Aggression and Competition in Two Boreal Animals

Aggression and competition in two boreal animals.

Jennifer E. Van Eindhoven

A thesis submitted in conformity with the requirements for the degree of Master of Science.

Ecology and Evolutionary Biology University of Toronto

Aggression and competition in two boreal animals.

Jennifer E. Van Eindhoven

Master of Science

Ecology and Evolutionary Biology

University of Toronto

2012

Abstract:

This thesis will focus on the use of agonistic behaviour in both direct and indirect competition in two arboreal species: the eastern chipmunk (Tamias striatus) and Cyphoderris monstrosa, a primitive orthopteran insect. Chipmunks are an example of indirect competitors as they are competing for finite resources for dealing with abiotic stresses to ensure their survival.

Chapter 2 of this project investigates the behavioural time budget for above ground activity at a time of critical importance for overwinter survival. The chipmunks in this study displayed a focus of their time budget on eating and collecting food in preparation of winter survival while they were above ground. Chapter 3 of this research project studies the physiological differences between males of Cyphoderris monstrosa which engage in aggressive territorial contests. The data suggest that metabolic scope is correlated with RHP. Males’ ability to mobilize energy reserves may be an important factor in contest outcomes.

Acknowledgements

I would like to start off by thanking my parents John Van Eindhoven and Debbie Spencer (married to my stepdad Dick) for always opening their warm homes and hearts whenever I need them. Whether its phone calls, dinners, emails, texts, baked goods, time spent away from technology, walks with the dogs you always know what I need to put everything into perspective. I love you so much and I am very grateful that I was able to go through this with you behind me 110% and being my biggest supporters.

A big thank you goes to Brendan Delehanty and Lanna Desantis for the technical assistance, guidance, general acceptance into the lab and for teaching me how to trap small mammals. Importantly, I would like to extend a thank you to my close friend and field assistant Bryan Taylor. Bryan, thank you for all of your hard work in the field and for the personal support when things became difficult both professionally and personally for me. I truly do value all of the work you did with me and how great of a friend you are.

I would also like to thank the members of the Mason Lab: Dean Koucoulas, Norman Lee, Paul DeLuca and Sen Sivalinghem for welcoming me into their lab. You all have been the source of thoughtful discussions and personal support. Maria Modanu you have been more than patient and kind when teaching me how to use the respirometry equipment which is a large part of the Cyphoderris project.

My committee members (Ken Welch, Maydianne Andrade and Andrew Mason) have all been a wealth of information and understanding. Ken you have been beyond helpful when trying to get through the respirometry data. You have also given rise to thought provoking discussions and challenging questions that keep me on top of things when preparing for meetings. Maydianne thank you so much; you were my saving grace when it came to dealing with a difficult situation, without your guidance and support I truly feel that the personal outcome of this degree would be drastically different. I honestly can’t express the gratitude that I feel. I need to say thank you to you, Andrew and the powers that be for providing me with an alternative when I didn’t know that there was one.

In turn, Andrew, thank you for accepting me into your lab. I honestly felt that I would end up somewhere that was suitable to obtain my degree and just bare down and get through it, however what I found was a lab full of people that mean a great deal to me, research that I find interesting and exciting and a supervisor that I am honoured to have. You’re guidance in the past year personally and professionally has meant a great deal to me and I am beyond thankful for that.

iii

Table of Contents

Acknowledgements ...... iii

Table of Contents ...... iv

List of Tables ...... vi

List of Figures ...... vii

Chapter 1: General Introduction ...... 1 Chapter 2: Indirect competition 1 Chapter 3: Direct competition 3

Chapter 2: The Behavioural Time Budget of the Eastern Chipmunk (Tamias striatus) While Preparing for the Induction of Winter Sleep………………………………………………………8 Introduction 8 Methods 12 Animals 12 Radio Telemetry 13 Behavioural Observations 13 Statistical Analysis 14 Results 15 Early October Behavioural Analysis 15 Late October Behavioural Analysis 16 Alteration in Behaviour over Time 16 Discussion 17

Chapter 3: Competitive signalling during aggressive encounters in the primitive acoustic insect Cyphoderris monstrosa (Orthoptera: Haglidae)...... 26 Introduction 26 Animal Contests: evolution 26 Animal Contests: energetics 27 Study Species 27 Methods 32 Animals 32 Fight Tournament 33 Energetic Rate Measurements 34 Song and Physical Characteristics 35 Statistical Analysis 37 Results 38 Tournament Results 38 Morphology Comparisons 38

Metabolic Measurements 39 Discussion 39 Behavioural Patterns 39 Morphometrics 40 Metabolism 41

Chapter 4: General Conclusion ...... 51

Literature Cited ...... 53

List of Tables

Table 1: Categories of chipmunk behaviour and descriptions 21

Table 2: Principal components analysis of male C. monstrosa morphology measurements 44

Table 3: Male C. monstrosa tactics on the outcome of staged territorial contests 45

Table 4: Male C. monstrosa status (winner/loser) with PCA of morphological measurements 46

List of Figures

Figure 1: Number of observations of chipmunks for each week of study 22

Figure 2: The daily time budget in Early October (Oct 5-8, 2011) 23

Figure 3: The daily time budget in Late October (Oct 18-21, 2011) 24

Figure 4: The change in displayed behaviour from Early to Late October 25

Figure 5: Distribution of Cyphoderris in North America 47

Figure 6: Relationship of duty cycle and weight 48

Figure 7: Carbon dioxide production of singing males and their duty cycles 49

Figure 8: Metabolic scope of Winners and Losers in relation to their duty cycles 50

vii

Chapter 1 General Introduction

Competition is any interaction between individuals or species, sharing limited resources, that is mutually detrimental to both participants. Individuals competing for the same resources or objects with one another where physical interference occurs is direct competition, competition for resources without direct contact between individuals or the resource itself is indirect competition (Eccard et al, 2011; English-Loeb et al., 1993). Competition for resources between different species (interspecific competition) can occur for the use of an area or access to food sources and it can present within a species (intraspecific) for mate selection and gathering of specific food items. Thus the type of competition an individual faces in securing sought after resources is related to various conditions and pressures. This research paper will focus on agonistic behaviour in both direct and indirect competition in two arboreal species.

Chapter 2: Indirect competition-a behavioural ecology approach

Canada’s landscape is in a continuous state of transformation by the change of seasons throughout the year. One of the most extreme seasons is winter because of very cold temperatures and snow coverage. Animals deal with this in three very distinct ways. The first option is avoidance; some species (like many bird species) choose to migrate to southern areas and remove themselves from the extreme weather. This involves a great deal of stress as the animals have to travel great distances to reach subtropical to tropical regions (Speakman and

Rowland, 1999). Geese species around the world travel approximately 3000 ground kilometers south every fall (Pennycuick et.al. 2011) and caribou move hundreds of miles from their winter ranges in the taiga and their summer ranges on the tundra (Frame et al., 2008). The other radical option is to remain in the area and enter hibernation. Through reducing energy requirements and

1 experiencing a severe drop in body temperature, these individuals are able to cope with the lack of available resources needed to sustain their normal daily energy requirements. Hibernators, for example insectorivous bats (Speakman and Rowland, 1999) and the North American woodchuck

(Marmota monax) (Peppas et al., 2009), usually prepare for this by storing high levels of fat, which is used as the energy source for their survival, before entering their shelter for the entire winter.

Lastly, there are species that are neither able to leave the area nor enter hibernation.. These species have to stay and modify their behaviours or physical requirements to be able to cope; each has their own unique way of doing so. Some species make physiological modifications like thickening their fur or feathers or increasing their fat stores. Some become more social and huddle together for heat conservation. Some reduce their winter energy demands by reducing body size and there are some that remain active, store food while resources are available to do so and consume it during these times (Brodin and Clark, 1997; Speakman and Rowland, 1999).

The eastern chipmunk (Tamias striatus) is a solitarily animal living in a single burrow with a breeding season in the spring and occasionally in the fall (Elliott 1978; Pyare et al 1993). After establishment in their shelter they spend the majority of their time (approximately 75%) within

25 m of the burrow entrance and actively defend a 9-12 m radius around their burrow (Elliott

1978). Chipmunks collect food items like maple, beech and oak tree seeds (acorns being the preferred food source) in their extensible cheek pouches and deposit them in their burrow for later consumption (Elliott 1978; Lacher and Mares 1996; Pyare et al 1993). The energetic expense of collecting this stash is great, as much time is spent collecting the food, predation avoidance is energetically draining and locating enough nuts may be challenging if the area is not rich in resources (Kawamichi 1996).

Chipmunks are an example of indirect competitors as they are competing for finite resources for dealing with abiotic stresses to ensure their survival. Chipmunks have to collect enough food to survive, but are be competing with others doing the same. They may or may not fight over these resources. Chapter 2 of this project investigates the behavioural time budget for both daily and above ground activity at a time of critical importance for overwinter survival.

Chapter 3: Direct competition-a physiological approach

In some cases individuals are in direct competition for resources and they participate in a fight. Game theory is the study of conflicts between individuals in which the individual’s actions are influenced by their own choices as well as the choices of their opponents (Reichert and

Hammerstein 1983). It is entitled “Game Theory” because it is similar to many parlor games (for example: chess, bridge and poker), in which the players have to read their opponents and anticipate the possible outcomes of the game given each other’s reactions and decision making

(Reichert and Hammerstein 1983). From an evolutionary perspective the “decision” made by a

“player” (the species or a population of the species) of the game refers to the long-term outcome of selection pressured by environmental conditions (Reichert and Hammerstein 1983). Game theory has been extensively used to analyze these contests as this provides a framework for interpreting the costs and benefits associated with different behavioural options available to competing individuals and therefore making predictions about adaptive strategies.

Stable biological processes have acted as the foundation of the Evolutionary Stable Strategy

(ESS) through the analysis of classic simple animal conflicts (Maynard Smith and Price 1973 and Parker 1974). This provides the basis for more complex systems when organism interactions are evaluated in a nonstable environment (for example a population that has a high level of dispersal) (Prior, Hines and Cressman 1993). In a contest between two individuals, fight

3 asymmetries are the differences that relate to the probability of winning between contenders; these asymmetries can be assessed among one another in contests that involve little physical interaction or in cases of physical combat, the asymmetries are displayed in the behaviour or efficiency of weaponry (Maynard Smith and Price 1973). Limited research is found that studies fighting asymmetries in the field at the population level or in interspecific interactions in relation to metabolic processes.

According to the evolutionary stable strategy models there are two contest types: symmetrical and asymmetrical. In symmetrical contests both contestants are evenly matched and the winner is the one who is prepared to fight longer. In asymmetrical contests the winner is dependent on the nature and strength of the asymmetry. There are two types of asymmetry:

“resource-holding potential” (RHP), in which the individuals assess the differences in fighting abilities, and payoff asymmetry (also known as resource value), in which the differences in the consequences of winning or losing the contest for each contestant are assessed (Haley 1994).

In contests resolved through the repeated mutual assessment of relative RHP instead of fighting (Hack 1997), the winner of the contest wins a particular resource on the site while compelling the loser to leave the site. This is thought, in some cases, to be related to body size and may also depend on transitory changes in motivation or physiological state (Nijman and

Heuts 2000). Individuals’ decisions on escalation or retreat should be based on the dominant asymmetry. If the asymmetries are difficult to perceive the length of contests may be increased in order to gain reliable information (Haley 1994). An example presented by Haley (1994) is of the work by Austad (1983) who looked at contests between two spiders; when different in size the contests were short but when they were approximately the same size the contest was much

4 longer. Body size is a common quality assessed as it is physically obvious but other species may look at other characteristics.

A dyadic contest is one in which both individuals value the resources equally (Briffa

2008). When contestants value the resources equally, the contest should end when the weaker one (with lower RHP) makes strategic decision to give up or withdraw (Humphries et al 2006).

The winner demonstrates superior fighting ability or RHP relative to its opponent (Briffa and

Sneddon 2007). To gain more mating opportunities selection pressures would operate to enhance the acquisition or defense of resources (Briffa and Sneddon 2007). Throughout the interaction the losers may give up because of changes in physical state or based on accumulation of information throughout the contest duration (Taylor and Elwood 2003). Mutual assessment may be involved but the individuals also determine their own thresholds of the energetic costs, if they were to continue in the contest, and this is involved in the decision to give up (Briffa 2008).

Motivation may help overcome inferior RHP (motivational asymmetry) (Brown et al 2007).

Chapter 2 of this project will examine direct competition among males of the primitive acoustic insect Cyphoderris monstrosa (Orthoptera: Haglidae) in securing mates.

The family Haglidae dates back to the Triassic period and has seven extant species in four genera (Mason 1994). In China there are two species of the genus Aboiloilomimus:

Aboiloilomimus guizhouensis and Aboiloilomimus ornatus (Liu et al 2009). In North America there is one genus with three species present, the species include: Cyphoderris monstrosa,

Cyphoderris buckelli, and Cyphoderris strepitans. Cyphoderris monstrosa and Cyphoderris buckelli are located in south-western Canada north-western United States and Cyphoderris strepitans is located in north-western United States (Morris and Gwynne 1978). The remaining

5 two genera and species Prophalangopsis obscura and Paracyphoderris erebeus are located in

Asia (Dodson et al., 1983).

Cyphoderris monstrosa has undergone the most extensive morphological and behavioural divergence of all of the species (Kumala et al., 2005). Of the three species in North America, C. monstrosa is the only species that engages in aggressive territorial contests which sometimes include escalated fights. The interactions displayed by this species are examples of direct competition as males directly compete for access to trees and thus reproductive success.

C. monstrosa is found in the boreal forest in southern British Columbia and is typically located in the subalpine Forest. They range from the Canadian Rockies, of southwestern Alberta, east through southern British Columbia, and north of the Canadian/United States Border by 700 miles and south almost as far as Northern California. They can also be found in western Montana and central Idaho. They feed on the cones of lodge pole pines (Pinus contorta) and western white spruce (Picea glauca). In order to obtain the cones the crickets climb trees and feast, using limbs as perches (Morris and Gwynne, 1978).

Males call for females by tegminal stridulation, while females are apterous (Dodson et al.,

1983; Morris and Gwynne, 1978). Males create song so females can find them in their natural environment (Snedden and Sakaluk, 1992). Calling begins in the late evening (dusk) and continues past midnight. Singing success and reproductive success are closely related in acoustic insects. Calling is an energetically costly way to obtain potential mates; males are under selection to optimize the effectiveness of their calls (Bennet-Clark, 1998).

Singing is directly related to reproductive success for C. monstrosa. House crickets, Acheta domesticus, who also call to attract their mates, display the highest rate of energy use while singing compared to other activities, presumably due to the relationship with securing mates and

6 the consequent selection on males to maximize song output. These male crickets don’t have specialized weapons, their body size is not a determinant of fighting success, and like most studied cricket species dominance relationships are temporally unstable; thus transient characteristics (such as physical condition) are important in determining fighting success (Hack

1997), and energy resources investing in singing may be a significant factor.

Male size and morphology, independently, have been found to have no influence on the outcome of fights between male C. monstrosa with the only predictor to be the duty cycle of the individuals call (Mason 1996). Chapter 2 of this research project studies the physiological differences between males of varying fighting success. Using a tournament design, in which a cohort of males engaged in three rounds of aggressive territorial contests, I tested for metabolic and morphological correlates of fighting abilities, to try to determine the underlying differences in males that are related to RHP.

Chapter 2

The Behavioural Time Budget of the Eastern Chipmunk (Tamias striatus) While Preparing for the Induction of Winter Sleep.

Introduction

Intraspecific competition is one of the major factors affecting survival and reproductive success in animals. In some cases, animals compete directly and engage in aggressive interactions over specific resources (e.g. food, territory, mating opportunities). In others, competition is indirect, with individuals competing for acquisition of the same (sometimes limiting) resources, but not necessarily interacting directly with competitors. For example, in food hoarding animals, the acquisition of finite food resources may be limited by the activity of other conspecific foragers via their effect on the availability of food. Aggressive interactions may or may not be involved. In this chapter, I examine the behavioural time budget of eastern chipmunks to address the question of whether they show increased aggression as the summer season progresses and food supplies become scarcer.

Hoarding food for later consumption is a foraging technique utilized by many taxa

(Clarke and Kramer 1994). Animals store food during times of high food availability, later consuming this supply at times of food scarcity (Wauters et.al., 1995). Food items that are stored are valued by the nutritional content (how much energy can be obtained through consuming it)

(Jansen et. al., 2004). The hoarder will then consume the stored food at a later time. Hoarded food may be consumed hours to days later; or up to months later, to increase survival or reproductive success (Vander Wall and Jenkins, 2003; Wauters et al, 1995).

There are two different strategies used when hoarding: larder-hoarding; in which a large amount of food is stored at one or a few sites that are visited repeatedly; and scatter-hoarding, in which food is distributed throughout the home range and each storage site is re-visited only once

(Clarke and Kramer 1994; Wauters et al., 1995). The hoarding strategy and time period needed for the collection and storage of the food items are specific to each hoarding species and sometimes vary among individuals. Red squirrels are examples of long term scatter hoarders; by storing pine cones, acorns, and beechnuts in the fall for consumption in the spring, they increase survival and reproductive success (Wauters et al, 1995). Similarly, over-winter survival is increased for Acorn woodpeckers (Melanerpes formicivorus) (Hitchcock and Houston, 1994) and pikas (Ochotona princeps) (Kawamichi 1976) through larder hoarding in the fall. Hoarding food increases food availability during periods of scarcity and it decreases the possibility that naïve foragers will find the cached food, thus making it advantageous for the hoarding individual

(Wauters et al, 1995).

Food hoarding does, however, allow for the alternative strategy of cache-robbing (Galvez et al, 2009, Haugaasen and Tuck Haugaasen, 2010). Cache robbery is most common amoung scatter-hoarders although robbing of large caches from larder-hoarders is documented and this poses a large risk for hoarders as it reduces the hoarder’s reserves and lowers its chances of survival and potentially future reproductive output (Galvez et al, 2009). The potential loss of a larder hoarder’s single cache represents a much more severe loss of resources than does the loss of one or a few caches belonging to a scatter hoarder (Vander Wall and Jenkins, 2003).

Chipmunks are a classic example of food hoarders, utilizing their characteristic extendable cheeks to collect food to then transport and stash it in their burrow. The eastern chipmunk (Tamias striatus) exhibits both methods of food hoarding. However scatter hoarding is far less common and occurs over short periods of time in the spring (Clarke and Kramer 1994;

Pyare et al 1993). Individuals collect seeds from maple, beech and oak trees; acorns (oak tree

9 seeds) are the preferred food source in the fall in preparation for winter (Elliott 1978; Lacher and

Mares 1996; Pyare et al 1993).

The eastern chipmunk is a solitarily animal living in a single burrow with a breeding season in the spring and occasionally in the fall (Elliott 1978; Pyare et al 1993). Males emerge from their burrows in February waiting for females to emerge and enter estrous. Males compete in aggressive encounters with other males for the chance to mate with a female (Elliot 1879;

Getty 1981; Yahner 1978). Young are reared in the mother’s burrow for about 51 days, after which they disperse and try to survive on their own, inhabiting pre-existing vacant burrows.

After establishment in their new shelter they spend the majority of their time (approximately

75%) within 25 m of the burrow entrance (Elliott 1978). Being located in a resource-rich area throughout the summer increases access to food which in turn aids in winter survival. Home ranges of individual chipmunks overlap and they display tolerance of neighbouring conspecifics.

When neighbours do encounter one another they rarely behave aggressively but sometimes fights do occur. Chipmunks usually move in ways that create space between themselves and other individuals, minimizing contact, and agonistic displays are rare (Elliot 1978; Getty 1981; Ickes

1974; Yahner 1978). This spacing behavior limits population density and access to space during establishment of the juveniles in the summer (Getty 1981). This is reflected in the difference of winter sleep initiation periods between adults and juveniles in chipmunks. Juveniles enter winter sleep a few days later than adults, which is believed to be caused by the extra needed time to obtain a suitable cache and build body fat (Kawamichi 1996).

The completion of food hoarding for chipmunks is a large factor in inducing winter sleep

(entering their burrow where they reduce activity and consume cached items as required), in in addition to attainment of sufficient body mass and the completion of offspring rearing

(Kawamichi 1996). Winter mortality is high. Thus preparation is critical to increase fitness.

Chipmunks will not enter winter sleep unless they have enough food resources stored to ensure they have access to enough energy to survive the entire winter. Once an individual has invested so much time and effort into building their cache, protective behaviours will heighten , to make certain that their food storage is secure. The energetic cost of collecting this stash is great, as much time is spent collecting the food, predation avoidance is energetically draining and locating enough seeds may be challenging if the area is not rich in resources (Kawamichi 1996).

Consequently, loss of this stash would be very costly. The construction of a food stash for winter survival is directly linked to the fitness of the individual, thus amplified aggressive protection of the food is justified, even in non-breeding seasons when agonistic behaviours are less common.

The relationship between aggression and access to space is complex as it involves long- term establishment patterns and interactions with neighbouring individuals (Getty 1981). Solitary animals have limited interaction with other individuals during their juvenile development, except with their mothers. As they grow and become more independent juveniles become exposed to their neighbours and then as they leave their natal nest they become fully emerged in social interactions with other individuals (Loew 1999). It is at this time that their defensive instincts and behaviours develop with little guidance from their parents (Elliot 1978; Loew, 1999).

Previous studies have found that the eastern chipmunk spend only a small fraction of their time (~1.5 – 2.3% of the total daylight time budget) in defensive or agonistic encounters

(Getty 1981, Yahner 1978), and spend half of their time eating or foraging (Getty 1981; Ickes

1974; Yahner 1978). These studies, however, have looked at the time budgets of the animals over seasons and years, yet did not address the question of whether time budgets are stable throughout the season. As both food availability and the time available for acquisition of

11 sufficient resources diminish as the season progresses, it is possible that competition may become more intense and chipmunks may devote a greater proportion of time to the defense of resources (or other agonistic interactions).

The purpose of this study is to understand how the eastern chipmunks spend their time preparing for winter. I s there a significant difference in the behavioural time budget of the eastern chipmunk directly before entering their burrows for the winter in comparison to the rest of the year? I hypothesize that chipmunks dedicate most of their time in the fall to ensuring winter survival by building and protecting their food cache. I predict that, while they are out of their burrow, during this time period chipmunks spend most of their time gathering food and then secondarily, protecting that food cache.

Methods

Animals

The study area was in the Ganaraksa Region Conversation Authority, near Pontypool,

Ontario. Two grids were constructed with intervals of 15 meters between markers, the first grid covered a 1.82 km2 area (44.53906"N, 78.321828"W) and the second grid covered an area of

3.24 km2 (44.53238"N, 78.322124"W). The forest is primarily dominated by oak (Quercus spp) and sugar maple (Acer sarccharum) trees with a small beech (Fagus grandifolia) population, and is ideal chipmunk habitat.

Trapping commenced in early June, 2010, and continued every other week until the end of October using Longworth traps. Traps were baited with sunflower seeds set before sunrise

(04:30-06:00) and checked within 4 hours of being set. Captured individuals were identified using Hauptner ear tags (administered the first time they were caught) and their weight, sex, and sexual condition were recorded.

Radio Telemetry

In September, 21 radio collars were placed on adult chipmunks (11 male, 10 female).

Collars were given a unique colour combination of heat shrink tubing (on the collar itself) to allow visual identification. The goal of this was to be able to track and observe natural behavior and determine burrow sites of a subsample of the population. If drastic behavioral changes were to occur, for example abandonment of burrow and cache and extreme relocation, the animals could still be tracked and observed. Of the 21 collared individuals, 15 collars were lost so that complete tracking data were obtained for only 6 specimens (4 male, 2 female).

Using radiotelemetry, each collared individual was tracked at 1.5 to 2 hour intervals between 0800 and 1800 to observe their behaviour. Researchers tried to stay as far away from the animal as possible to reduce bias in the subject’s behaviour. Visual confirmations of the animal were difficult as they have excellent camouflage and run very fast. This tracking data was used to determine the burrow location before the behavioural observations began, to assess changes in the daily movement during the behavioural experiment and to confirm the location of animals during the behavioural observations.

Behavioural Observations

Behavioural data collection on the 6 remaining collared individuals began in October, when records indicate that the chipmunks are likely to enter their burrow for the winter season

(Elliott 1978) In the first week of data collection a 9 inch pie plate (filled with peanut butter, sunflower seeds, hamster mix, granola mix and dried corn) was placed within 2 meters from the burrow of each of the individuals. Game cameras (Scout Guard, Model: SG560V) recorded the interactions of the chipmunks around the trays for 3 days (October 5-7, referred to as “Early

October”). Cameras were motion-activated and set to record for one minute durations, with 1

13 second between successive videos. Trays were refilled between 0730 and 0830 every morning during the test. Using radiotelemetry, behavioural data were collected for each day the tray was available and the day that it was removed to confirm the activity of the individual when not on camera. Behavioural tests were conducted again two weeks later (Oct 18-22, referred to as “Late

October”) employing trays and game cameras as well as direct observation of activity.

Observations after another two week interval indicating that all animals involved in this study were not active and the data collection period was over.

For analysis of video data, focal animals were identified by their coloured collars.

Observed behaviour was classified into four categories: aggressive, eating/collecting, grooming and other. Descriptions of these behavioural categories are presented in Table 1. The total number of appearances in the videos was recorded as well. To determine the behavioural time budget for each individual, the number of videos in each category (each being 1 minute long) was divided by the total length of day light hours (Early October= 11:28, Late October= 10:47) to obtain percentages for each behavioural category. Burrow residency was determined by calculating the difference in the total day light and above ground activity; radiotelemetry data was used to confirm that while off camera the individuals were in their burrows.

Statistical Analysis

Due to the small sample size, permutation tests were conducted to test for differences in the behavioural time budgets (proportion of time spent in each behavioural category) for each interval of observation. I randomly shuffled and re-sampled (without replacement) the observed values for each of the four behavioural categories within individuals. I then used an analysis of variance (ANOVA) to test for differences in the percentage of time animals were aggressive, eating or collecting food, grooming or other activities, with a null hypothesis that there was no

14 difference between behaviours in each week. I then calculated the difference between the randomized value set and the observed data. This was completed 10000 times and I counted the number of times that the absolute value of the permuted difference exceeded the observed difference. This number divided by 10000 represents the equivalent of a p-value in standard statistical tests. As there is no test statistic, I simply report this p-value. For post-hoc analysis

Student’s t-tests were calculated for each possible behavioural category pairing. Using the previously generated randomized data set Student’s t-tests were calculated and compared to the observed data; this was repeated 10000 times and a p-value was then generated. The null hypothesis for this test was that there would be no difference in the amount of time an individual displays a specific behaviour. A similar test was conducted to determine the change in the duration of display behaviour from Early October to Late October with a null hypothesis that the amount of time the behaviour is displayed will not change between weeks.

A student t-test was performed on the number of sightings for each individual to evaluate activity levels of the chipmunks between the two intervals. The null hypothesis is that the animals will be as active in Late October as Early October. A permutation test was also used by randomly shuffling (without replacement) the observed data, performing a student t-test with the randomized data and comparing that to the observed data. This was completed 10000 times and a p-value was then generated as mentioned previously. Randomizations were carried out in

Microsoft Excel using the PopTools add-in. All tests were two-tailed.

Results

Early October Behavioural Analysis

The activity of chipmunks significantly decreased from 346 minutes in Early October to 32 minutes in Late October (Permutation test: p = 0.0134, Figure 1). During the week of highest

15 activity, Early October, time budgets differed significantly among individuals (Permutation test: p<0.01, Figure 2). Figure 2a highlights the large difference in the above and below ground activity. Overall, individuals spent a significantly greater proportion of time in their burrows than undertaking any other observed behaviour (post-hoc pair-wise permutation comparisons: aggressive Permutation test: p=0.0022, eating/gathering Permutation test: p=0.002, grooming

Permutation test: p=0.0025 and other Permutation test: p=0.0023). Individuals spent significantly greater proportions of time above ground eating/gathering as opposed to either aggressive behaviour or grooming (Permutation test: p=0.0433) and grooming (Permutation test: p=0.0224) (Figure 2b).

Late October Behavioural Analysis

During Late October there remained a significant difference between the above and below ground activity (Permutation test: p<0.001) (Figure 2). Burrow residency was significantly different than aggressive behaviour (Permutation test: p=0.0025), eating/gathering

(Permutation test: p=0.0025), grooming (Permutation test: p=0.0019) and other behaviours

(Permutation test: p=0.0027) as displayed in Figure 2a. Among above ground activity there were no significant differences across the four behavioural categories. Eating/gathering had the highest range (0-1.7%) but this was not significantly different than the rest of the behaviours

(Figure 3b).

Alteration in Behaviour over Time

The behavioural budgets changed over time as the animals were preparing for winter hibernation (Figure 4). The amount of time chipmunks spent in their burrow increased from

Early October to Late October and all of the surface activities/behaviours deceased. The magnitude of change (from Early October to Late October) was significantly different among the

16 behaviour categories (Permutation test: p<0.001). The increase in burrow residency differs from the decrease in aggressive behaviours (Permutation test: p=0.0018), eating/gathering

(Permutation test: p=0.0454), grooming (Permutation test: p=0.0021) and other behaviours

(Permutation test: p=0.0013). As the chipmunks were getting near to the end of the season they began spending more time in their burrows and less time above ground.

Discussion

The goal of this study was to determine the behavioural time budget of chipmunks as they prepare for winter sleep and determine if there is a change over time. By conducting this research, significant insight into how these animals shift focus to winter survival have been obtained. The video footage collected in Early October (Oct 5-8) captured significantly more activity than in Late October (Oct 18-21) (Figure 1), suggesting that the timing of the study was appropriate for studying the animal as they are preparing to induce winter sleep.

Early October was the interval of the highest activity. Amoung all four behaviours displayed above ground, eating/gathering was displayed significantly more often than aggressive behaviours and grooming (Figure 2). This supports the prediction that eating/collecting would be the most consistently displayed behaviour. However I also predicted that aggressive behaviours would account for the second greatest proportion of above ground activity, and that the proportion of time spent on aggressive behaviours would be significantly greater than remaining classes of above ground activity and it was not. During this time period the chipmunks spent most of their time in their burrows, underground.

Burrow residency was the dominant behaviour in Late October (Figure 3). During this week of observation half of the recorded animals did not have any above ground activity, implying that these animals had entered the burrow for winter sleep. The active surface animals

17 focused their time budgets on ‘Eating/Collecting’ food, as expected. However there were no significant differences with the rest of the behavioural categories. As the animals were nearing winter hibernation I predicted that there would be a heightened aggressive response in association with cache protection. As the time window for food collection was becoming smaller the risk of pillaging would increase by those animals with a small or no cache. Thus more aggressive forms of cache protection would be displayed. This was not displayed, yet the amount of burrow residency may be an indication of cache protection as the individuals are guarding that cache by never leaving it. Alternatively, the time spent on aggressive behaviour did not increase as potential cache robbing pressure may decrease with the decreasing surface activity of conspecifics.

When studying the change of behaviours between weeks there is no significant difference amoung the surface behavioural categories, but there was an increase in burrow residency

(Figure 4). It was predicted that the amount of time eating and collecting food would increase, as would aggression, and grooming and other behaviours would decrease; the degree of change would then be drastic enough to be significantly different. This prediction was not supported by the data. Overall the time allotments of each of the surface behaviours remained similar between the two weeks but their overall time budgets, when burrow residency is included, indicate that animals were spending more time underground. Thus the change in time budgets from Early

October and Late October indicate the onset of winter hibernation.

In previous studies during the non-breeding season, aggression was observed to occupy only 1.5 % (Getty 1981) or 2.3% (Yahner 1978) of the eastern chipmunks daily time budget.

These values are much lower than the findings of this paper (23.79% of above ground activity in

Early October and 5.2 % in Late October). This may be due to the way the time budget was

18 constructed as this study only takes into consideration the documented behaviours above ground and the previous studies both included time spent in the day in the individual’s burrow. This may also be explained by a potential shift in focus to gathering food and protecting it directly before inducing winter sleep.

The observation season in this study occurred during a mast year for the oak trees in the area. Mast years, years in which the oak trees produce seeds (acorns), occur only every couple of years and are associated with several measures of abundance and canopy cover (Schroeder and

Vangilder 1997). 2010 was a mast year. Therefore acorns, a preferred food item, are available in abundance and finding them is extremely easy relative to non-mast years, even for inexperienced juveniles. This therefore has resulted in a time budget break down during a favorable year in which there is no food limitation and minimized the effects of artificially feeding them with the trays. This is not always the case for some animals and in some years, food items can be hard to locate and the construction of their cache is limited. In these circumstances, pillaging is to be expected. A time budget analysis of the eastern chipmunk during a food deficient year would be an interesting comparison study as I would speculate that the behavioural time budgets would be significantly different and likely support the predictions of this study.

Study of the animals stress levels as they prepare for hibernation in both mast year and non-mast years would be a good accompaniment to this study. The impacts of food availability on overall survival of the individuals in the following spring and the possible effects it would have the spring mating season are warranted as these studies would extend this study by elaborating on the importance of maintaining an appropriate time budget (i.e. allocation of effort to foraging versus defense, etc.) to ensure winter survival and wellbeing.

Also an area of future development would be in the use of more advanced camera systems. This may allow the separation of eating (immediately consuming the food) and collection (for later consumption) while the animals are putting food in their mouths. This would then strengthen the relationship between the time spent gathering food and their survival. In this study I look at the amount of time dedicated to obtaining nutritional requirements and I cannot separate it further into short term (eating) or long term (collecting) requirements. These findings indicate that acquiring food to meet requirements of long term survival is the main focus of the eastern chipmunks’ time budget while outside of their burrow.

In conclusion, the eastern chipmunks involved in this study displayed a focus of their time budget on eating and collecting food in preparation of winter survival while they were above ground. Surface activity decreased from Early October to Late October as the animals spent more time in their burrows. This burrow residency would aid in predator avoidance and provide a high level of cache protection. These behaviours aid in the winter survival of the animal, which is the largest environmental challenge posed to these animals each year.

Table 1: Categories of eastern chipmunk (Tamias striatus) behaviour and their description (Yahner, 1978).

Behavioural category Description

any combative behaviour with another individual, includes chasing, being Aggressive chased, withdraw and fight

Handling of food item with forepaws and consumption or placing into mouth or Eating/Gathering check pouches Grooming licking or rubbing the body with forepaws Other behaviours unlike the categories such as running, climbing and sitting still Burrow time spent, off camera, underground in burrow

Amount of Above Ground Activity (minutes) Activity Ground Above of Amount

Time Period

Figure 1: The average time of above ground activity for all individuals (Tamias striatus) over a two day observation period in both weeks of observations (Early October, Oct 5-8 and Late October, Oct 18-21, 2011), standard deviations displayed (n=6).

Percentage of Daily Time Budget Time Daily of Percentage

Behavioural Categories

Behavioural Time Budget Time Behavioural

Percentage of Above Ground Above of Percentage

Behavioural Categories

Figure 2: The percentages of the daily time budget of the eastern chipmunk (Tamias striatus) in Ganaraska Forest, Ontario Canada, for above and below ground activity (a) and above ground (b) for each studied behaviour in Week 1 (Oct 5-8, 2011), standard deviations displayed (n=6).

Percentage of Daily Time Budget Time Daily of Percentage

Behavioural Categories

Percentage of Above Ground Behavioural Time Budget Time Behavioural Ground Above of Percentage

Behavioural Categories Figure 3: The percentages of the daily time budget of the eastern chipmunk (Tamias striatus) in Ganaraska Forest, Ontario Canada, for above and below ground activity (a) and above ground (b) for each studied behaviour in Week 2 (Oct 18-21, 2011), standard deviations displayed (n=6).

Change in Percentage of Daily Time Budget Time ofDaily Percentage in Change

Behavioural Categories

Figure 4: The change in displayed behaviour of the eastern chipmunk in the Ganaraska Forest, Ontario Canada, from Week 1 (Oct 5-8) to Week 2 (Oct 18-21) (Week 2-Week 1) with standard deviations displayed (n=6).

Chapter 3 Mediators of signalling during aggressive encounters in the primitive acoustic insect Cyphoderris monstrosa (Orthoptera: Haglidae)

Introduction Animal contests: evolution

Game theory has been extensively used in evolutionary biology to examine the outcome of animal contests. During a contest each opponent has several behavioural options available to display. The strategies in behaviours can be analyzed for their costs and benefits with respect to contest outcome (Reichert and Hammerstein 1983). Two categories of aggressive contests and their associated models have been defined: mutual assessment or self-assessment, based on whether contestants use information obtained during a contest to measure their opponent’s or their own ability (usually referred to as resource-holding potential, or RHP – an indicator of an individual’s probability of winning a contest) (Parker 1974). In both categories participants are forced to a conclusion wherein one participant is the winner and one is the loser.

Contests can also be classified as either symmetrical or asymmetrical. A symmetrical contest involves contestants whom are evenly matched, with a winner being the one who fights longer.

In asymmetrical contests the winner is dependent on the nature and strength of the asymmetry

(Maynard Smith and Price 1973, Maynard Smith and Parker 1976). Asymmetries can then be divided further into terms of RHP, in which the individuals assess the differences in fighting abilities, or payoff asymmetry, in which the differences in the consequences of winning or losing the contest for each contestant are assessed (Haley 1994).

In mutual assessment models, contests that involve individuals with little differences between them will likely last longer as the participants require more time to understand the differences in each other’s RHP. Contest dynamics are based on the dominant asymmetry; if this is difficult to perceive then the contest will continue longer (Enquist and Leimer 1987; Taylor et

26 al. 2001). There are general rules when determining the use of tactics in an aggressive encounter, often suggesting an escalation sequence from low cost to high cost tactics (Hack 1997).

Costsassociated with aggressive encounters include injury risks and energy expense (Hack

1997). In a competitive encounter the contestants assess correlates of RHP (body size and weaponry) and the first one to give up is the loser (Briffa 2008). The end of a contest occurs when the weaker participant (low RHP) strategically decides to withdraw and the winner therefore demonstrates (honestly or not) a relatively high RHP (Briffa and Sneddon 2007;

Humphries et al 2006).

Animal contests: energetics

Contests for mates and territory are highly influenced by the energetic costs associated with the displayed behaviour (Davies and Houston 1984). Contests are energetically costly and longer contests consume more energy. Energetic expenditure is an important factor shaping the contest strategies used (Hack 1997). In aggressive encounters, losers may be physically exhausted, have lower energy reserves, and spend more time eating to replenish their energy storage (Hack 1997).

Study species

The family Haglidae dates back to the Triassic period and are the primitive ancestor to the

Tettigoniidae and Gryllidae; it has seven extant species in four genera (Mason 1994). In China there are two species of the genus Aboiloilomimus: Aboiloilomimus guizhouensis and

Aboiloilomimus ornatus (Liu et al 2009). In North America there is one genus with three species present, the species include: Cyphoderris monstrosa, Cyphoderris buckelli, and Cyphoderris strepitans. Cyphoderris monstrosa and Cyphoderris buckelli are located in western Canada and

Cyphoderris strepitans is located in the western United States. The remaining two genuera and species Prophalangopsis obscura and Paracyphoderris erebeus are located in Asia (Dodson et

27 al., 1983). The communication, competition and mating ecology of the Tettidgoniidae and

Gryllidae have been the subject of much study. Cyphoderris, however, have not received as much research attention and the findings of work so far suggest that Cyphoderris combine some characteristics of the acoustic communication system of both groups, which, along with their phylogenetic relationship, suggests they are primitive in some respects (Mason 1994).

Cyphoderris are active at much colder temperatures (as low as 2° Celsius) than is typical for acoustic insects. Endemic to mountainous regions of western North America, they start emerging in late May or early June and sing when evening temperatures are just above freezing. Individual males establish a singing perch shortly after dusk; they then begin singing and do so for several hours (Mason 1994). The breeding season begins at first emergence of the males and females (in late May or early June) and nocturnal activity is present until the end of the summer (August)

(Mason 1994). The three species are all found in the mountainous ranges of Canada and the

United States however each species has a unique destination for perching, feeding and mating.

The distributions of the three North American species are found in Figure 5 (Morris and

Gwynne, 1978).

Cyphoderris buckelli are located in the Dry Forest and Columbia Forest of southern British

Columbia. They feed on the flowers of understory plants such as the service berry (Amelanchier sp.), arrow-leaf balsam-root (Balsamorhiza) and the tall Oregon grape (Berberis). Their range is more restricted and they have a different ecological niche than the Cyphoderris monstrosa and in areas of population overlap they do not compete for food resources (Morris and Gwynne, 1978).

Cyphoderris strepitans is similar in morphology and calling song to Cyphoderris buckelli but the males have a defining structure of their terminalia, who protrudes more than the other species. C. strepitans is predominately a sagebrush species, although they are found in open

28 forest habitats near sagebrush prairies. Their range is restricted to the mountainous areas of

Colorado and Wyoming.

Cyphoderris monstrosa has undergone the most extensive morphological and behavioural divergence of all of the species (Kumala et al., 2005). They have the most extensive range

(Morris and Gwynne, 1978), are found in dry forests in southern British Columbia and are typically located in the subalpine Forest of the Kananaskis Valley in southern Alberta. They range from the Canadian Rockies, of southwestern Alberta, east through southern British

Columbia, and north of the Canadian/United States Border by 700 miles and south almost as far as Northern California. They can also be found in western Montana and central Idaho. They feed on the cones of lodge pole pines (Pinus contorta) and western white spruce (Picea glauca), which is a characteristic species of the habitat area (Mason 1994). As a means of obtaining the cones the animals climb the trees at dusk and feed at night, presumably emerging from the leaf litter of the forest floor where they presumably reside throughout the day (Morris and Gwynne,

1978). However localization of individuals during daylight hours remains unspecified.

In each of the species both sexes are flightless and the population has fewer males than females (Dodson et al., 1983). Males produce a sustained acoustic call by tegminal striduation, as in other ensiferan insects (Dodson et al., 1983, Morris and Gwynne, 1978). Females are apterous.

Males produce song to attract females (Snedden and Sakaluk, 1992). Calling begins in the late evening and continues past midnight. Cyphoderris strepitans and Cyphoderris buckelli call from low shrubs and C. monstrosa climb up the tree trunks and call from there (Mason 1994). Females mount males and start to eat the fleshy hind wings, males use their pinching organ (gin trap) to hold the female to him and then the genitilia deposit a spermatophore (a gelatinous mass transferred from the male after copulation) to the female. Once copulation has ended, the female

29 leaves the area and consumes the spermatophore (Dodson et al., 1983). Because the spermatophore is thought to represent as a gift filled with nutrients, the males are considered to have a dual investment in mating and reproduction (i.e. wing and spermatophore feeding)

(Dodson et al., 1983).

Nuptial gifts function as mating effort if the gift increases paternity by allowing males to transfer more sperm or substances that induce refractoriness in females. They can also function as parental effort if the gift directly increases the fitness or number of offspring sired by the male

(Gwynne, 2008). Virgin males, with wings intact and fleshy, have a mating advantage as their gift is more appealing to the females (Snedden 1996). In a study to determine the role of the hind wings in reproduction, males without fleshy hind wings were mounted as quickly as males with their hind wings intact but were less likely to transfer the spermatophore (Eggert and Sakaluk,

1994). Males without the hind wings were less successful in mating and had increased amounts of injuries to the rest of their bodies, illustrating the importance of the nuptial gift (Eggert and

Sakaluk, 1994).

Non-virgin males are also at a disadvantage as they call less because of their depleted energy storage after mating (Sakaluk et al., 1987). After a night of calling males lose weight; this may explain why they do not call every night (Dodson et al., 1983). Non-virgins also have shorter calling durations, and calling duration decreases as the season progresses (Sakaluk and Snedden,

1990). Females should choose virgins as they have more resources to offer (Johnson et al., 1999) and have more energy to put into calling (Sakaluk et al., 1987).

Singing success and reproductive success are closely related in acoustic insects. Singing in acoustic orthoptera is the most energetically demanding behavioural activity (Prestwich 1994).

But it has been shown for a number of species that a male’s probability of attracting a mate is

30 related to his calling effort. Males are therefore under selection to optimize the effectiveness of their calls (Bennet-Clark, 1998).

In C. monstrosa, the function of calling is not only for mate attraction but it is also important for male competition. The call is an irregularly interrupted trill, so that different males vary in the amount of sound they produce (short chirps, long pauses; long chirps, short pauses, etc.). Once the male has fought off other males on the tree he then calls for the duration of the evening’s active period (or until a female arrives). When there are two males present on the same tree they typically engage in a contest involving a (sometimes prolonged) period of competitive acoustic signalling (singing) which may escalate to overt fighting. The winner of the contest can be predicted by the amount of singing by each male; the male with the higher duty cycle (proportion of time singing) is the winning male (Mason 1996). Males become aggressive when another male is located near them. When males encounter one another they typically start to call at one another. The weaker caller typically gives up and leaves after a period of competitive singing. However, when opponents are more evenly matched in terms of duty cycle

(game theory: symmetric contests, Reichert and Hammerstein 1983) then an escalated fight

(involving physical contact) begins. A fight usually results in the losing male retreating off of the tree entirely and the winner remaining on the tree (usually perched on one of the lower limbs)

(Mason 1994).

In many species, when contests occur between individuals they mutually assess the RHP of their opponent and this is often directly related to size and weaponry (Hack 1997). However, in a previous study of aggressive interactions in C. monstrosa, size had no influence in the outcome of the fights and there were no obvious morphological correlates with competitive abilities (Mason 1996). Butterflies have no weapons and with limited physical contact the

‘superior competitor’ is the one who’s RHP is high and contests are settled by persistence (Kemp and Wiklund 2001). In Gryllus bimaculatus lighter animals won 30% of fights (Hofman and

Schildberger 2001). This may be the similar to C. monstrosa. Large body size may help a competitor overpowerhis opponent; large or strong weaponry may help inflict injury and high stamina (measures of energy reserves and aerobic capacity) may sustain longer periods of competition and injuries avoidance (Batchelor and Briffa 2010). Physiological variables are important indicators of RHP and agonistic signals are influenced by the individual’s energetic constraints (Briffa and Snedden 2007).

In this study, I examined the energetics of competitive acoustic signalling in C. monstrosa. The relative competitive ability (RHP) of males was assessed using a tournament design, in which pairs of males competed for a singing perch. Then, for a sample of these males,

I estimated the costs of signalling by measuring their resting and singing metabolic rates (as CO2 production). I tested the hypothesis that successful males, those who were able to maintain high singing duty cycles in contests, were those with lower metabolic rates during singing.

Methods

Animals

Cyphyoderris monstrosa males and females were collected in an area outside of

Kamloops, British Columbia (N50.75462°, W120.12046°) in May 2011 (at night between 2100 and 0100 hours) and transported to the University of Toronto Scarborough campus. The males were marked with identification tags glued to their pronota and housed in an environmental chamber (15C, 15:9 light:dark cycle) in individual plastic cages with a piece of cardboard egg carton for shelter. Animals were fed every day with a piece of apple, a kitten chow kibble and

32 bee pollen and were supplied with a cotton plugged water vial for moisture. Males used in this study were virgins, as indicated by their intact and undamaged hindwings.

Fighting Tournament

Fighting trials were held from June 24-July 11 from 0930-1500. All males fought 3 rounds in the tournament. Opponents were randomly assigned with the constraint that there was an equal distribution of the different possible pairings of winners and losers in the second and third rounds

(i.e., winners:winners, winners:losers, losers:losers). Thirty-two males started in the first round of the tournament and 26 males completed all three rounds of the tournament (the remaining 8 males died during the study).

Contests were staged in an arena made from a section of spruce (Picea sp.) trunk approximately 46 cm long. This log was placed, standing on end, on a rotatable platform (lazy susan) to allow observation and video recording of all aspects, with minimal disturbance to the animals. Males released onto the log would typically take up singing perches on the log and engage in territorial contests to defend it against intrusion by other males.

All males were brought up to the room with the staged fighting stump at the onset of scotophase (0900) after being fed and allowed to acclimate for half an hour. Males had to be singing for a minimum of 3 minutes to be considered ready for a fight trial. Two singing males were randomly selected and placed at the bottom of the fighting log for them to voluntarily leave their plastic cages and participate in the contest. Of the males fighting, the number of previous contests for each male was controlled to be equal. Males had to participate in at least one interaction for the competition to be classified as a fight. An interaction was defined as an orientation to, and approach of one male to the other with or without physical contact. Typically, an interaction involved an approach (one of the contestants walks towards the other), which may

33 have led to physical contact and an escalated fight (antennation, biting grappling, kicking), or a chase in which one male retreats without physical contact. A trial ended when one of the males exited the log, whether forcibly (i.e. ejected during escalated combat – see below) or voluntarily

(following one or more interactions). The winner was deemed to be the animal still on the log at the end of the trial and the loser was the one who quit the log. See Mason (1996) for a detailed description of C. monstrosa territorial contests.

I recorded the following data from these territorial contests.

1) Status- based on their record in three contests, each male was classified as either a winner

(winner of 2 or more contests) or loser (loser of 2 or more contests).

2) Approaches (an interaction without physical contact) by each male towards the other.

3) Initiation of physical contact.

4) Approach wins – the male that chases his opponent away from an interaction without

physical contact.

5) Bout wins – the male that forces his opponent to retreat from an interaction with physical

contact.

Energetic Rate Measurements

I measured the metabolic rates (as CO2 production) of males that had participated in fight trials. All metabolic rate measurements were conducted under red light, in the same room as the fight trials, from July 18-Sept 7. For metabolic measurements, individual males were enclosed in a chamber that included a speaker (for playback of synthetic song – see below) and a microphone

(for recording of the male’s song). Metabolic measurements were carried out for animals in three different behavioural states: resting, disturbed resting and singing. A resting trial was defined as one in which the animal was making no sound and not moving. Disturbed resting was similar to

34 resting, but with an artificial call playing in the chamber. A singing trial was defined as a trial while the males were singing with an artificial song playing in the background. For each trial the weight of the animal was recorded using an (Ohaus Explorer) electronic balance.

All metabolic rates are reported in mLCO2/min. Energetic measurements were obtained via stop-flow respirometry using a Qubit systems (Ontario, Canada) 8 channel gas controller (G245), a Qubit systems 8 channel gas switcher (GS244) and a Li-Cor (Nebraska, USA) CO2 analyzer

(LI-6252). Two different chambers were utilized for the resting and disturbed resting trials, one was a chamber made of metal and plexiglass (volume of 207 mL) and the other was a round

Ziplock container (dimensions) with a screw on lid sealed with Teflon tape. For the singing trials the Ziplock chamber was the only one used. CO2 was removed from incoming air by filtering the air through soda lime. Flow rate through the chambers was set to 300 mL/min and did not appear to disturb the animals. Air from the test chamber was passed through a magnesium perchlorate filter to remove moisture then passed into the CO2 analyzer which measured respired CO2, in mL/min.

Dwells (i.e. duration of continuous measurement on each respirometer channel) lasted from

300 (resting and disturbed resting trials) to 600 (singing trials) seconds. For use in statistical analyses the last 50 seconds of the trial were used for the CO2 measurements in the resting and disturbed resting trials. The 50 seconds of CO2 data used for the singing trial were at a time when the animals were constantly singing and were near the end of the trial well after the chambers had been flushed of residual air from the previous trial.

For song playback to males during respirometry, artificial C. monstrosa songs were synthesized using a Tucker-Davis Technologies DSP (RL2, TDT System 3). Playback stimuli consisted of synthetic trills of 12 kHz sound pulses, 8 ms duration with 1ms rise/fall times. Trill

35 durations and inter-trill intervals were variable (Gaussian distributions: mean±sd duration

1±0.2s; interval 0.5±0.2s). For recording the acoustic output of males during metabolic measurements, chambers were fitted with a microphone (Knowles, NR-23158) connected to a digital recorder (Sound Devices 722) via a custom-built pre-amplifier.

Song and Physical Characteristics

Only a subset of males (n=16) completed the Fighting Tournament and had the full data set for the energetic rates. These individuals were dissected to obtain morphological measurements and their recorded calls were analyzed. A recording of each males call was obtained while they were in the chamber for the singing trial; this recording was analyzed using a MatLab script

(Mathworks, Natick MA) to obtain a measurement of the proportion of time males spent producing sound (duty cycle). Duty cycles were measured as the total duration of continuous (i.e. no pauses greater than 50 ms) sound production divided by the total duration of the measured recording. The sound clips were used from the same period that the CO2 measurements were recorded in the singing trial to ensure overlap with the energetic data.

Physical properties of each male were collected lastly for this subset of males. Males were dissected and photographed using Act1 (Nikon). A number of features related to overall body size (hind femur length (FL), pronotum width (P), head width (HW), maxilla span); two structures used in sound production (stridulatory file length (FI) and harp area(HA)); and the width of the gin-trap (a structure used during copulation (GTW)) were measured. Measurements were taken using the measurement tool in ImageJ and were recorded in millimeter and mm3.

These measurements were compared for repeatability with a subset carried out independently.

Statistical Analysis

For experience effects in contest outcomes were tested for using a Fisher’s Exact test to compare the number of winners and losers with prior winning and losing experience. A principal components analysis was applied to the morphometric measurements to reduce the data to a smaller number of uncorrelated variables. For this analysis, bilateral measurements (e.g. right and left femur length) were averaged to give one value per male. The first three principal components accounted for 96% of the variance in morphometric measurements and these were used in further analysis (Table 2). Then a logistic regression was performed with male status

(winner/loser) as the dependent variable and the first three principal components as predictors.

In a separate analysis, the effect of tactics within contests on the contest outcomes was examined. With individual contests as the unit of analysis, then a logistic regression was performed with contest outcome as the dependent variable and male tactics (approach, initiation, approach win, bout win – see above) as predictors. Previous work (Mason 1996) has demonstrated the singing duty cycle is a predictor of success in Cyphoderris monstrosa territorial contests. Possible physiological correlates of variation in singing among males, as well as physiological traits that might distinguish winning and losing males, was then the focus. Using regression analysis the metabolic data was examined to determine the effects of body weight and singing activity on metabolic rate. Then the factorial metabolic scope was calculated (ratio of active to resting metabolic rate). Since active metabolic rate varies with duty cycle (ie amount of singing), winners and losers were compared while controlling for differences in duty cycle using an analysis of covariance.

Results

Tournament Results

All males in the data set (n=16) completed three fights in the tournament. The tournament was used to determine the classification of the fighting success status (Status), based on the number of contest wins. Males that won two or more fights were classified as Winners (n=9), while those winning fewer than two fights were classified as Losers (n=7). There was no evidence of an experience effect in contest outcomes (Fisher’s exact two-tailed test p > 0.9).

There were, however, differences in the behaviour (tactics) of males in contests that were correlated with contest outcome. Then a logistic regression of contest outcome was performed with all variables, reduced the model by removing non-significant variables and compared model fits using the Akaike Information Criterion (AIC). The best fit model (Table 3) included Initiate,

Approach, Approach Win and Bout Win, but only Approach Win and Bout Win were significant.

This indicates that Winners were significantly more likely to force their opponent to retreat in individual encounters (with or without physical contact) during a contest.

Morphology Comparisons

Principal components analysis yielded six uncorrelated components, the first 3 of which accounted for ~96% of the variance in the data (Table 2). The first principal component (PC1) was loaded most heavily on song-related structures (harp area). PC2 was mainly a measure of size (femur length and pronotum). PC3 was mainly a contrast between femur length and pronotum width, suggesting a measure of leg length relative to body size (gangliness). Logistic regression of male status on PC1-3 showed no effect of male morphology on territorial contests

(Table 4).

Metabolic Measurements

There was no relationship between metabolic rate (measured as CO2 production) and body weight (R2 = 0.0147, p = 0.89, Figure 6). Therefore metabolic rate data were not weight- corrected in further analyses.

There was no significant difference in the metabolic rates of winners and losers (Welch two sample t-test, t = 1.4671, df=6.94, p = 0.19). Also, as expected given the energetically demanding nature of sound production in orthopteran insects (Prestwich and Walker 1981) there was a strong relationship between duty cycle and metabolic rate (R^2 = 0.65, p < 0.0002, Figure

7). We compared the residuals of this regression for winners and losers, but found no significant difference in the metabolic cost of singing between males of different status (Welch two sample t-test, t = 0.974, df=13.579, p = 0.38). Similarly, an analysis of covariance, including Status as a factor showed no significant effect.

Finally, analysis of covariance of factorial metabolic scope, with Status as a factor and

Duty Cycle as a covariate showed a significant interaction between Duty Cycle and Status

(Figure 8). Winners showed a steeper increased in metabolic rate, relative to resting, with increasing Duty Cycle. As there was no difference in resting rates between Winners and Losers, this suggests that Winners are showing a greater increase in metabolic rate during singing than are Losers.

Discussion

Behavioural Patterns

Contest history does not influence future contest outcomes. C. monstrosa did not retain winner-loser effects between trials in the Fight Tournament. Therefore males approached every contest in the tournament with equal determination and motivation This is not always the case

39 for contests as displayed with male jumping spiders (Phidippus clarus) who are more likely to lose a contest after losing in a previous contest or win after winning in the last contest

(Kasumovic et al 2009). Cumulatively, the effects of multiple contests for an individual can also increase or decrease the likeliness of winning a contest (Stuart-Fox et al 2006). However, this was not displayed in the male C. monstrosa involved in this study. Future work may involve the potential of short term experience effects that males might experience over one night when males may have to defend their perch tree multiple times.

There were significant differences in the fighting tactics between the Winners and Losers and these were consistent with previous work on C. monstrosa territorial contests (Mason 1996).

Contests consist of a period of variable duration in which both males are on the disputed territory

(log) and singing. During this on-going singing competition, males engage in a series of direct interactions in which one male (usually the eventual contest loser) approaches his opponent which elicits an attack from the other male. This suggests the male may have some uncertainty in the information gathered in the acoustic contest and needs to test the opponent in another way.

These escalated interactions are resolved when one of the males retreats from the immediate vicinity, with or without physical contact between the contestants.

Morphometrics

Unlike many species that engage in intraspecific contests size and weight do not act as predictors for contest outcomes. Weight differences between the contestants are commonly assessed as an indicator (sometimes in combination with another trait) of the participants RHP

(Batchelor and Briffa 2010; Bridge et al 2000; Clark and Moore 1995; Elias et al 2008). The

Mediterranean field cricket (Gryllus btmaculatus) also shows no relationship between contest outcome and body weight. However there is a significant relationship between contest outcome

40 and antennal fencing (behavioural differences) demonstrating readiness to fight and fight ability through mandible flaring (Hofmann and Schildberger 2001). There are no relationships to physical differences between the Winners and Losers. Therefore individual differences in weaponry do not appear to be significant in these contests. However, Mason (1996) found that males assessed RHP acoustically by the individual’s duty cycle; males with high duty cycles are more likely to win a contest. Male C. monstrosa do not assess their opponents on the basis of fixed physical traits but rather the acoustic abilities (Mason 1996), which may be correlated with physiological condition or energy reserves. It is possible that individual variation in male morphology could be relevant to male mating success in the context of female choice, even though it does not appear to affect male-male competition.

Metabolism

Not surprisingly, high duty cycles (more sustained singing activity) are associated with high metabolic rates (as measured by CO2 consumption rate). As duty cycle increases CO2 production rates correspondingly increase. The previously established relationship between song duty cycle and competitive ability, combined with our results that show no significant effect of variation in male size or morphology, suggests that C. monstrosa territorial contests may be purely energetic competitions (e.g. male duty cycles signal their relative energy reserves or ability to sustain higher energetic costs of signaling). The relationship between metabolic costs and contest results in our data, however, was more complex. There were no overall differences between Winners and Losers in terms of their resting or singing (controlling for duty cycle) metabolic rates. But Winners and Losers differed in factorial metabolic scope

(CO2singing/CO2resting) (Fry 1947). Winners increased metabolic rate (for a given level of singing) by a greater factor over rest than losers. This differs from my initial hypothesis that individuals

41 who sustain high duty cycles are more energetically efficient singers with greater capacities store energy for escalating contests. This may suggest that a male’s ability to rapidly mobilize metabolic resources, rather than differences in males’ absolute levels of energy reserves, may be critical to competitive ability.

As the Haglidae are mountain dwelling species the ability to increase body temperature may be advantageous in the cool nights that they are active in. However, during the experiment the contests were staged at room temperature. It may be possible that this obscured some of the normal pattern because we were effectively helping the Losers to get warmed up. Trilling species increase in body temperature, possibly altering bodily functions (Hoback and Wagner 1997). The variation in calls is great in C. monstrosa from chirp like calls to continuous trill like calls; therefore there may be an effect of the call variation on the body temperature of the individuals.

Hinds and colleges (1993) found that endotherms had an aerobic scope during locomotion that is almost twice of the scope observed in cold conditions, suggesting further that the temperature an individual is exposed to may alter the functioning of normal processes.

The artificial circumstances of the contests may be another source of difficulty in distinguishing differences between Winners and Losers. Our contests were open-ended (ie we didn’t impose a time-limit), but we don’t know whether they were longer or shorter than most of them in nature. As the ambient temperature is colder during a contest in their natural setting, there is reason to believe that natural interactions are usually longer as the animals themselves would be colder and energy mobilization may be slower; Losers may take longer periods of time to be too fatigued to fight, then they did in our contests at room temperature. Also, the dynamics of natural contests are not well-known, but since males emerge each evening and move to calling sites on trees, it is likely that contests would tend to occur early in the evening’s active period

42 when males are establishing their nightly territories. If this is the case, then males would not have had an extended period of calling and the capacity for rapid energy mobilization (and warm-up) may be an important factor. The artificially elevated ambient temperature during out laboratory contest would, therefore, tend to diminish detectable differences among males. But, it is unlikely that this would generate the basic trend displayed in the results. However, it is unlikely that this would alter the basic trend displayed in the results.

I conclude that the contest outcomes are not only predicted by duty cycle but also at the cellular level with the individuals metabolic rates. Winners have a higher metabolic scope, demonstrated by greater rates of increase in metabolic rate with increasing duty cycle than losers.

They are able to mobilize energy at a time of importance during the contest. At the height of the

RHP analysis between contestants Winners are able to strengthen their abilities and deter the

Losers. Although the initial hypothesis was not supported the findings provide another alternative that gives insight into individual variation that has never been explored in this species before.

Table 2: Principal components analysis of male C. monstrosa morphology measurements. Measurements were recorded for femur length (FL), gin trap width (GTW), head width (HW), pronotum width (P), File length (FI) and harp area (HA) (all measurements were recorded in mm and mm2).

Importance of components: PC1 PC2 PC3 PC4 PC5 PC6 Standard deviation 0.769 0.569 0.237 0.132 0.106 0.0851 Proportion of Variance 0.587 0.321 0.056 0.0173 0.011 0.00719 Cumulative Proportion 0.587 0.908 0.964 0.981 0.992 1.000 Standard deviations 0.769 0.568 0.237 0.132 0.106 0.0851

Rotation: PC1 PC2 PC3 PC4 PC5 PC6 FL -0.250 0.780 0.459 -0.169 0.257 0.150 GTW -0.00668 -0.0251 -0.195 -0.0668 -0.183 0.961 HW -0.125 0.202 0.191 0.767 -0.563 -0.0112 P -0.119 0.504 -0.837 0.0223 -0.0421 -0.165 FI -0.212 0.0193 0.0954 -0.609 -0.740 -0.165 HA -0.928 -0.307 -0.0622 0.0790 0.183 0.0132

Table 3: Logistic regression model examining the effect of male tactics on the outcome of staged territorial contests in Cyphoderris monstrosa (n=69).

Deviance Residuals: Min 1Q Median 3Q Max -2.2093 -0.7776 -0.4879 0.6663 2.0916

Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -1.041 0.573 -1.816 0.0694 Approacher -1.027 0.720 -1.427 0.154 Approach winner 2.433 0.802 3.033 0.00242 ** Bout winner 1.727 0.759 2.277 0.0228 *

Null deviance: 95.640 on 68 degrees of freedom Residual deviance: 69.272 on 65 degrees of freedom AIC: 77.272

Table 4: Logistic regression of male C. monstrosa status (winner/loser) with principal components of morphological measurements (n=16).

Deviance Residuals: Min 1Q Median 3Q Max -1.4312 -1.155 0.525 1.091 1.446

Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) 0.494 0.608 0.814 0.416 pc1 0.226 0.861 0.263 0.793 pc2 1.461 1.409 1.037 0.300 pc3 -2.277 2.788 -0.817 0.414

Null deviance: 21.930 on 15 degrees of freedom Residual deviance: 20.123 on 12 degrees of freedom AIC: 28.123

Vancouver Kamloops

Figure 5: Distribution of Cyphoderris buckelli, Cyphoderris monstrosa and Cyphoderris strepitans across western Canada and United States of America (Morris and Gwynne, 1978).

DutyCycle

Weight (mg)

Figure 6: Duty cycles (proportion of time spent singing) of all individuals (Winners and Losers) against their weight. There was no relationship between the two variables (n=16).

2Singing

CO V

Duty Cycle

Figure 7: Carbon dioxide production of singing Cyphoderris monstrosa males plotted against Duty cycles (proportion of time spent singing) of all individuals (Winners and Losers) the line of best fit includes all data points for Winners and Losers (n=16).

) )

2resting

/VCO

2singing Factorial metabolic scope (VCO scope metabolic Factorial

Duty Cycle

Figure 8: Metabolic scope (V.CO2singing/V.CO2rest) of Winners and Losers against their duty cycles (proportion of time spent singing). Winners have a higher scope per unit of singing than Losers (n=16).

Chapter 4 General Conclusions

This study has investigated how two different animals react to intense situations in which intraspecific competition may lead to agonistic behaviour. Chipmunks, Tamias striatus, prepare for winter sleep by collecting food in the fall to ensure that they have the required resources to survive the winter and, in doing so, compete with conspecifics for finite food resources. In the insect, Cyphoderris monstrosa, males engage in territorial contests to gain access to females and increase their reproductive success.

Chipmunks in Early October spent significantly more time in their burrows than above ground, when they were above ground they spent more time eating and gathering food than all of the other behavioural categories. In Late October chipmunks increased their burrow residency and decreased all of the above ground behaviours. The results did not support the hypothesis that the animals would increasingly spend more time in aggressive protection of their growing cache as time got closer to the induction of winter sleep. Rather the findings show that while above ground the animals spent little time in agonistic encounters over both intervals of observations.

In Chapter 3 my hypothesis was also not supported by the data. I hypothesized that animals with high duty cycles would use less energy while singing to support increased fighting abilities. The results showed, however, that successful males did not have lower singing metabolic rates. Instead, successful males were those who increased their metabolic rate during singing by a greater factor relative to their resting rates (CO2singing/CO2resting). These results suggest that the physiological basis of C. monstrosa territorial contests is the accumulation and mobilization of energy resources.

Although the two study species are very different in their ecology, physiology and behaviour they both demonstrate agonistic behaviours when the conditions are necessary to do

51 so. This study shows that Chipmunks do not show significant levels of aggression in comparison to other behaviours, while they are above ground, during a mast year. C. monstrosa display territorial behaviours with other males and the data suggests that the underlying metabolic condition which separates Winners and Losers is the metabolic scope of CO2 for the males from their resting rate to the singing rate. In conclusion, aggression can be studied in a diverse range of animals and for each system the functioning and mechanisms driving the behaviour deserve individual investigation.

Literature Cited

Austad ,S.N.1983. A game theoretical interpretation of male combat in the bowl and doily spider (Frontinella-Pyramitela). Animal Behaviour. 31: 59-73.

Batchelor, T.P., and Briffa, M. 2010. Influences on resource-holding potential during dangerous group contests between wood ants. Animal Behaviour. 80: 443-449.

Bennet-Clark, H.C. 1998. Size and scale effects as constraints in insect sound communication. Philosophical Transactions: Biological Sciences. 353: 407-419.

Bridge, A.P., Elwood, R.W., and Dick, J.T.A. 2000. Imperfect assessment and limited information preclude optimal strategies in male-male fights in the orb-weaving spider Metellina mengei. Proceedings of the Royal Society of London, Series B. 267: 273-279.

Briffa, M. 2008. Decisions during fights in the house cricket, Acheta domesticus: mutual or self- assessment of energy, weapons and size? Animal Behaviour. 75: 1053-1062.

Briffa, M. and Sneddon, L.U. 2007. Review: Physiological constraints on contest behaviour. Functional Ecology. 21: 627-637.

Brodin, A. and Clark, C.W. 1997. Long-term hoarding in the Paridae: a dynamic model. Behavioural Ecology. 8: 178-185.

Brown, W.D., Chimenti, A.J. and Siebert, J.R. 2007. The payoff of fighting in House crickets: motivational asymmetry increases male aggression and mating success. Ethology. 113: 457-465.

Clarke,M. and Kramer, D. 1994. Scatter-hoarding by a larder-hoarding rodent: intraspecific variation in the hoarding behaviour of the eastern chipmunk, Tamias striatus. Animal Behavior. 48: 299-308.

Clarke, D.C., and Moore, A.J. 1995. Variation and repeatability of male agonistic hiss characteristics and their relationship to social rank in Gromphadorhina portentosa. Animal Behaviour. 50: 719-729.

Davies, N.B. and Houston, A.I. 1984. Territory economics. In: Behavioural Ecology (Ed. By J.R. Krebs and N.B. Davies). Pp 148-169. Sunderland, Massachuetts: Sinauer.

Dodson, G.N., Morris, G.K., and Gwynne, D.T.1983. Mating behaviour of the primitive orthopteran genus Cyphoderris (Haglidae). In: Orthopteran Mating Systems: Sexual competition in a diverse group of insects (Gwynne, D.T. and Morris G.K. Eds.) pp 305- 318. Bolder Colorado: Westview Press.

Eccard, J.A., Fey, K., Caspers, B.A. and Ylönen, H. 2011. Breeding state and season affect interspecific interaction types: indirect resource competition and direct interference. Oecologia. 167: 623-633.

Eggert, A.K., and Sakaluk, S.K. 1994. Fluctuating asymmetry and variation in the size of courtship food gifts in decorated crickets. American Naturalist. 144: 708-716.

Elias, D.O., Kasumovic, M.M., Punzalan, D., Andrade, M.C., and Mason, A.C. 2008. Assessment during aggressive contests between male jumping spiders. Animal Behaviour. 76: 901-910.

Elliott, L. 1978. Social behaviour and foraging ecology of the eastern chipmunk (Tamias striatus) in the Adirondack mountains. Smithsonian Contributions to Zoology. 263: 1-107.

English-Loeb, G.M., Karban, R. and Hougen-Eitzman, D. 1993. Direct and indirect competition between spider mites feeding on grapes. Ecological Applications. 3: 699-707.

Enquist, M. and Leimer, O. 1987. Evolution of Fighting Behaviour: The Effect of Variation in Resource Value. J. theor. Biol. 127: 187-205.

Frame, P.F., Cluff, H.D., and Hik, D.S. 2008. Wolf reproduction in response to caribou migration and industrial development on the central barrens of mainland Canada. Arctic. 61: 134-142.

Fry, F.E.J. 1947. Effects of environmental factors on animal activity. University of Toronto Studies, Biological Series. 55: 1-62.

Galvez, D., Kranstauber, B., Kays, R.W. and Jansen, P.A. 2009. Scatter hoarding by the Central American agouti: a test of optimal cache spacing theory. Animal Behaviour. 78: 1327- 1333.

Getty, T. 1981. Territorial behaviour of the eastern chipmunk (Tamias striatus): encounter avoidance and spatial time-sharing. Ecology. 62: 915-921.

Hack, M.A. 1997. The energetic costs of fighting in the house cricket, Acheta domesticus L. Behavioral Ecology. 8: 28-36.

Haley, M.P. 1994. Resource-holding power asymmetries, the prior residence effect, and reproductive payoffs in male Northern Elephant Seal fights. Behavioral Ecology and Sociobiology. 34: 427-434.

Haugaasen, T. and Tuck Haugaasen, J.M. 2010. Cache pilferage may be prominent in Neotropical forests. Mammalia. 72: 423-425.

Hinds, D.S., Baudinette, R.V., MacMillen, R.E., and Halpern, E.A. 1993. Maximum metabolism and the aerobic factorial scope of endotherms.

Hitchcock, C.I. and Houston, A.I. 1994. The value of a hoard: not just energy. Behavioral Ecology. 5: 202-205.

Hoback, W.W., and Wagner, W.E. 1997. The energetic cost of calling in the variable field cricket, Gryllus lineaticeps. Physiological Entomology. 22: 286-290.

Hofman, H.A., and Schildberger, K. 2001. Assessment of strength and willingness to fight during aggressive encounters in crickets. Animal Behaviour. 62: 337-348.

Humphries, E.L., Hebblethwaite, A.J., Batchelor, T.P., and Hardy, I.C.W. 2006. The importance of valuing resources: host weight and contender age as determinants of parasitoid wasp contest outcomes. Animal Behaviour. 72: 891-898.

Ickes, R.A. 1974. Agonistic behaviour and the use of space in the eastern chipmunk, Tamias striatus. Ph.D. dissertation.

Jansen, P.A., Bongers, F. and Hemerik, L. 2004. Seed mass and mast sedding enhance dispersal by a neotropical scatter-hoarding rodent. Ecological Monographs. 74: 569-589.

Johnson, J.C., Ivy, T.M., and Sakaluk, S.K. 1999. Refmale reamting propensity contingent on sexual cannibalism in sagebrush crickets, Cyphoderris strepitans: a mechanism of cryptic female choice. Behavioral Ecology. 10: 227-233.

Kasumovic, M.M., Elias, D.O., Punzalan, D., Mason, A.C., and Andrade, M.C. 2009. Experience affects the outcome of agonistic contests without affecting the selective advantage of size. Animal Behaviour. 77: 1533-1538.

Kawamichi, T. 1976. Hay Territory and dominance rank of pikas (Ochotona princeps). Journal of Mammalogy. 57: 133-148.

Kawamichi, M. 1996. Ecological factors affecting annual variation in commencement of hibernation in chipmunks (Tamias sibiricus). Journal of Mammalogy. 77: 731-744.

Kemp, D.J., and Wiklund, C. 2001. Fighting without weaponry: a review of male-male contest competition in butterflies. Behavioural Ecology Sciobiology. 49: 429-442.

Kumala, M., McLennan, D.A., Brooks, D.R., and Mason, A.C. 2005. Phylogenetic relationships within hump-winged grigs, Cyphoderris (Insecta, Orthoptera, Tettigonioidea, Haglidae). Canadian Journal of Zoology.83: 1003-1011.

Lacher, T.E. and Mares, M.A. 1996. Availability of Resources and us e of space in eastern chipmunks, Tamias striatus. Journal of Mammalogy. 77: 833-849.

Liu, X., Zhou, M., Bi, W. and Tang, L. 2009. New data on taxonomy of recent Prophalangopsidae (Orthoptera: Hagloidea). Zootaxa. 2026: 53-62.

Loew, S.S. 1999. Sex-biased dispersal in eastern chipmunks, Tamias striatus. Evolutionary Ecology. 13: 577-577.

Mason, A.C. 1994. Hearing and communication in the monster Haglid. University of Toronto, Biological Sciences.

Mason, A.C. 1996. Territoriality and the function of song in the primitive acoustic insect Cyphoderris monstrosa (Orthoptera: Haglidae). Animal Behaviour. 51: 211-224.

Maynard Smith, J. and Parker, G.A. 1976. The logic of asymmetric contests. Animal Behaviour. 24: 159-175.

Morris, G.K., and Gwynne, D.T. 1978. Geographical distribution and biological observations of Cyphoderris (Orthoptera: Haglidae) with a description of a new species. Psyche (Camb.). 85: 147-167.

Nijman, V. and Heuts, B.A. 2000. Effect of environmental enrichment upon resource holding power in fish in prior residence situations. Behavioural Processes. 49: 77-83.

Parker, G.A. 1974. Assessment strategy and the evolution of fighting behaviour. Journal of Theoretical Biology. 29: 221-240.

Pennycuick, C.J., Griffin, L.R., Colhoun, K. and Angwin, R. 2011. A trial of a non-statistical computer program for monitoring fuel reserves, response to wind and other details from GPS tracks of migrating geese. J Ornithol. 152: S87-S99.

Peppas, A.P., Yan, L., Shen, Y.T., Vatner, D.E., Vatner, S.F. and Kudej, R.K. 2009. Seasonal variation in ischemia tolerance in a true mammalian hibernator. Circulation. 120: S846- S846.

Prestwich, K.N. 1994. The Energetics of Acoustic Signaling in Anurans and Insects. American Zoologist. 34: 625-643.

Prestwich, K. N. and T. J. Walker. 1981. Energetics of singing in crickets: Effects of temperature in three trilling species (Orthoptera: Gryllidae). J. Comp. Physiol. 143:199-212

Prior, T.G., Hines, W.G.S., and Cressman, R. 1993. Evolutionary games for spatially dispersed populations. Journal of Mathmatical Biology. 32: 55-65.

Pyare, S., Kent, J.A., Noxon, D.L. and Murphy, M.T. 1993. Acorn preference and habitat use in eastern chipmunks. American Midland Naturalist. 130: 173-183.

Reichert, S.E. and Hammerstein, P. 1983. Game theory in the ecological context. Ann. Rev. Ecol. Syst. 14: 377-409.

Sakaluk, S.K., Morris, G.K., and Snedden, W.A. 1987. Mating and its Effect on Acoustic Signalling Behaviour in a Primitive Orthopteran, Cyphoderris strepitans (Haglidae): The Cost of Feeding Females. Behavioral Ecology and Sociobiology. 21: 173-178.

Sakaluk, S.K. and Snedden, W.A. 1990. Nightly Calling Durations of Male Sagebrush Crickets, Cyphoderris strepitans: Size, Mating and Seasonal Effects. Oikos. 57: 153-160.

Schroeder, R.L. and Vangilder, L.D. 1997. Tests of wildlife habitat models to evaluate oak-mast production. Wildlife Society Bulletin. 25: 639-646.

Snedden, W.A. 1996. Lifetime mating success in male sagebrusch crickets: sexual selection constrained by a virgin male mating advantage. Animal Behaviour. 51: 1119-1125.

Snedden, W.A., and Sakaluk, S.K. 1992. Acoustic signaling and its relation to male mating success in sagebrush crickets. Animal Behaviour. 44: 633-639.

Speakman, J.R. and Rowland, A. 1999. Preparing for inactivity: how insectivorous bats deposit a fat store for hibernation. Proceedings of the Nutrition Society. 58: 123-131

Stuart-Fox, D. 2006. Testing game theory models: fighting ability and decision rules in chameleon contests. Proceedings of the Royal Society of London, Series B. 273: 1555- 1561.

Taylor, P.W., Hasson, O., and Clark, D.L. 2001. Initiation and resolution of jumping spider contests: roles for size, proximity and early detection of rivals. Behavioral Ecology and Sociobiology. 50: 403-413.

Taylor, P.W. and Elwood, R.W. 2003. The mismeasure of animal contests. Animal Behaviour. 65: 1195-1202.

Vander Wall, S.B. and Jenkins, S.H. 2003. Reciprocal pilferage and the evolution of food- hoarding behaviour. Behavioral Ecology. 14: 656-667.

Wauters, L.A., Suhonen, J., Dhondt, A.A. 1995. Consequences of Hoarding Behaviour in the Eurasian Red Squirrel. Proceedings: Biological Sciences. 262: 277-281.

Yahner, R.H. 1977. The adaptive nature of the social system and behavior in the eastern chipmunk, Tamias striatus. Behav. Ecol. Sociobiol. 3: 397-427.