were not assessed
using the ethogram score for this analysis.
In the bioassay, ants responded differently in the proportion of ants aggressive to the
blank control, nestmate hydrocarbons and non-nestmate hydrocarbons (Repeated
measure ANOVA, Fi,3 = 70.430, p < 0.005; n = 4; Figure 3.1.1; Table 3.1.1, 3.1.2).
Colonies showed a significantly higher proportion of aggressive ants to non-nestmate
hydrocarbons compared to the nestmate hydrocarbons and the blank control (LSD
post-hoc,/? <0.005). There was no statistical difference between the proportion of
aggressive ants towards nestmate hydrocarbon stimuli and the blank control (LSD
post-hoc, p > 0.05).
Collectively, 1 out of a total of 42 ants responded to the nestmate hydrocarbon stimuli
with aggression while 55 out of a total of 99 ants that came contact with the non-
nestmate hydrocarbon stimuli displayed aggression. Zero out of a total of 29 ants
responded to the blank control with aggression.
In this experiment, more than one non-nestmate sample was tested for each focal
colony (Table 3.1.3).
3.1.2 Analysis 2: Does T. caespitum use cuticular hydrocarbon profiles as nestmate
recognition cues? Comparison of levels of aggression using an ethogram
In analysis 1,1 compared the proportion of aggressive ants towards nestmate
hydrocarbons, non-nestmate hydrocarbons, and blank controls to show that they
respond differently to nestmate and non-nestmate hydrocarbon stimuli. In this
analysis, I compared the mean ethogram scores (Table 2.1) for the same 4 focal
colonies tested in the analysis 1 of experiment. A higher ethogram score indicated a
higher level of behavioral aggression. This experiment was performed in order to
measure the differences in the level of aggressions towards nestmate and non-
nestmate along a continuous scale versus a binomial scale as used in analysis 1. As in
analysis 1, there were statistically significant differences in the levels of aggression
among colonies towards non-nestmate hydrocarbon stimuli and nestmate hydrocarbon
stimuli. The repeated measure ANOVA test showed there were significant
differences in the mean ethogram scores towards three experimental groups (Fi,3 =
38.039, p < 0.01). LSD post hoc test showed there was a significant difference
between nestmate hydrocarbon stimuli and the non-nestmate hydrocarbon stimuli and
blank control (p < 0.005). The blank control did not differ from nestmate
hydrocarbon stimuli (LSD post-hoc,/? > 0.05, figure 3.1.2).
Evidence for unicoloniality in T. caespitum?
It was fortunate that multiple non-nestmate hydrocarbon samples were tested in the
experiment at each focal colony. During the data analysis for analysis 1 and 2,1
noticed that focal colonies did not respond to some non-nestmate samples with high
levels of aggression. This might provide evidence that T. caespitum exhibits
unicolonial colony structure in which different colonies can share workers or
collaborate without aggression. The Argentine ant Linepithema humile is well known
for its unicoloniality (Giraud et al., 2002), in which intraspecific aggression is
reduced and workers and queens can move freely between different colonies (Passra,
1994). Ants at colony 4 were tested for aggression in response to a blank control,
nestmate hydrocarbons, non-nestmate hydrocarbons from colony 1 and colony 5.
Ants at colony 5 were tested for aggression in response to a blank control, nestmate
hydrocarbons, non-nestmate hydrocarbons from colony 1 and colony 4. Ants at
colony 4 did not respond with aggression to hydrocarbon stimuli from colony 5 while
they responded with aggression to the hydrocarbon stimuli from colony 1.
Interestingly, ants at colony 5 did not respond with aggression to the hydrocarbon
stimuli from colony 4 either although they exhibited aggression towards hydrocarbon
stimuli from colony 1 (Table 3.1.2.) Ants at colony 3 were tested for aggression in
response to a blank control, nestmate hydrocarbons, non-nestmate hydrocarbons from
colony 6, colony 8, and colony 10. Ants at colony 6 were tested for aggression in
response to a blank control, nestmate hydrocarbons, non-nestmate hydrocarbons from
colony 3, colony 8, and colony 10. Although ants at colony 3 showed aggressive
responses to all the non-nestmate hydrocarbon stimulus introduced, mean ethogram
scores indicating the level of aggression towards hydrocarbon stimuli from colony 6
was noticeably lower than towards the hydrocarbon stimulus from other 2 colonies.
Response data from colony 6 showed similarity with response data from colony 3 in
which the mean ethogram score indicating the level of aggression towards
hydrocarbon stimuli from colony 3 was noticeably lower compare to the ethogram
scores towards hydrocarbon stimuli from colony 8 and colony 10 (table 3.1.2).
3.1.3 Analysis 3: Does T. caespitum recognize heterospecific species as well as
non-nestmate of conspecific using cues present in cuticular hydrocarbon
These data were collected from a second experiment conducted during the summer of
2006. The data showed that T. caespitum also recognizes cues present in cuticular
hydrocarbons during the recognition of heterospecific ants. Tetramorium caespitum
responded with high levels of aggression towards both non-nestmate hydrocarbon
samples and hydrocarbon samples from the heterospecifc ants, genus Camponotus
(unidentified species) and P. occidentalis in comparison to blank control and
nestmate. (Repeated measures ANOVA, Fi,n = 997.496. p < 0.001; n = 14; Figure
3.1.3, Table 3.1.4). The glass beads coated with non-nestmate hydrocarbons elicited
higher ethogram scores compared to the glass beads coated with nestmate
hydrocarbon (LSD post-hoc test,/? < 0.001; Figure 3.1.3, Table 3.1.4). The glass
beads coated with heterospecific species hydrocarbons also elicited a higher mean
ethogram score compared to the glass beads coated with nestmate hydrocarbon (LSD
post-hoc test,/? < 0.001; Figure 3.1.3, Table 3.1.4).
Because of the uneven aggressive responses to the different colonies in experiment 1,
two different colonies of conspecifics were tested for nestmate recognition test.
However, the LSD post-hoc analysis showed no statistical differences in the mean
level of aggression displayed towards the two non-nestmate samples (LSD,/? > 0.05).
The post-hoc analysis also showed that there were no differences in the mean
ethogram scores of the focal ants towards the hydrocarbons of P. occidentalis and
Camponotus (p > 0.05; Figure 3.1.3, Table 3.1.4). The response towards the non-
nestmate samples and the heterospecific samples did not differ significantly (LSD, p
> 0.05). Also, the pentane-coated glass beads that were used as control did not differ
statistically from the glass beads coated with nestmate HC extract (LSD post hoc test,
p > 0,05) (Figure 3.1.3, Table 3.1.4).
Figure 3.1.1. The proportion of individuals who responded with aggression between
nestmate and non-nestmate hydrocarbon. There is a statistical
difference between nestmate HC and non-nestmate HC. Bars represent
mean values +/- s.e.m. n= 141.
Table 3.1.1. The number of ants contacted HC, number of ants responded with
aggression, and their proportions. n= 141 from 8 different colonies.
Data have been combined for all colonies tested.
Nestmate HC Non-nestmate HC
Number of ants contacted 42 99
Number of ants responded with aggression 1 55
Proportion 0.0238 0.556
Mean ethogram score
Figure 3.1.2. The degree of aggression towards hydrocarbon stimuli of nestmate and
non-nestmate. There are no statistical differences between nestmate
HC and pentane control. Bars represent mean values +/- s.e.m. Lines
above bars denote a lack of statistical differences among treatments
according to LSD post-hoc test. The lines above the bar graphs
indicate there are no statistical differences between the groups. n=4.
Scores are the mean ethogram scores.
Table 3.1.2. Descriptive statistics for comparison of mean ethogram scores for
nestmate recognition. n=4.
Mean ethogram score Std error
Blank .000 .000
Nestmate stimuli .104 .079
Non-nestmate stimuli 1.252 .163
Table 3.1.3. Comparisons of mean ethogram scores to the HC profile from different
colonies. Because of the availability of HC extract, Colony-9 is
excluded from the statistical analysis for the ethogram comparison.
Focal ants colony Hydrocarbon sources
Colony-1 Colony-4 Colony-5
Colony-4 1.7037 0.4286 0
Colony-5 1.4 0 0.0833
>> e o Colony-6 Colony-3 Colony-8 Colony-10 Colony-9
O C/5 c S3 Colony- 3 0.75 0 1.3333 1.75
CJ O u* Colony- 6 0.3333 0.6667 2 2 0.6667
Mean ethogram scores
Nestmate HC Pentane Control Non-nestmate 1 Non-nestmate 2
Figure 3.1.3. The degree of aggression towards each experimental group. There are
no statistical differences between nestmate HC and pentane control, or
between non-nestmate 1, non-nestmate 2, Camponotus, and P.
occidentalis. Bars represent mean values +/- s.e.m. Lines above bars
denote a lack of statistical differences among treatments according to
LSD post-hoc test. The lines above the bar graphs indicate there are no
statistical differences between the groups. n=14.
Table 3.1.4. Descriptive statistics for nestmate and heterospicies recognition.
N Mean ethogram score Std. Error Std. deviation
Nestmate HC 14 .0899 .03679 .13766
Pentane control 14 .0903 .04225 .15807
Non-nestmate 1 HC 14 1.4993 .08944 .33465
Non-nestmate 2 HC 14 1.5398 .06205 .23218
Campanotus 14 1.6842 .06211 .23238
P. Occidentalis 14 1.5168 .13658 .51105
3.2 Results for Aim 2: To measure the effect of social isolation and
abiotic factoron profile of individual cuticular hydrocarbon
composition and nestmate recognition.
Experiment: Do social isolation or different environmental condition alter the
nestmate recognition behavior ofT. caespitum?
All environmental condition treatments caused changes in the expression of the
recognition cues, but there was no difference in ethogram scores among the
experimental treatments. In other words, all conditioning in high or low temperatures
and high or low humidity was able to change the recognition of nestmates. I
hypothesize that social isolation from other colony fragments also leads to changes in
recognition cues. ANOVA analysis was conducted using week (time) and
environmental conditions as factors in the analysis.
The duration of isolation period had an effect on recognition mechanisms. The level
of aggression, as measured by the ethogram score, towards isolated individuals
increased as duration of isolation period increased (AVOVA, F3,i6 = 4.627, p < 0.05).
The post-hoc test analysis showed statistical difference in the mean ethogram score at
fourth week (p < 0.01). In other words, four weeks of isolation from the main colony
was sufficient to lead to an aggressive response towards former nestmates.
The difference in their environment, temperature and humidity levels, did not affect
the mean ethogram score over time (ANOVA, F3,i6 = 0.115, p > 0.05). HOBO
devices were placed in each treatment group and temperatures and relative humidities
were recorded every 6 minutes for 48hr to approximate average climate setting.
HOBO data suggested that those 4 treatment groups had different climate setting
during the experiment was in progress. HOBO devises recorded average temperature
of 26.1794C and average relative humidity of 85.8264% for treatment group-1,
average temperature of 25.82650 and average relative humidity of 59.5871% for
treatment group-2, average temperature of 20.82390 and average relative humidity
of 61.9850% for treatment group-3, and average temperature of 20.8961 and average
relative humidity of 25.1307% (Figure 3.2.1). Despite differences in relative
humidity and temperature among the treatments, I found no effect of environmental
conditions on the recognition response.
Mean ethogram score
Sample Sample Sample Sample
from day 1 from day 7 from day 14 from day 21
Figure 3.2.1. Decrease in mean ethogram score of focal ants towards isolated
individuals. Each bar indicates mean ethogram scores. There is a
statistical difference between 1 week and 4 week.
Table 3.2.1. The change in the level of aggression toward isolated nestmate with
duration of time. The scores are mean ethogram scores. The level of
aggression increased as time of isolation prolonged.
Environmental Condition One week of isolation 2 weeks of isolation 3 weeks of isolation 4 weeks of isolation
High temp & humid 0.250 0.000 0.725 1.417
High temp & dry 0.000 0.867 0.800 1.417
Low temp & humid 0.200 0.950 0.750 1.000
Low temp & dry 0.000 0.433 0.875 1.300
Figure 3.2.2. The average temperatures in C and average relative humidity in %
among 4 treatment group. The data recorded using HOBO devises.
3.3 Results for Aim 3: To investigate the anti-desiccation properties of
the cuticular hydrocarbon.
Experiment: Do hydrocarbons work as protection against water loss?
These data show that cuticular hydrocarbons offer significant anti-desication
properties to the ants. Independent t- test (n = 12) showed there were statistically
significant differences between snake skins treated with cuticular hydrocarbons
extracted from T. caespitum and skins without cuticular hydrocarbon treatment in the
amount of water lost at all three time interval points. The snake skins treated with the
cuticular hydrocarbon extract lost significantly less water than the control group after
72hr (p = ,002)(Figure 3.3). After 24hrs, the experimental group lost 0.049333333g
of water in average that account for 0.63% loss in total mass while the control group
lost 0.091066667g of water in average that account for 1.17% loss in total mass,
about twice as much as for the hydrocarbon treatment. At 48hr interval, the
experimental group lost 0.044583333g of water in average that account for 1.21%
loss in total mass while the control lost 0.0927g of water in average that account for
2.37% loss in total mass. At 72hr interval, the experimental group lost 0.059566667g
of water in average that account for 1.97% loss in total mass while the control lost
0.14145g of water in average that account for 4.19% loss in total mass, more than
twice as much as the experimental treatment.
Figure 3.3.1. The amount (g) of water loss between two experimental groups at 72hr
interval. There is a statistical difference between two experimental
groups. Bars represent mean values +/- s.e.m. The asterisk above the
line indicates statistical differences.
Table 3.3.2. Means and raw scores of water losses in grams.
24hr interval Mean difference 48hr interval Mean difference 72hr interval Mean difference
HC 0.0416 0.049333333 0.0367 0.044583333 0.0702 0.059566667
0.0554 0.0533 0.1035
0.0316 0.0189 0.0391
0.0271 0.0248 0.0263
0.0741 0.0719 0.058
0.0662 0.0619 0.0603
Control 0.0961 0.091066667 0.0931 0.0927 0.197 0.14145
0.0807 0.0814 0.164
0.0991 0.0935 0.1898
0.0846 0.0967 0.1015
0.1154 0.1104 0.1113
0.0705 0.0811 0.0851
The average water loss (g) between the snake skins treated with HC
extract from T. caespitum and snake skins without HC treatment.
I conclude the following about my data:
1. Tetramorium caespitum uses cues present in cuticular hydrocarbons to
recognize non-nestmate conspecific ants and heterospecific ants. Focal ants
responded to the hydrocarbon stimuli of non-nestmate conspecifics and from
heterospecific ants with significantly more aggression compared to the stimuli
of nestmate hydrocarbons or the blank control, as measured using the
ethogram scale. A significantly higher proportion of individuals exhibited
agonistic responses towards hydrocarbon stimuli from non-nestmate
conspecifics and from heterospecific species compared to those from
nestmates or the blank control.
2. The duration of isolation significantly affected the nestmate recognition of T.
caespitum. Four weeks of isolation was sufficient to lead to an aggressive
response towards former nestmates.
3. Cuticular hydrocarbons from T. caespitum prevented approximately 1.54
times the water loss in comparison to snake skin barriers not treated with
4.2 Recognition Behavior and Hydrocarbons
Recognition cues are expressed in phenotypic traits, including chemicals. Variation
in such traits provides information that allows evaluators to discriminate the group
membership of the cue-bearer. The mechanism of discriminating non-self from
self, or nestmate from non-nestmate, is thought to depend upon the detection of
differences in the relative abundance of cuticular hydrocarbons (Greene & Gordon,
2007, Lucas et al., 2005). Many other researchers have demonstrated that the
cuticular hydrocarbons are an expression component of nestmate recognition among
several species of ants (Lucas et al., 2005, Wagner et al., 2000, Lahav et al., 1999).
Based on the knowledge from historical studies, I predicted that the pavement ant T.
caespitum also utilize cuticular hydrocarbon profiles as nestmate recognition cues.
This study demonstrated the ability of T. caespitum to use social recognition cues
present in cuticular hydrocarbons in both nestmate recognition and heterospecific
recognition (Figure 3.1.1, Figure 3.1.2, Figure 3.1.3). Specifically, the cuticular
hydrocarbons were demonstrated as the expression component of nestmate
recognition among T. caespitum. The action component of nestmate recognition
among T. caespitum was observed as well using ethogram scaling. Aggressive
behavioral responses of focal ants demonstrated the ability of T. caespitum to respond
to non-nestmates differently from nestmates.
My study is one of only a handful of studies in ants that demonstrates nestmate
recognition cues are present in cuticular hydrocarbons. Wagner et al. (2000) showed
that the cuticular hydrocarbon fraction of cuticular lipid extract alone was sufficient
to elicit aggression toward non-nestmates among the harvester ant Pogonomyrmex
barbatus. My data comply with their study examining relation between hydrocarbons
and nestmate recognition in ants. Other researchers have also suggested that the
hydrocarbon profiles were recognition cues for ants (Lahav et al., 1999, Lucas et al,
2005, Thomas et al., 1999, Wagner et al., 2000).
T. caespitum was more aggressive toward glass beads coated with non-nestmate
hydrocarbon than towards glass beads coated with nestmate hydrocarbon as measured
as both proportion of aggressive ants (Figure 3.1.1 and Table 3.1.1) and in ethogram
score (Figure 3.1.2, Figure 3.1.3, & Table 3.1.2, Table 3.1.4). Because the glass
beads coated with pentane that was used as solvent in the extraction process did not
differ from nestmate, the nestmate recognition cue of T. caespitum is present in
cuticular hydrocarbon profiles.
The significant difference between the proportions of aggressive workers towards
nestmate stimuli and non-nestmate stimuli suggests that the majority, if not all, of
workers are able to discriminate non-nestmate from nestmate by detecting the cues
presented in cuticular hydrocarbons. The proportion of aggressive ants towards non-
nestmate hydrocarbon stimuli was 55.6% while the proportion of aggressive ants
towards nestmate hydrocarbon stimuli was 2.38% (Table 3.1.1). This also suggests
that a colony responds aggressively to non-nestmates as a whole. However, the
proportion of focal aggressive ants towards non-nestmate hydrocarbon stimuli was
still far from 100%. The possible explanations can be geographic affinity or
unicoloniality, or possibly because of limitations in my experimental methods.
Some argue that the importance of geographic affinity and the history of interaction
between members of different colonies have important roles in development of the
action component of nestmate recognition in particular species of ants (Melissa et al.,
1999, Gordon, 1989). Gordon (1989) stated that workers of P. barbatus showed
higher aggression towards workers of P. barbatus from neighboring colonies than
towards workers from colonies that were more than 150m apart. Thomas et al. (1999)
demonstrated that non-nestmate aggression of Australian meat ant Iridomyrmex
purpureus was higher when the density of surrounding conspecific nest was also
high, and they were less aggressive towards non-nestmates of conspecific from far
colonies compare to non-nestmates of conspecific from close colonies by using
The other possible explanation for presence of non-aggressive workers can be
unicoloniality. Argentine ants (Linepithema humile) are well known for their
unicoloniality (Giraud et al., 2002) in which intraspecific aggression is reduced and
workers and queens can freely move between different colonies (Passra, 1994). The
bottleneck effect of an introduced population is thought to be the cause of
unicoloniality among L. humile (Suarez et al., 1999). The polygynous colonial
structure might also lead to unicioloniality in some species of ants by reducing the
genetic relatedness among nestmates (Steiner et al., 2003, van der Hamme et al.,
2002). T. caespitum obtain both of those characteristic of possible explanation for
origin of unicoloniality, they are widespread (McGlynn, 1999) and it is possible that
they experienced bottleneck effect in their evolutionary path, and they form
polygynous colonial structure (Steiner, et al., 2003, Brian, et al., 1967) that can
reduce genetic relatedness among nestmates. Steiner et al. (2003) has reported the
data that possibly suggests unicoloniality among T. caespitum.
Geographic patterns were not analyzed in this study, so the possibility of an effect
due to geographic affinity cannot be ignored. Data obtained from the experiment of
experiment 1 of the nestmate recognition showed a trend favoring the presence of
certain degree of unicoloniality among T. caespitum in which there were uneven
aggression in the ethogram scores towards different non-nestmate hydrocarbon
stimulus (see section 3.1). However, there were no statistical differences between
two different non-nestmate of conspecific in experiment 2 of nestmate recognition (p
> 0.05, Figure 3.1.3).
My results are among the few showing that ants use cues in hydrocarbons to
recognize other species of ants. There were no significant differences in the level of
aggression against either non-nestmate of conspecific species or heterospecific
species. In other words, the ants responded similarly to members of their own species
that do not belong to their colony and to other species of ants. Lucas (2005)
demonstrated that P. villosa and P. inversa responded differently to the P. subversa,
but P. subversa did not respond differently to P. villosa and P. inversa. All three
species showed higher levels of aggression toward heterospecifics than toward
conspecifics. My results do not agree with those of Lucas et al. (2005). Marier et al.
(2004) also showed difference in the agonistic behavior of Crematogaster scutellaris
towards non-nestmate conspecific and towards heterospecific species. Crematogaster
scutellaris responded with classical aggressive behavior such as grips to non-nestmate
of conspecific while they often engage in the secretion of toxic venom to the
hetrospecific intruders. Therefore, my study was not specifically able to investigate
whether T. caespitum can discriminate non-nestmate conspecifics from heterospecific
species because the ants did not respond differently to stimuli. Although it was not
clear if T. caespitum identified heterospecific species from non-nestmate of
conspecific, this study still demonstrated that they can recognize nestmates from
heterospecific species by detecting cues presented in the cuticular hydrocarbon.
Further, it suggests that the expression component of nestmate recognition of ants
rely mainly on the cuticular hydrocarbons.
Recent studies have also revealed more detailed mechanisms about the nestmate
recognition cue. Lucas et al. (2005) showed that internally branched methyl- and
dimethylialkanes are involved in recognition behavior of Pachycondyla villosa, P.
inversa and P. subversa. Structural complexity of classes of hydrocarbon such as
methyl-alkanes, n-alkanes, and n-alkenes are more important in the nestmate
recognition mechanisms of ant species rather than hydrocarbons of single class alone
(Greene & Gordon, 2007). However, this study did not investigate detailed
mechanisms of the nestmate recognition behavior of ants and the complexity of role
4.3 Significance of Social Interaction
The hydrocarbon profile is known to have plasticity. Profiles can change in response
to the type of food ants eat and to various environmental factors, including
temperature and humidity (Liu et al., 2001, Tissot et al., 2001, Wagner et al., 2001,
Vander Meer & Morel, 1998). The colonial signature of hydrocarbon profiles is
updated within the colony through antennation and trophollaxis, creating a gestalt
colony odor (Lucas et al., 2004, Boulay et al., 2000, Soroker et al., 1995).
From data in those previous studies, I predicted that prolonged isolation and
differences in environmental conditions would alter nestmate recognition of T.
caespitum. I found that two weeks of isolation altered their nestmate recognition
behavior. Ants responded to former nestmates with higher levels of aggression as the
duration of isolation was prolonged (Table 3.2.1). However, repeated-measure
ANOVA suggested there were no differences in the environmental conditions on the
nestmate recognition cue, indicating that the nestmate signatures among experimental
populations were being altered through time and must have been maintained through
physical contact such as trophallaxis. It is also possible that all environmental
conditions caused changes to the recognition cue that led to recognition response,
although this is unprecedented in the literature. Those physical contacts must also
have updated their templates as well.
Boulay et al. (2000) showed the aggressive responses of the resident nestmate
towards isolated nestmate in C. fellah. This study demonstrated similar results to
mine, in which workers of T. caespitum among each experimental population
responded aggressivly towards isolated nestmate that were isolated for 28 days.
Because of the plasticity in the hydrocarbon profiles (Liu et al., 2001, Tissot et al.,
2001, Wagner et al., 2001, Vander Meer & Morel, 1998), perception component,
templates, also has to be updated in order to recognize nestmates. The significant
change in the hydrocarbon profiles according to time, the diverging effect of isolation
of population on the hydrocarbon profiles, and the importance of social interaction for
maintaining Gestalt colony odor are well reported (Katazav-Gozansky et al. 2004,
Boulay & Lenoir, 2001, Tissot et al., 2001, Boulay et al., 2000). The reduction of
aggression towards non-nestmate conspecific among isolated workers of C. fellah
while being attacked from their former nestmates also suggest that importance of
social interaction to maintain or reminding each individuals their own identity
(Boulay & Lenoir, 2001).
Although this study could not investigate changes in profiles of hydrocarbons or
neural activity of the antennae, I demonstrated the importance of social interaction in
maintaining nestmate recognition mechanisms in T. caespritum by observing
alterations of the behavioral or action component of nestmate recognition. In this
study, workers of each experimental group were able to keep physical contact such as
trophallaxis and antennuation. Archiving each sample taken every 7 days allowed me
to test if the hydrocarbon profiles of a population constantly changed. The data
showed the experimental populations were aggressive towards their own nestmate
hydrocarbon from 28 days earlier. In other words, the colonial signature of cuticular
hydrocarbon profiles can diverge within 4 weeks. Data from this study suggested that
the alteration of colonial hydrocarbon signature is gradual and it is modified
constantly (Figure 3.2.1, Table 3.2.1), which complies with results reported
previously from other species of ants (Katazav-Gozansky et al., 2004, Boulay &
4.4 Physiological Function of Hydrocarbons
The cuticular hydrocarbons protect insects from desiccation (Hadley, 1984).
However, few researchers have quantified or demonstrated the anti-desiccation
properties of the cuticular hydrocarbon. In this study, I conducted a novel experiment
that measured the anti-desiccation property of cuticular hydrocarbons. I predicted
that the treatment of snakeskin with cuticular hydrocarbon extracts would reduce the
amount of water loss through evaporation. The result showed that there was
significant decline in the amount of water loss through the skin, thus demonstrating
that hydrocarbons had anti-desiccation properties. The snake skin barriers without
hydrocarbons applied to them lost about twice as much water as those with
Water is essential to life and water within the body has to be balanced in order to
maintain the physiological functions. T. caespitum prefers a temperature range of 10-
40C (Holldobler & Wilson, 1990), and the rate of evaporative cooling differs among
varying climate conditions. Therefore, the component or amount of the cuticular
hydrocarbon must be the environmentally dependent. These have suggested that the
cuticular hydrocarbon profiles must be plastic (Liu et al., 2001, Tissot et al., 2001,
Wagner et al., 2001, Vander Meer & Morel, 1998). As Hadley suggested (1984), the
primary function of the cuticular hydrocarbon is thought to be protection from water
loss. Nestmate recognition that utilizes hydrocarbons as perception components
might be secondary in the evolutionary processes. If the primary function of cuticular
hydrocarbons was to maintain water balance in insects, cuticular hydrocarbons have
to have certain degree of plasticity in order to adopt climate change. Therefore,
nestmate recognition utilizing hydrocarbons as secondary evolutionary process
supports the plasticity of cuticular hydrocarbons and importance of social interaction
for maintenance ofGestalt colony odor.
The studies conducted for this thesis project successfully supported the hypotheses
stated prior to the study: 1) T. caespitum utilizes cuticular hydrocarbon for their
nestmate and species recognition, 2) social isolation alters nestmate recognition
mechanisms, in other words, social interaction is necessary to maintain colonial
structures, and 3) protection from water loss is the primary function of cuticular
I demonstrated the ability of T. caespitum workers to utilize cuticular hydrocarbon
profile for nestmate and species recognition and the importance of social interaction
to maintain nestmate recognition ability at colonial levels. Future studies are needed
to investigate if T. caespitum are able to identify non-nestmate conspecific ants from
heterospecific ant species, or if both are the same recognition response to non-self.
I also found evidence for unicoloniality. To further investigate unicoloniality,
geographic comparison of aggressive behavior and DNA analysis might be useful,
and chemical analyses of hydrocarbon profiles from focal samples and isolation study
are still need to be done. The chemical analyses might help further investigate
environmental effect on alteration of cuticular hydrocarbons and also add to the
supporting evidence of unicoloniality among pavement ant T. caespitum. Also, this
study clearly demonstrated anti-dessication propertiy of cuticular hydrocarbons.
Adams, E.S. (1991). Nest-mate recognition based on heritable odors in the termite
Microceroermes arboreus. Proceedings of the National Academy of Sciences of the
United States of America, 88, 2031 -2034
Bonavita-Cougourdan, A., Theraulaz, G., Bagneres, A.G., Roux, M., Pratte, M., Provost,
E., & Clement, J.L. (1991). Cuticular hydrocarbons, social organization and ovarian
development in a polistine wasp: Polistes dominulus Christ. Comparative
biochemistry and physiology, 100, 667-680
Boulay, R., Hefetz, A., Sorker, V., & Lenoir, A. (2000). Camponootus fellah colony
integration: worker individuality necessitates frequent hydrocarbon exchanges.
Animal behavior, 59, 1127-1133
Boulay, R. & Lenoir, A. (2001). Social isolation of mature workers affects nestmate
recognition in the ant Camponotus fellah. Behavioural Processes, 55, 67-73
Breed, M.D., Guzman-Novoa, E., & Hunt, G.J. (2004). Defensive behavior of honey
bees: Organization, genetics, and comparisons with other bees. Annual review of
entomology, 49, 271-298
Breed, M.D., Perry, S., & Bjostad, L.B. (2004). Testing the blank slate hypothesis: why
honey bee colonies accept young bees. Insectes Sociaux, 51, 12-16
Brian, M.V., Elmes, G., & Kelly, A.F. (1967). Population of the ant Tetramorium
caespitum Latreile. Journal of Animal Ecology, 36, 337-342
Buchwald, R. & Breed, M.D. (2005). Nestmate recognition cues in a stingless bee,
Trigonafulviventris. Animal Behavior, 70, 1331-1337
Cheney, D.L., Seyfarth, R.M. & Smuts, B.B. (1986). Social relationships and social
cognition in non-human primates. Science, 234, 1361-1366
Cranshaw, W.S. (2003). Home & Garden. Retrieved January 4, 2007, from the
Colorado State University website:
Cuvillier-Hot, V., Lenoir, A., Crewe, R., Malosse, C., & Peeters, C. (2004). Fertility
signaling and reproductive skew in queenless ants. Animal Behavior, 68, 1209-1291
Darlington, J.P.E.C. (1994). Nutrition and evolution in fungus-growing termites. In J.H.
Hunt & C.A. Nalepa (Ed.), Nourishment and Evolution in Insects Societies (pp. 105-
130. Boulder, CO: Westview Press.
Dietemann, V., Peeters, C., Liebig, J., Thivet, V., & Holldobler, B. (2003). Cuticular
hydrocarbons mediate discrimination of reproductives and nonreproductives in the
ant Myrmecia gulosa. Proceedings of the National Academy of Sciences of the
United States of America, 100 (18), 10341-10346
Espelie, K.E., Gamboa, G.J., Grudzien, T.A., & Bura, E.A. (1994). Cuticualr
hydrocarbons of the paper wasp, Polistes fuscatus: a search for recognition
pheromones. Journal of Chemical Ecology, 20, 1677-1687
Faulkes, C.G. & Abbott, D.H. (1997). Proximate mechanisms regulating a reproductive
dictatorship: a single dominant female controls male-rats. In N.G. Solomon & J.A.
French (Eds.), Cooperative Breeding in Mammals (pp. 302-334). Cambridge
Faulkes, C.G. & Bennett, N.C. (2001). Family values: group dynamics and social
control of reproduction in African mole-rats. TREND in Ecology & Evolution, 16 (4),
Fletcher, D.J.C. (1987). The behavioral analysis of kin recognition: perspective on
methodology and interpretation. In D.J.C. Fletcher & C.D. Michener (Eds.),
Kin recognition in animals (pp. 19-54). New York: Wiley.
Gibbs, A.G. (2002). Lipid melting and cuticular permeability: new insights into an old
problem. Journal of Insect Physiology, 48, 391-400
Giraud, T., Pedersen, J.S., & Keller, L. (2002). Evolution of supercolonies: The
Argentine ants of southern Europe. Evolution, 99(9), 6075-6079
Gobin, B., Ito, F., & Billen, J. (2003). The subepithelial gland in ants: a novel exocrine
gland closely associated with the cuticle surface. Acta Zoologica (Stockholm), 84,
Goldsby, A.R., Kindt, T.J., Osborne, B.A., & Kuby, J. (2003). Immunology. (5th ed.).
New York: W.H. Freeman and Company.
Gordon, D.M. (1987). Group-level dynamics in harvester ants: young colonies and the
role of patrolling. Animal Behavior, 35, 833-843
Gordon, D.M. (1989). Dynamics of task switching in harvester ants. Animal Behavior,
Greene, M.J. & Gordon, D.M. (2003). Cuticular hydrocarbons inform task decisions.
Nature, 423: 32
Greene, M.J. & Gordon, D.M. (2007). Structural complexity of chemical recognition
cues affects the perception of group membership in the ants Linephithema humile and
Aphaenogaster cockerelli. The Journal of Experimental Biology, 210, 897-905
Greene, M.J. & Gordon, D.M. (2007). Interaction rate informs harvester ant task
decisions. Behavioral Ecology, 18 (2), 451-455
Hackman, R.H. (1984). Cuticle: Biochemistry. In J. Bereiter-Hahn, A.G. Matolsty, &
K.S. Richards (Eds.), Biology of the integument, Vol. 1 (pp.583-610). Berlin: Spring-
Hadley, N.F. (1984). Cuticle: Ecological significance. In J. Bereiter-Hahn, A.G.
Matolsty, & K.S. Richards (Eds.), Biology of the integument, Vol. 1 (pp.583-610).
Hamilton, W.D. (1964). The genetical theory of social behavior. Journal of
Theoretical Biology (1,11), 7, 1-52
Haverty, M.I. & Thorne, B.L. (1989). Agonistic behavior correlated with hydrocarbon
phenotypes in dampwood termites, Zootermopsis (Isoptera: Termopsidae). Journal of
Insect Behavior, 2, 523-543
Haverty, M.I., Copren, K.A., Getty, G., & Lewis, V.R. (1999). Agonistic behavior and
cuticular hydrocarbon phenotypes of colonies of Reticulitermes (Isoptera:
Phinotermitidae) from northern California. Annals of the Entomological Society of
Hefetz, A., Errard, C., Chambris, A., & Le Negrate, A. (1996). Postpharyngeal gland
secretion as a modifier of aggressive behavior in the myrmicinc ant Manica rubida.
Journal of Insect Behavior, 9, 709-717
Holldobler, B. (1995). The chemistry of social regulation: multicomponent signals in
ant societies. Proceedings of the National Academy of Sciences, U.S.A., 92, 19-22
Holldobler, B. & Wilson, E.O. (1990). The Ant. Cambridge, Massachusetts: The
Belknap Press of Harvard University Press.
Holldobler, B. & Wilson, E.O. (1994). Journey to the ants: a story of scientific
Howard, R.W. (1993). Cuticular hydrocarbons and chemical communication. In D.W.
Stanley-Samuelson & D.R. Nelson (Eds.), Insect Lipids: Chemistry, Biochemistry and
Biology (pp. 179-226). Lincoln: University of Nebraska Press.
Howard, R.W. & Blomquist, G.J. (2005). Ecological, behavioral, and biochemical
aspects of insect hydrocarbons. Annual Review of Entomology, JO, 371-393
Johanson, C.A., Vander Meer, R.K., & Lavine, B. (2001). Changes in the cuticular
hydrocarbon profile of the slave maker ant queen, Poyergus breviceps Emery, after
killing a Formica host queen (Hymenoptera: Formicidae). Journal of Chemical
Ecology, 27(9), 1787-1803
Jungnickel, H., da Costa, A.J.S., Tentschert, J., Patricio, E.F.L.R.A., Imperatiz-Fonseca,
V.L., Drijifhout, F., & Morgan, E.D. (2004). Chemical basis for inter-colonial
aggression in the stingless bee Scaptotrigona bipunctata (Hymenoptera: Apidae).
Journal of Insect Physiology, 50, 761-766
Kaib, M., Jmhasly, P., Wilfert, L., Durake, W., Franke, S., Francke, W., Leuthold, R.H.,
& Brandi, R. (2004). Cuticular hydrocarbons and aggression in the termite
Macrotermes subhyalinus. Journal of Chemical Ecology, 30(2), 365-385
Katav-Gonzansky, T., Boulay, R., Vander Meer, R., & Hefetz, A. (2004). In-nest
environment modulates nestmate recognition in the ant Camponotus fellah.
Naturwossenschaften, 91(4), 186-190
Kirchner, W.H. & Minkeley, N. (2003). Nestmate discrimination in the harvester
termite Hodotermes mossambicus. Insect Society, 50, 222-225
Lahav, S., Soroker, V., & Hefetz, A. (1999). Direct behavioral evidence for
hydrocarbons as ant recognition discriminators. Naturwissenschaften,
Liang, D., Blomquist, G.J., & Silverman, J. (2001). Hydrocarbon-released nestmate
aggression in the Argentine ant, Linepithema humile, following encounters with
insect prey. Comparative Biochemistry and Physiology, 129 (B), 871-882
Liebere, A.E. & Starks, P.T. (2004). The action component of recognition system: a
focus on the response. Annales Zoologici Fennici, 41, 747-764
Liang, D. & Siverman, J. (2000). You are what you eat: Diet modifies cuticular
hydrocarbons and nest-mate recognition in Argentine ant, Linepithema humile.
Naturwissenschaften, 87, 412-416
Liu, Z.B., Bangneres, A.G., Yamane, S., Wang, Q.C., & Kojima, J. (2001). Intra-
colony, inter-colony and seasonal variations of cuticular hydrocarbon profile in
Formica japonica (Hymenoptera, Formicidae). Insecte Sociaux, 48, 342-346
Lockey, K.H. (1988). Lipids of insect cuticle: origin, composition and function.
Comparative Biochemistry and Physiology, 89(B), 595-645
Lucas, C., Fresneau, D., Kolmer, K., Heinze, J., Delabie, J.H.C., & Pho, D.B. (2002). A
multidisciplinary approach to discriminating different taxa in the species complex
Pachycondyla villosa (Formicidae). Biological Journal of the Linnean Society, 75,
Lucas, C., Pho, D.B., Fresneau, J.M., & Jallon, J.M. (2004). Hydrocarbon circulation
and colonial signature in Pachyconfyla villosa. Journal of Insect Physiology, 50,
Lucas, C., Pho, D.B., Jallon, J.M., & Fresneau, D. (2005). Role of cuticular
hydrocarbons in the chemical recognition between ant species in the Pachycondyla
villosa species complex. Journal of Insect Physiology, 51, 1148-1157
Marlier, J.F., Quinet, Y., & de Biseau, J.C. (2004). Defensive behaviour and biological
activities of the abdominal secretion in the ant Crematogaster acutellaris
(Hymenoptera: Myrmicinae). Behavioral Processes, 67, 427-440
Mateo, J.M. (2004). Recognition system and biological organization: The perception
component of social recognition. Annales Zoologici Fennici, 41, 729-745
McGlynn, T.P. (1999). The worldwide transport of ants: geographic distribution and
ecological invasions. Journal of Biogeography, 26, 535-548
Nelson, D.R. & Blomquist, G.J. (1995). Insect waxes. In R.J. Hamilton (Ed.), Waxes:
Chemistry, Molecular Biology and Functions (pp.1-90). Dundee, Scotland: Oily
Ozaki, M., Wada-Katsumata, A., Fujikawa, K., Iwasaki, M., Yokohari, F., Satoji, Y.,
Nisimura, T., & Yamaoka, R. (2005). Ant nestmate and non-nestmate discrimination
by a chemosensory sensillum. Science, 309, 311-314
Passera, L. (1994). Characteristics of tramp species. In D.F. Williams (Ed.), Exotic
ants: Biology, impact, and control of introduced species (pp. 23-43). Boulder CO:
Reeve, H.K. (1989). The evolution of conspecific acceptance thresholds.
The American naturalist, 133, 407-435
Robinson, G.E. (2001). From society to gene with honey bee: A combination of
environmental, genetic, hormonal and neurobiological factors determine a bees
progression through a series of life stages. In P. W. Sherman & J. Alcock (Eds.),
Exploring animal behavior: Readings from American Scientist (3rd ed., pp.60-66).
Sunderland, Massachusetts: Sinauer Associates, Inc.
Ruther, J., Sieben, S., & Schricker, B. (2002). Nestmate recognition in social wasps:
manipulation of hydrocarbon profiles induces aggression in the European hornet.
Naturwissenschaften, 89, 111-114
Scharf, M.E., Ratliff, C.R., & Bennett, G.W. (2004). Impacts of residual insecticide
barriers on perimeter-invading ants, with particular reference to the odorous host ant,
Tapinoma sessile. Journal of Economic Entmology, 97, 601-605
Seeley, T.D. (2001). The honey bee colony as a superorganism. In P.W. Sherman & J.
Alcock (Eds.), Exploring animal behavior: Readings from American Scientist (3rd ed.,
pp.294-301). Sunderland, Massachusetts: Sinauer Associates, Inc.
Sendova-Franks, A.B., Franks, N.R., & Britton, N.F. (2002). The role of competition in
task switching during colony emigration in the ant Leptothorax albipennis. Animal
Behavior, 63, 715-725
Sherman, P.W. & Holmes, W.G. (1985). Kin recognition: issue and evidence.
Fortschritte der Zoologie, 31, 437-460
Sherman, P.W., Reeve, H.K., & Pfenning, D.W. (1997). Recognition systems.
In J.R. Krebs & N.B. Davies (Eds.), Behavioral ecology: an evolutionary approach
(pp. 69-96). Oxford, United Kingdom: Blackwell Scientific Publication.
Singer, T.L. (1998). Role of hydrocarbons in the recognition systems of insects.
American Zoologist, 38, 394-405
Sledge, M., Boscaro, F., & Turillazzi, S. (2001). Cuticular hydrocarbon and
reproductive status in the social wasp Polstes dominulud. Behavioral Ecology and
Sociobiology, 49 (5), 401-409
| Soroker, V., Vienne, C., Nowbahari, E., & Hefetz, A. (1994). The postpharyngeal
| gland as a Gestalt organ for nestmate recognition in the ant Cataglyphis niger.
j Naturwissenschaften, 81, 510-513.
i Soroker, V., Hefetz, A., Cojocaru, M., Billen, J., Franke, S., & Francke, W. (1995).
! Structural and chemical ontogeny of the post-phaiyngeal gland in the deseart ant
: Cataglyphis niger. Physiological Entomology, 20, 323-329
| Steiner, F.M., Schilick-Steiner, B.C., & Buschinger, A. (2003). First record of
j unicolonial polygyny in Tetramorium cf. caespitum (Hymenoptera, Formicidae).
| Insectes Sociaux, 50, 98-99
Thomas, M.L., Parry, L.J., Allan, R.A., & Eiger, M.A. (1999). Geographic affinity,
cuticular hydrocarbons and colony recognition in the Australian meat ant
Iridomyrmex purpureus. Naturwissenschaften, 86, 87-92
j Tissot, M., Nelson, D.R., & Gordon, D.M. (2001). Qualitative and quantitative
l differences in cuticular hydrocarbons between laboratory and field colonies of
j Pogonomyremex barbatus. Comparative Biochemistry and Physiology, 130 (B), 349-
Tsutsui, N.D. (2004). Scent of self: The expression component of self/non-self
recognition systems. Annales Zoologici Fennici, 41, 713-727
UC IPM Online (2005). Key to identify common household ants. Retrieved March 13,
2007, from the University of California website:
van der Hammen, T., Pedersen, J.S., & Boomsma, J.J. (2002). Convergent developmnt
of low-relatedness supercolonies in Myrmica ants. Heredity, 89, 83-89
Vander Meer, R.K. & Morel, L. (1998). Nest-mate recognition in ants. In R.K. Vander
Meer, M.D. Breed, K.E. Espelie, & M.L. Winston (Eds.), Pheromone Communication
in Social Insects: Ants, Wasps, Bees and Termites (pp. 79-103). Boulder, CO:
Venkataraman, A.B., Swarnalatha, V.B., Nair, P., Vinutha, C., & Gadagkar, R. (1990).
Nestmate discrimination in the social wasp Ropalidia marginata. In G.K. Veeresh &
T. Shivashankar (Eds.), Social insects: An Indian Perspective (pp. 161-171).
Bangalore: Indian Chapter.
Wagner, D., Brown, M.J.F., Broun, P., Cuevas, W., Moses, L.E., Chao, D.L., & Gordon,
D.M. (1998). Task-related differences in the cuticular hydrocarbon composition of
harvester ants, Pogonomyrmex barbatus. Journal of Chemical Ecology, 24 (12),
Wagner, D., Tissot, M., Cuevas, W., & Gordon, D.M. (2000). Harveter ants utilize
cuticular hydrocarbons in nestmate recognition. Journal of Chemical Ecology, 26
Wagner, D., Tissot, M., & Gordon, D. (2001). Task-related environment alters the
cuticular hydrocarbon composition of harvester ants. Journal of Chemical Ecology,
Wheeler, W.M. (1928). The social insects: Their origin and evolution. London: Kegan
Paul, Trench, and Trubner & Co.
Wild, A. (2007). Myrmecos.net. Retrieved March 13, 2007, from:
Wilson, E.O. (1975). Sociobiology: The new synthesis. Cambridge, Massachusetts:
The Belknap Press of Harveard University Press.
Winston, M.L. & Slessor, K.N. (2001). The essence of royalty: Honey bee queen
pheromone. In P.W. Sherman & J. Alcock (Eds.), Exploring animal behavior:
Readings from American Scientist (3rd ed., pp.294-301). Sunderland, Massachusetts:
Sinauer Associates, Inc.