Thermal Ecology and Management of the Invasive Tawny Crazy Ant, Nylanderia Fulva (Mayr) (Hymenoptera: Formicidae)

THERMAL ECOLOGY AND MANAGEMENT OF THE INVASIVE TAWNY CRAZY ANT, NYLANDERIA FULVA (MAYR) (HYMENOPTERA: FORMICIDAE)

MICHAEL THOMAS BENTLEY

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA 2016

To my family and friends that helped make this achievement possible

ACKNOWLEDGMENTS

I would like to thank all of those who helped me complete this journey. First, I owe my deepest appreciation and gratitude to Dr. Faith Oi, my supervisory committee chair, for providing me the opportunity to return to graduate school and to make this degree possible.

Thank you for your motivation, support, understanding, and patience that helped keep me going through even the toughest of times. I am eternally grateful for all that you have done for me both professionally and personally.

Thank you to my other committee members, Dr. David Oi, Dr. Katie Sieving, and Dr.

Daniel Hahn. Your guidance and expertise within and outside of the field of entomology have provided an education that I will carry far beyond the field of science. It was an honor and a privilege to have been mentored by such remarkable scientists.

I would also like to acknowledge my parents, Mike and Jill Bentley, for their never- ending love and support of my academic endeavors. I am beyond fortunate to have always had the love and encouragement of these amazing people to help me reach the end of this long and difficult academic journey. It is an honor to share this degree with them both, because without them none of this would have ever been possible.

Lastly, thanks to all my additional family and friends that have motivated, supported, and sacrificed to help me reach this final chapter in my formal education. I have asked a lot from all of you these past few years, and you have gladly given more than I could have ever expected in return. Thank you.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 10

ABSTRACT ...... 13

CHAPTER

1 LITERATURE REVIEW ...... 15

Taxonomy and History of Nylanderia ...... 18 Biology and Habits of N. fulva ...... 22 Life Stages ...... 22 The Colony ...... 23 Nest Sites ...... 24 Communication ...... 24 Foraging Behavior ...... 26 Pest Status and Management of N. fulva ...... 28

2 THE THERMAL BREADTH OF NYLANDERIA FULVA (MAYR) (HYMENOPTERA: FORMICIDAE) IS NARROWER THAN THAT OF SOLENOPSIS INVICTA (BUREN) AT THREE THERMAL RAMPING RATES: 1.0, 0.12 AND 0.06 °C ...... 33

Introduction ...... 33 Materials and Methods ...... 35 Collection and Maintenance of Field Colonies...... 35 Critical Thermal Limits...... 35 Data Analysis...... 36 Results...... 37 Discussion ...... 38

3 TUNNELING PERFORMANCE INCREASES AT LOWER TEMPERATURES FOR SOLENOPSIS INVICTA (BUREN) BUT NOT FOR NYLANDERIA FULVA (MAYR) ...... 45

Introduction ...... 45 Materials and Methods ...... 47 Collection of Field Colonies ...... 47 Two-Dimensional Tunneling Assays and Experimental Design ...... 48 Analysis ...... 49 Results...... 49 Conclusions...... 50

5 4 COMPETITIVE SUCCESS FOR CARBOHYDRATE AND PROTEIN RESOURCES INCREASES AT LOWER TEMPERATURES FOR NYLANDERIA FULVA (MAYR) AND SOLENOPSIS INVICTA (BUREN) WHEN COLONIES ARE EQUAL-SIZED ...... 59

Introduction ...... 59 Materials and Methods ...... 60 Collection of Field Colonies ...... 60 All Assays ...... 61 Impact of Temperature on Foraging Behavior without Competition: ...... 61 Impact of Temperature on Foraging when Forced to Compete for Resources: ...... 62 Analysis ...... 62 Results and Discussion ...... 64 Without Competition ...... 64 In the Presence of Competition ...... 65

5 NUMERICAL ADVANTAGE IMPROVES COMPETITIVE SUCCESS FOR RESOURCES AT LOWER TEMPERATURE FOR NYLANDERIA FULVA (MAYR) AND AT HIGHER TEMPERATURES FOR SOLENOPSIS INVICTA (BUREN) ...... 77

Introduction ...... 77 Materials and Methods ...... 77 Collection of Field Colonies ...... 77 Impact of Temperature and Worker Ratio When Competing for Resources: ...... 78 Analysis...... 79 Results and Discussion ...... 80 250, 500, or 1000 N. fulva Vs a Competitor (500 S. invicta) ...... 80 500 S. invicta vs a Competitor (250, 500, or 1000 N. fulva) ...... 81

6 SOLENOPSIS INVICTA (BUREN) ARE DISPLACED FROM SUCROSE AND PROTEIN RESOURCES BY LOW AND HIGH NUMBERS OF NYLANDERIA FULVA (MAYR) ...... 97

Introduction ...... 97 Materials and Methods ...... 98 Collection of Field Colonies ...... 98 Solenopsis invicta Introduction Experiments ...... 99 Data Analysis...... 100 Results and Discussion ...... 101

7 FIELD POPULATIONS OF NYLANDERIA FULVA (MAYR) READILY CONSUME A DILUTED FORMULATION OF MAXFORCE™ QUANTUM LIQUID ANT BAIT IN CHOICE AND NO-CHOICE ASSAYS ...... 114

Introduction ...... 114 Materials and Methods ...... 114 2014 No-Choice Bait Assay ...... 115 2015 Choice Bait Assay ...... 116

6 Results and Discussion ...... 116

8 CONCLUSIONS ...... 124 APPENDIX:SAS CODE FOR REPEATED MEASURES ANALYSES OF VARIANCE (ANOVA) EXAMINING THE EFFECTS OF TEMPERATURE AND SPECIES ON NUMBERS OF RESOURCES OCCUPIED AND NUMBERS OF ANTS AT RESOURCES IN BOTH NO-COMPETITION AND COMPETITION EXPERIMENTS WHEN WORKER-RATIO WAS NOT MANIPULATED ...... 130

SAS Code for the Analysis of Data for Nylanderia Fulva and Solenopsis Invicta when Ants Foraged without Competition: ...... 130 SAS Code for the Analysis of Data for N. Fulva and a Competitor (S. Invicta) when N. Fulva was Given First Access to Resources: ...... 130 SAS Code for the Analysis of Data for S. Invicta and a Competitor (N. Fulva) when S. Invicta was Given First Access to Resources: ...... 131

LIST OF REFERENCES ...... 133

BIOGRAPHICAL SKETCH ...... 144

7 LIST OF TABLES

Table page

2-1 A mixed-model analysis of variance (ANOVA) summary table evaluating the critical thermal minima (CTmin) and critical thermal maxima (CTmax) for Nylanderia fulva and Solenopsis invicta at three rates of temperature change: 0.06, 0.12, 1.0 °C min-1 ...... 41

2-2 Critical thermal minima (CTmin) and critical thermal maxima (CTmax) values (± SE) for Nylanderia fulva and Solenopsis invicta at three rates of temperature change: 0.06, 0.12, 1.0 °C min-1 ...... 42

3-1 A mixed model analysis evaluating the tunneling performance of Nylanderia fulva and Solenopsis invicta at 15.0, 18.0, 20.0, and 22.0 °C ...... 53

3-2 A two-way ANOVA summary table by species evaluating the tunneling performance of Nylanderia fulva and Solenopsis invicta at 15.0, 18.0, 20.0, and 22.0 °C ...... 54

4-1 Repeated measures analyses of variance table for the main effects of temperatureand resource on the number of resources occupied and the number of ants at resources for Nylanderia fulva and Solenopsis invicta without competition ...... 70

4-2 Repeated measures analyses of variance table for the main effects of temperature and bait type on the number of resources occupied and the number of ants at resources when the competitor was Solenopsis invicta ...... 71

4-3 Repeated measures analyses of variance table for the main effects of temperature and bait type on the number of resources occupied and the number of ants at resources when Nylanderia fulva was the competitor ...... 72

5-1 Analyses of variance table for main effects of worker ratio, temperature, and resource on the cumulative number of resources occupied and the cumulative number of ants at resources when the competitor was S. invicta ...... 84

5-2 Comparison of the effect of temperature by resource over all worker ratios (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) when the competitor was S. invicta ...... 85

5-3 Comparison of the effect of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) by resource over all temperatures when the competitor was S. invicta ...... 86

5-4 Analyses of variance table for main effects of worker ratio, temperature, and resource on the cumulative number of resources occupied and the cumulative number of ants at resources when the competitor was N. fulva ...... 87

8 5-5 Comparison of the effect of temperature by resource over all worker ratios (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) when the competitor was N. fulva ...... 88

5-6 Means comparison of the effect of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) over all temperatures when the competitor was N. fulva ...... 89

6-1 Analysis of variance table showing the effects of forager intensity and resource type on the number of Solenopsis invicta counted at Vienna sausage and honey resources ...104

6-2 Analysis of variance table showing the effects of forager intensity and resource type on the number of Nylanderia fulva counted at Vienna sausage and honey resources when N. fulva were present in high numbers or low numbers ...... 105

LIST OF FIGURES

Figure page

2-1 Mean critical thermal values (± SE) for Nylanderia fulva (NF) and Solenopsis invicta (SI) at three rates of temperature change per min ...... 43

2-2 Regression of critical thermal minima (CTmin) and critical thermal maxima (CTmin) values for workers of Nylanderia fulva and Solenopsis invicta ...... 44

3-1 Tunneling arena ...... 55

3-2 Mean total (±SE) tunneling distances (cm) per temperature for Nylanderia fulva (NF) and Solenopsis invicta (SI)...... 56

3-3 Mean (±SE) daily tunneling distances (cm) for Nylanderia fulva (Nf) and Solenopsis invicta (Si) over all temperatures...... 57

3-4 Tunneling performance at 20.0 °C after three days ...... 58

4-1 Diagram of the foraging arena used to evaluate the number of resources occupied and number of ants at resources for Nylanderia fulva and Solenopsis invicta ...... 73

4-2 In the absence of competition, mean number (±SE) of resources occupied by ants and mean number of ants at resources when Nylanderia fulva and Solenopsis invicta were placed separately in an area with two resources at 15.0, 25.0 and 35.0 °C...... 74

4-3 Mean number (±SE) of resources occupied by ants (A-C) and of ants at resources (D-F) when Nylanderia fulvaand competitor S. invicta were placed in an arena with two resources at 15.0, 25.0 and 35.0 °C ...... 75

4-4 Mean number (±SE) of resources occupied by ants (A-C) and of ants at resources (D-F) when Solenopsis invicta and competitor Nylanderia fulva were placed in an arena with two resources at 15.0, 25.0 and 35.0 °C ...... 76

5-1 Diagram of the foraging arena used to evaluate the foraging behavior of Nylanderia fulva and Solenopsis invicta at 15.0, 25.0, and 35.0 °C ...... 90

5-2 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 15.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources ...... 91

5-3 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 25.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources ...... 92

5-4 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 35.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources ...... 93

5-5 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 15.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources ...... 94

5-6 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 25.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources ...... 95

5-7 Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 35.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources ...... 96

6-1 A diagram showing the placement of sliced Vienna sausage (ca. 1.0 g) (Original Vienna Sausage, Cherry Hill, NJ) baits ...... 106

6-2 Plastic excluder ...... 107

6-3 Cumulative count of Solenopsis invicta (± SE) (n = 6 per treatment) on Vienna sausage and honey resources ...... 108

6-4 Mean number of Solenopsis invicta (± SE) (n = 6 per time point for each treatment or control) on Vienna sausage and honey resources ...... 109

6-5 Mean number of Nylanderia fulva (± SE) (n = 6 per time point for each treatment) on Vienna sausage and honey resources ...... 110

6-6 Cumulative count of Solenopsis invicta (± SE) (n = 6 per treatment) on honey and Vienna sausage and honey resources ...... 111

6-7 Cumulative count of Nylanderia fulva (± SE) (n = 6 per treatment) on honey and Vienna sausage resources ...... 112

6-8 Cumulative count of Nylanderia fulva (± SE) (n = 6 per treatment) on Vienna sausage and honey resources ...... 113

7-1 Diagram of 19 bait stations ...... 118

7-2 Diagram of 16 pairs of bait stations ...... 119

7-3 Cumulative mean (± SE) volume of diluted Maxforce™ Quantum liquid ant bait depleted from 19 bait stations ...... 120

7-4 Cumulative mean (± SE) volume of diluted Maxforce™ Quantum liquid ant bait and 25.0% sucrose solution depleted from 16 pairs of bait stations...... 121

7-5 Mean (± SE) number of Nylanderia fulva (n = 19 per service date) sampled at each bait station ...... 122

7-6 Mean (± SE) number of Nylanderia fulva (n = 16 per service date) sampled at each bait station in 2015 ...... 123

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

THERMAL ECOLOGY AND MANAGEMENT OF THE INVASIVE TAWNY CRAZY ANT, NYLANDERIA FULVA (MAYR) (HYMENOPTERA: FORMICIDAE)

Michael Thomas Bentley

December 2016

Chair: Faith M. Oi Major: Entomology and Nematology

Few invasive insects have been more deleterious to non-native ecosystems than ants. Nylanderia fulva (Mayr) is an invasive pest ant established in the southern US. While conventional pest control treatments have struggled to provide consistent long-term control of N. fulva, field observations indicate that when temperatures drop to <10-15 oC foraging activity decreased, and thus, homeowner complaints. The overarching goal of this work was to understand how temperature impacts the biology and behavior of N. fulva compared to Solenopsis invicta (Buren), the red imported fire ant. To determine how temperature could shape their distribution and activity, the impacts of body size and thermal ramping rate were evaluated. Overall, N. fulva had a narrower thermal breadth than S. invicta. Each year, ant populations rebound when temperatures warm. Both ant species were observed to tunnel in soil. Solenopsis invicta’s gallery system is used for thermoregulation, but the extent of N. fulva’s ability to tunnel had not been documented. Thus, tunneling performances were compared between species at four temperatures: 15.0, 18.0, 20.0, and 22.0 °C. Overall, N. fulva tunneled significantly less than S. invicta. Nylanderia fulva tunneled furthest at warmer temperatures; whereas, S. invicta tunneled furthest at cooler temperatures. Temperature also affects rates of

foraging and competition. Competition assays were done to test the effects of temperature, worker ratio, species and resource type. In laboratory assays at 15.0°C, as the ratio increased to favor N. fulva, more N. fulva were recorded at resources. At 35.0°C, S. invicta was the better competitor in ratios up to 1:1, but were displaced when ratios increased to 2:1 in favor of N. fulva. Field assays confirmed laboratory observations: colony fragments of S. invicta were displaced from resources by high and low intensities of N. fulva. Finally, N. fulva’s consumption of a diluted formulation of Maxforce™ Quantum liquid ant bait was measured in field trials. As expected, N. fulva workers depleted more diluted bait when a competing carbohydrate source was not present, emphasizing the case for sanitation. These studies improve our knowledge of N. fulva’s thermal ecology, which can advance our understanding of geographic distributions and management strategies.

CHAPTER 1 LITERATURE REVIEW

An estimated 50,000 pest species have been introduced into the U.S. (Pimentel et al.

2005), with approximately 1,000 of those species inhabiting Florida (Frank and McCoy 1995).

Nearly 15% of the invasive species found throughout the U.S. are arthropods, accounting for agricultural losses that exceed $120 billion annually (Pimentel et al. 2005). Greater than $13.5 billion of these costs are attributed to control efforts and loss resulting from insect damage

(Dowell and Krass 1992, Pimentel 1997, Vinson 1999, Pimentel et al. 2005). Few insects have had greater deleterious impacts on non-native ecosystems than invasive ants.

Ants have successfully established in a range of ecological niches and have colonized every continent except Antarctica (Hölldobler and Wilson 1990). Greater than 13,000 ant species are identified, with new species confirmed annually (Ward 2014). Ants serve many beneficial roles in their ecosystems. Their large populations are food for mammals, reptiles, and other insects (Redford and Dorea 1984, Gotelli 1996). Several ant species also play a vital role in seed dispersal and soil turnover (Hölldobler and Wilson 1990).

When ants inhabit non-native ecosystems, their presence can have significant negative effects at many trophic levels. Even small populations of invasive ants have been known to disrupt mutualistic relationships among native ant species that can lead to a cascade of adverse effects throughout the food chain (Ness and Bronstein 2004, Styrsky and Eubanks 2007, Ness et al. 2009). The most dramatic ecological impacts occur when native ants are displaced by invasive ants, disrupting mutualistic interactions between native ants and other organisms such as seed dispersal or protection from predators (Holway et al. 2002). These negative impacts are most evident in fragile island ecosystems that are considered far more vulnerable due to their isolation and low numbers of native ant species (McGlynn 1999, Holway et al. 2002).

Less than 1% of all ant species identified are considered invasive. Currently, A. gracilipes, L. humile, Pheidole megacephala (Fabricius), Solenopsis invicta (Buren), and

Wasmannia auropunctata (Roger) rank 6th, 48th, 68th, 86th and 100th, respectively, as invasive species (Global Invasive Species Database 2013). Their relative positions among other species may change depending on which variables are considered, such as gross economic loss or total ecological impact. It is important to note that this list is not exhaustive, and additional ant species should be considered as important invasive pests.

Invasive ant species share similar traits such as high reproductive rates, ability to survive human-mediated dispersal, and polygyny that contribute to their competitive success when exploiting resources and expanding territories (Passera 1994, Moller 1996). High reproductive rates facilitate the production of large colonies. Invasive ants are able to devote a larger percentage of their colony to geographic expansion and resource location (Holway 1999). There are two forms of dispersal among ants: winged dispersal of female reproductives and budding.

Winged dispersal is far less common among most invasive ant species, with an important exception being the monogyne form of S. invicta which is the predominant form in the U.S.

(Tschinkel 2006). Budding is a method of colony formation in which a few queens separate from the natal colony with a small group of workers, and locate to a new nesting site (Holway 1998,

Holway 1999, Holway et al. 2002, Abbott 2005). Budding is the more common dispersal method among polygynous species such as Linepithema humile (Mayr) and Anoplolepis gracilipes

(Smith). Colonies that reproduce by budding have lower rates of spread in the absence of human- mediated transport compared to colonies that reproduce by mating flights (Suarez et al. 2001).

Shifts in colony structure and behavior can occur after ants arrive in a new area (Tsutsui and Suarez 2003). Colonies of many highly successful invasive ants transition from small,

territorially-oriented colonies to organized, cooperative “supercolonies” after ants become established. Holway and Suarez (1999) identify reduced intraspecific aggression that can lead to a loss in territoriality, increasing colony cooperation as one of the shifts in behavior. Recent studies also have linked the formation of supercolonies to a decrease in genetic diversity within many invasive ant species because of a population bottleneck during founding (Ross et al. 1996,

Tsutsui et al. 2000, Tsutsui and Case 2001, Tsutsui and Suarez 2003).

Solenopsis invicta (Buren) is among the most studied examples of behavioral changes between native and invasive ants. Solenopsis invicta exist in both monogyne and polygyne forms

(Ross et al. 1996, Tschinkel 1998). Polygyne S. invicta colonies indigenous to Argentina contained higher numbers of genetically related queens and had lower nest densities than introduced colonies in the U.S. (Ross et al. 1996). The greater genetic variation in queens of native polygyne S. invicta colonies versus introduced polygyne colonies correlated a loss in intraspecific aggression in introduced colonies, resulting in unrelated queens to be adopted into existing or founding colonies more often in non-native ranges (Ross et al. 1996). As a result, polygyne nest densities increase, allowing for a numerical advantage in invasive S. invicta colonies over native ant species and thus a greater likelihood of displacement (Macom and Porter

1990, Tschinkel 1998, Holway and Suarez 1999).

Recently, another South American ant, Nylanderia fulva (Mayr), has invaded the southern U.S. Nylanderia fulva’s aggressive foraging behavior and large polygynous colonies, which exhibit high rates of new colony formation through budding, allow this invasive ant to quickly build up large infestations (Zenner-Polania 1990a, Zenner-Polania 1994). In Texas,

LeBrun et al. (2013) reported that N. fulva populations displaced another ecologically dominant invasive species, Solenopsis invicta, as well as reduced species abundance and richness of native

ants in sampled locations. Nylanderia fulva populations may also pose a threat to small wildlife and ground nesting birds that can easily be overwhelmed by their dense foraging populations. It is possible that N. fulva may also displace native ant species where it has become established in

Florida, Mississippi and Louisiana.

While conventional pest control treatments have struggled to provide consistent long- term control of N. fulva, field observations indicate that seasonal cold exposure leads to a decrease in foraging activity and thus homeowner complaints (Hill 2013, D. Calibeo, personal communication). Unpublished results from seasonal baiting studies have recorded large population decreases during fall and winter seasons, with numbers rebounding in early spring and summer (D. Calibeo, personal communication). These unpublished results are supported by field observations as well as feedback from homeowners and pest management professionals who indicate that “winter temperatures” in north Florida are their only source of long term relief from N. fulva populations. Given N. fulva’s native range is tropical, it is reasonable that a decrease in activity is experienced during periods when air temperatures are low. However, we currently lack data evaluating the impact of temperature on N. fulva’s activity, thermoregulatory behavior, and interspecific competition to fully understand the effects of seasonal cold on N. fulva populations. Given that seasonal cold temperatures are thought to effect N. fulva’s activity, and thus rates of homeowner complaints, further examination of the impacts of temperature on

N. fulva’s activity are needed.

Taxonomy and History of Nylanderia

Comprised of more than 130 extant species, the genus Nylanderia is one of the largest and ecologically significant belonging to the genus-group Prenolepis (Hymenoptera:

Formicidae) (LaPolla et al. 2010). The genus Prenolepis was originally described by Mayr in

1861, two years before Motschoulsky (Motschoulsky 1863 in LaPolla et al. 2010) first described

the genus Paratrechina. Nylanderia was first described as one of three subgenera within the genus Prenolepis by Emery (Emery 1906 in LaPolla et al. 2010). In 2010, the genus Prenolepis underwent a major taxonomic revision in which Nylanderia and Paraparatrechina were re- elevated to genera (LaPolla et al. 2010). Of the ant species formerly assigned to the genus

Paratrechina, 133 species were reassigned to the genus Nylanderia and 28 species assigned to the genus Paraparatrechina. Paratrechina longicornis (Latreille) is the only ant species remaining within the genus Paratrechina. Nylanderia fulva (Mayr), N. bourbonica, and N. vividula (Nylander) are considered the most important species of this genus in North America because of their status as pests in nurseries, greenhouses, laboratories, insectaries, homes, and other buildings (Trager 1984).

Globally, Nylanderia species are among the most abundantly encountered ants where that genus-group is known to occur (Bolton et al. 2006) and can be found in nearly all mid- and low- latitude geographic locations (LaPolla et al. 2011). Nylanderia species are well adapted to survive in most environmental conditions, but thrive in forested environments with high humidity and tropical temperatures (LaPolla et al. 2011). Many Nylanderia species are known to be rapid and efficient foragers, well-adapted to resource location on forest floors where leaf litter conditions are favored nesting sites (LaPolla et al. 2011).

Before the revision of the genus Paratrechina by LaPolla et al. (2010), a pest ant found in

Florida that was initially identified as Paratrechina pubens Forel was first described in the

Caribbean on St. Vincent Island by Auguste Forel in 1893 (Wetterer and Keularts 2008). In July

1905, samples of this ant were collected on the Bermuda Islands, but originally catalogued as

Prenolepis sp. and later identified as P. pubens (Wetterer 2007). Paratrechina pubens also was reported in Martinique (Forel 1912 in Wetterer and Keularts 2008). In 1928, Marlatt suggested

that N. pubens originated from Brazil (Marlatt 1928 in Wetterer and Keularts 2008, Wetterer

2007). By 1951, N. pubens was reported from Cuba, Mexico, Central and South America, and the West Indies (Wetterer 2007). The first records of N. pubens in the U.S. came from Miami,

Florida in 1953 (Trager 1984). It was not recognized as a pest species in the U.S. until 1990 when large collections of what were then identified as N. pubens were made from a hospital, commercial buildings, and later the University of Miami campus (Klotz et al. 1995, Deyrup et al.

2000). In 2004, the common name, “Caribbean crazy ant” was proposed for the Nylanderia sp. in

Florida based upon its suspected origins (Warner and Scheffrahn 2004); however, we now believe these ants to have been N. fulva. It is thought that the initial 1990 reports of N. pubens in

Florida might have been N. fulva (Trager 1984, Gotzek et al. 2012). However, this hypothesis cannot be confirmed because reference male specimens of many early accounts of either species do not exist.

Taxonomic uncertainty and improper nomenclature have surrounded N. fulva and N. pubens. In 2002, a pest management professional, Tom Rasberry, discovered a large population of unknown ants in Pasadena Texas. Of the samples submitted to Meyers (2008), a taxonomic distinction was made between N. pubens and N. fulva. There were 3 to 4 pairs of macrosetae on the mesonotum of a Nylanderia sp., compared with two pairs present on N. pubens. These characters combined with additional phylogenetic analyses prompted Meyers (2008) to declare the unknown ant as Nylanderia (formerly Paratrechina) sp. nr. pubens. Examination of additional samples proved the macrosetae characters to be insufficient; however, Nylanderia

(formerly Paratrechina) sp. nr. pubens designation persisted in the literature for a few years.

In 2009, “N. sp. nr. pubens” were reported along the coast of Mississippi, and Louisiana in 2010 (Hooper-Bùi et al. 2010, MacGown and Layton 2010). Zhao et al. (2012) then utilized

molecular markers to demonstrate that the Nylanderia sp. from Texas and Florida were the same species, but the question of which species remained unresolved. Without the macrosetae characters described in Meyers (2008), workers of N. fulva and N. pubens were virtually indistinguishable. The only differentiating characters remaining was the setal pattern on the male parameres that Gotzek et al. (2012) used in combination with morphometric, phylogenetic, and molecular techniques to finally resolve the identity of the ant as N. fulva (Gotzek et al. 2012). In

2013, the Entomology Society of America approved “tawny cray ant” as the common name for

N. fulva, where “fulva” is Latin for tawny (D. H. Oi, personal communication).

Nylanderia fulva is believed to be native to Brazil (Arcila et al. 2002). Names used to describe this species have included the hairy crazy ant, Caribbean crazy ant (primarily in

Florida), the Rasberry crazy ant (primarily in Texas), hormiga loca, formiga cuiabana, or doceira.

In 1971, N. fulva was first reported as a pest on coffee plantations and cattle farms in Puerto

Boyacá, Columbia after attempts to utilize it as a biological control agent against leaf-cutter ants and venomous snakes were unsuccessful (Zenner-Polania 1990b). Ant populations on these farms overwhelmed cattle, blinded calves, and contributed to the drying of grasslands. It was later determined that a mutualistic relationship between N. fulva and the plant feeding homopteran Antonina sp. was likely the cause of grassland destruction (Lopez 1975 in Zenner-

Polania 1990b). Over the next 10 years, N. fulva was reported as a pest in several locations throughout Columbia, likely being transported through soil and dung shipments (Zenner-Polania

1990b). Columbia residents experienced relief around 1980 when N. fulva populations reportedly declined, but in 1987 large numbers of ants were again reported in homes, coffee plantations, and cattle farms.

Within the U.S., N. fulva have been confirmed in a total of 81 counties throughout

Florida (27), Mississippi (3), Georgia (3), Louisiana (19 parishes), Alabama (1) and Texas (28)

(Oi 2015). Nylanderia fulva populations appear to have the broadest distribution in Texas and

Florida. The 27 counties in Florida where N. fulva populations have been identified include

Alachua, Bay, Brevard, Broward, Clay, Collier, DeSoto, Dade, Duval, Hardee, Hillsborough,

Indian River, Lake, Lee, Manatee, Marion, Martin, Nassau, Orange, Osceola, Palm Beach,

Pasco, Pinellas, Polk, St. Johns, Saint Lucie, and Sarasota counties (Klotz et al. 1995, Warner and Scheffrahn 2004, MacGown 2013). The 28 Texas counties where populations of N. fulva have been confirmed include Augustine, Bastrop, Bexar, Brazoria, Brazos, Cameron, Comal,

Fayette, Fort Bend, Chambers, Galveston, Hardin, Harris, Hays, Hidalgo, Jefferson, Jim Hogg,

Liberty, Matagorda, Montgomery, Nueces, Orange, Polk, Travis, Victoria, Walker, Wharton, and

Williamson (Meyers and Gold 2008, Meyers 2008, McDonald 2012, MacGown 2013, Oi 2015).

Biology and Habits of N. fulva

Life Stages

Nylanderia fulva queens are brown in color and tend to be marginally larger than workers and males, because they are physogastric. They range in body length without antenna from 5.06 to 5.30 mm. Female and male alates are produced periodically, but mating flights have not been observed. During oviposition, queens will seek high-humidity environments to deposit clusters of 17 to 25 eggs within leaf litter or crevices. Eggs are milky white in color with average height and width of 0.18 x 0.281 mm (Zenner-Polania 1990a). Late instar N. fulva larvae measure approximately 1.84 mm long; whereas, pupae are approximately 2.12 mm in length (Zenner-

Polania 1990a, Gotzek et al. 2012).

Nylanderia fulva workers are monomorphic, measuring approximately 2.17 to 2.41 mm in length and typical of the genus. Antennae have 12-segments and are clubless. Pubescence is

uniformly dense across mesosoma and the mesopleuron. Macrosetae on the mesosoma are long, flexuous and acuminate. Mandibles and maxillary palps are large in proportion to the head, with mandibles having six teeth (Zenner-Polania 1990a, Gotzek et al. 2012, MacGown 2013).

Nylanderia fulva males are marginally larger than workers, ranging from 2.41 to 2.77 mm in length and possess two pairs of abdomen-length wings which most easily distinguishes them from the worker caste. Their body structure is slightly more developed when compared to workers with a longer, slender abdomen and more sclerotized thorax (Meyers and Gold 2008).

Though color is relatively similar among males and workers, antennae on males are 13- segmented with reduced mandibles. Nylanderia fulva males are the caste known to possess taxonomic characters that separate N. pubens from N. fulva. In N. fulva males, the parameres are triangular and less sclerotized with fewer macrosetae along the paramere margins compared to the parameres of N. pubens which are more rounded, well-sclerotized, and have dense, fanlike macrosetae along the paramere margins (Gotzek et al. 2012).

The Colony

Ants utilize a complex social structure comprised of a caste system and a complimentary division of labor (polyethism) in order to coordinate functions within the colony. Ant colony members belong to one of three castes: workers, soldiers, and reproductives. Workers are primarily responsible for nest construction, brood care, and resource allocation. Defense, protection and resource retrieval are often done by soldiers. Lastly, reproductives are responsible for the perpetuation of their species. Another form of polyethism known as “age” or temporal polyethism is the coordination of behaviors based upon age of the individual. As colony members age, tasks transition from interior nest activities such as brood care to exterior activities such as foraging.

Nylanderia fulva colonies are polygynous and polydomous, capable of producing extremely large populations. Like other ant species, temperature influences colony growth and activity. Field studies suggest that N. fulva has low worker numbers and little brood production during winter months (McDonald 2012). As seasons transition into spring (≥ 15.0 °C), N. fulva colonies increase brood production and thus, worker populations. Colony growth peaks during summer months, before decreasing again as colonies transition back into winter growth cycles.

Nest Sites

Nest location and nest architecture are important for thermoregulation. Nest site selection of N. fulva is variable and opportunistic. Nests of N. fulva have been reported in leaf litter, rotting vegetation, mulch, bark, potted plants, and various man-made structures (Meyers 2008,

Meyers and Gold 2008). Ant species that relocate nests frequently, such as the nomadic army ant

Onychomyrmex hedleyi (Emery) (Miyata et al. 2003) and L. humile (Heller and Gordon 2006), select the type and amount of cover depending on ambient temperatures. Other ant species, such as S. invicta, construct subterranean nests that enables ants to transport brood into areas of the nest where temperatures are ideal (Penick and Tschinkel 2008). Solenopis invicta transport brood throughout the nest based on the nutritional condition of the colony (Porter and Tschinkel 1993).

Colonies provided with a surplus of resources transported brood to optimum colony temperatures around 31.0 °C; whereas, resource-limited colonies transported brood to slightly cooler temperatures closer to 30.0 °C, likely to reduce metabolic costs during starvation.

Communication

Ants are primarily ground-dwelling organisms. To efficiently coordinate tasks, ants utilize various forms of close-contact communication such as tactile, acoustical, and chemical means. Tactile communication involves the repeated antennation and foreleg contact with nestmates or other ants, and is commonly used in nestmate recognition. Acoustical signaling is

achieved through vibrations created from stridulation or through body tapping. Acoustical communication between ants can result in alarm, recruitment, and mate-signaling (Markl 1973,

Hölldobler and Wilson 1990). For some insects that engage in mutualistic relationships with ants, such as the treehopper Publilia concava Say (Morales et al. 2008) and larvae of the

Australian common imperial blue butterfly, Jalmenus evagoras (Donovan) (Travassos and Pierce

2000), acoustical signaling is used to attract ants during times of distress. Lastly, chemical communication occurs between ants through the production of semiochemicals to elicit alarm, attraction, recruitment, caste determination, and sexual communication responses from nestmates or intruders depending on necessity. Chemical communication plays a central role across ant species in coordinating interactions and functions at the individual and colony level.

Loose trail formations and strong group responses to stimuli indicate the presence of effective recruitment and trailing pheromones in N. fulva (Meyers and Gold 2008, McDonald

2012). However, the presence and concentrations of these pheromones is not yet identified.

Recently, research has focused on the potential role of defensive chemicals in N. fulva’s displacement of S. invicta in many areas (Chen et al. 2013, LeBrun et al. 2014, Zhang et al.

2015). Chen et al. (2013) determined that N. fulva maintained twice the quantity of formic acid when compared to P. longicornis. Lebrun et al. (2014) demonstrated that after N. fulva are exposed to S. invicta venom, N. fulva can detoxify the venom by applying abdominal exocrine gland secretions to its cuticle. Zhang et al. (2015) determined that formic acid synergized attraction of N. fulva workers to secretions of the Dufour’s gland, suggesting the mixture of these semiochemicals could be important to N. fulva’s strong recruitment of nestmates to resources or conflicts. These data collectively suggest that N. fulva’s use of chemicals may increase their defensive capabilities and resource procurement when compared to other ant species, as well as

contribute to N. fulva’s ability to displace S. invicta populations in non-native ranges. However, these findings do not address why N. fulva populations decline in many areas within approximately 10 years of their initial infestation. Furthermore, these data do not indicate how some S. invicta populations are able to move back into areas overrun by N. fulva after already being previously displaced.

Foraging Behavior

Foraging ants are predominantly scavengers. To sustain the often massive colony size workers to probe their environment for resources. When resources are encountered, most ant species will elicit the help of other nestmates to retrieve and utilize found items via chemical communication. Once resources are depleted, worker ants return to foraging, and the cycle continues (Wilson 1971). Constantly foraging for resources can be energetically costly. Several different species of ants have been observed to change foraging intensity daily and seasonally in response to the energy needs of the colony (Hölldobler and Wilson 1990).

Foraging behavior of ants is strongly influenced by internal factors of the colony such as nutritional needs, and by abiotic factors such as temperature (Howard and Tschinkel 1980, Cerdá et al. 1998a, Cook et al. 2011). The nutritional needs of a colony can change in response to conditions of starvation or resource abundance, and due to changes in colony size. To respond to these changes, ant colonies will alter rates of food exchange, recruitment levels, and the number of foragers, thus directly impacting their foraging behavior (Howard and Tschinkel 1980,

Mailleux et al. 2006).

Circadian rhythm is known to significantly impact ant foraging. Circadian rhythms among ants is believed to have evolved as a strategy to avoid interspecific competition for resources (Wilson 1971). For ant species sharing similar niches, distinctive daily foraging schedules across ant populations can easily be observed. Wilson (1971) reported consistent

patterns of Myrmecia, Rhytidoponera, Dacryon, and Iridomyrmex foragers dominating the vegetation in early to mid-morning hours. As late afternoon hours approached, those species began to retreat, allowing for nocturnal species, such as Colobostruma, Iridomyrmex, and

Camponotus to forage.

Temperature is one of the most important environmental factors known to influence the foraging activity of ants (Porter and Tschinkel 1987, Cerdá et al. 1998b, Heller and Gordon

2006, Barbieri et al. 2014). All ants have upper and lower thermal thresholds of activity that limit their ability to forage. Differences in thermal tolerance between ant species can also impact competition for resources by limiting the abundance and foraging activity for one species over another (Cerdá et al. 1998a, Cerdá et al. 1998b, Retana and Cerdá 2000).

Field observations suggest that foraging strategies of N. fulva are similar to those of other invasive unicolonial ant species such as L. humile and Anoplolepis gracilipes (Smith). Having large colonies with low intraspecific aggression is one factor that enables these species to establish dense foraging populations that cover expansive territories in search of resources

(Holway 1999, Lester and Tavite 2004). Their large populations also lend to successful mass recruitment strategies to quickly locate and retrieve resources (Holway et al. 2002).

Nylanderia fulva are omnivores, increasing the potential and abundance of viable resources encountered within native or non-native habitats (Holway et al. 2002). Similar to L. humile, N. fulva adults obtain the majority of the carbohydrates necessary to sustain large colonies from nectar-producing plant parts or from honeydew-producing hemipterans such as aphids, plant hoppers, and scales (McDonald 2012, Sharma et al. 2013). Proteins can include virtually any digestible animal parts. Field observations of a large N. fulva colony infesting a

coffee plantation recorded ants consuming immature and adult insects, birds, snakes, lizards, and even baby calves (Zenner-Polania and Ruiz-Bolanos 1985 in Zenner-Polania 1990a).

It appears that food type can impact ant behavior. In laboratory trials, when N. fulva were fed a carbohydrate-restricted diet, Horn et al. (2013) reported an increase in aggressive behavior between N. fulva and S. invicta. Similar observations have been reported in Formica aquilonia

(Yarrow) and L. humile on carbohydrate-based diets (Grover et al. 2007, Sorvari et al. 2008).

While this does provide insight into the causes of potential aggressive interactions between N. fulva and competing ant species, more research is needed to identify additional factors contributing to N. fulva’s high foraging efficiency and interspecific aggression.

Pest Status and Management of N. fulva

The pest status of N. fulva in both native and introduced habitats is well documented

(Zenner-Polania 1994, Meyers and Gold 2008, McDonald 2012, Hill 2013). Large colonies can pose a serious ecological threat to local fauna, displacing or even eliminating native species

(Zenner-Polania 1994). In Texas, N. fulva forced breeding populations of endangered ground nesting birds and other small mammals from nesting sites (Meyers 2008). Zenner-Polania (1994) reported that N. fulva populations invading Cimitarra, Columbia caused a 98.85% decrease in native ant species abundance. The primary cause of species displacement was attributed to resource competition based upon the observation that N. fulva populations were so large it was virtually impossible for any other ant species to exist in the area.

Nylanderia fulva do not sting; however, workers project formic acid from their acidopore which can act as an irritant to some individuals. In addition to their potential as a nuisance to people, N. fulva can also cause damage to electrical equipment. Residents have suffered power outages as a result of ants infesting electrical boxes and circuit breaker systems (Meyers 2008,

MacGown and Layton 2010). Nylanderia fulva also have been found infesting phone lines,

computers, and municipal sewage equipment (Meyers 2008, McDonald 2012). Furthermore, N. fulva have caused damage to commercial bee hives (Harmon 2009), and have infested landscape goods such as mulch and potted plants that are unable to be sold (Drees 2009). Even residential property values have diminished where N. fulva populations are present (Meyers 2008,

McDonald 2012, Friedman 2013).

Control strategies for N. fulva have had limited success. Early attempts to manage large infestations in Columbia included chemical and non-chemical strategies. Initially, repeated applications of carbaryl and aldrin were made to many of the agriculturally managed areas where

N. fulva populations were present with little success (Zenner-Polania 1990b). These products were soon restricted. Many economical, non-chemical management strategies also were explored. Water-filled moats and sticky materials were employed as economical control strategies for farmers. Physical barriers such as water-filled moats to prevent ants from crossing to a structure and sticky substances applied to the bases of trees to inhibit access to honeydew producing insects proved unsuccessful. Moats would fill with dead ants daily, allowing incoming ants to walk over dead ants, gaining access to structures. Sticky material applications to trees did show some levels of efficacy. However, this method is unrealistic in most agricultural settings where tree branches commonly contact neighboring trees, allowing treated areas to be bypassed

(Zenner-Polania 1990b).

The investigation of baits as management tools for invasive ants are well document

(Lofgren et al. 1975, Lofgren 1986, Klotz et al. 1997, Klotz et al. 1998a, Hooper-Bui and Rust

2000, Rust et al. 2002, Klotz et al. 2003, Warner 2005, Oi and Oi 2006). Granular baits that use corn grit as a carrier for the active ingredient dissolved in soybean oil have been effective in controlling ants such as S. invicta (Banks et al. 1973, Hu and Frank 1996, Oi and Oi 2006,

Furman and Gold 2006), but have had limited success with N. fulva. Meyer (2008) and

McDonald (2012) demonstrated potential control with Whitmire Advance® Carpenter Ant Bait

(Whitmire Micro-Gen, St. Louis MO), in both laboratory and field experiments. However, populations returned to pre-treatment levels in 3-4 weeks in field trials. Sugar-based liquid ant baits have been highly preferred by unicolonial ant species such as L. humile, and have demonstrated success in suppressing ants (Knight and Rust 1991, Klotz et al. 1998b, Tollerup et al. 2004). Although sugar-based baits are also highly palatable to workers of N. fulva, suppression of colonies over the long term have yet to be demonstrated.

Contact insecticides have been used to manage N. fulva populations, but their success in effectively managing this pest ant is inadequate. Most contact insecticides can only be applied as a crack and crevice treatment indoors. Exterior applications are limited due to water-quality concerns. Exterior insecticide sprays and granular products containing fipronil, such as

Termidor® SC (BASF, Florham Park, NJ) and Top ChoiceTM (Bayer Environmental Science,

Montvale, NJ) demonstrated temporary success (Meyers 2008). The combined use of some contact insecticides and baits has been attempted in order to extend the duration of control.

Meyers (2008) reported marginal effectiveness with dinotefuran, N-methyl-N’nitro[N”-

[(tetrahydro-3- furanyl)methyl]guanidine, and Whitmire Advance® Carpenter Ant Bait against

N. fulva populations. However, product-combination control strategies do have merit when used in an integrated pest management program and warrants further investigation of additional product combinations (Klotz et al. 2002, Klotz et al. 2003).

The use of pathogens as biological control agents to manage N. fulva also is being investigated. Currently, the Nylanderia fulva virus in the order Picornavirales (has been isolated

(Valles and Oi 2014) and one new microsporidian (Myrmecomorba nylanderian gen. et. sp. nov.) has been described (Plowes et al. 2015). The effectiveness of both pathogens is not yet known.

Integrated pest management strategies are being recommended for the control of N. fulva.

Management of this pest ant may be possible through the use of sanitation, habitat modification, and pesticide applications as needed (Calibeo and Oi 2011). The elimination of excess debris against and around properties can reduce potential breeding sites. Additionally, keeping tree limbs, hedges, and tall grasses managed and off of structures will prevent interior access to homes. Though these strategies are still being evaluated for long term success, the combination of insecticide applications with sanitation measures may be sufficient to reduce N. fulva populations to tolerable levels.

Nylanderia fulva populations in North Florida are most difficult to manage from early summer through fall when ant populations reach their peak. While pest management professionals have struggled to find reliable solutions, winter temperatures have consistently provided suppression of N. fulva populations. Hill (2013) found that satellite nest sites and foraging activity were greatly reduced during the winter. Management strategies that target N. fulva populations during periods of cold could be more effective when N. fulva numbers are low, and nest sites containing reproductives are centralized in smaller areas. Data on the thermal ecology of N. fulva can improve monitoring and management efforts.

The goal of my dissertation is to examine N. fulva’s thermal ecology to better understand how temperature might impact their success as an invasive species. Specifically, studies will include: 1.) Determining N. fulva’s thermal tolerance at both upper and lower critical limits at three thermal ramping rates for various sized workers of N. fulva and S. invicta, 2.) Measuring a possible cold avoidance strategy, subterranean tunneling, for N. fulva, and comparing those data

to a known tunneling species, S. invicta, over a range of cool temperatures (15.0 to 22.0 °C), 3.)

Determining the impact of three temperatures (15.0, 25.0 and 35.0 °C) on the number of resources occupied and the total number of ants at resources for N. fulva and S. invicta separately and when confined together; 4.) Measuring the effect of temperature and worker ratio of N. fulva to S. invicta on the number of resource occupied and the total number of ants at resources, and

5.) Evaluating the impact of high and low numbers of N. fulva on S. invicta’s ability to occupy resources. In addition to the thermal ecology studies, N. fulva’s depletion of a diluted formulation of Maxforce™ Quantum ant bait will be measured in the presence and the absence of a sugar-bait (25.0% sucrose solution) to further develop the utilization of this management strategy.

CHAPTER 2 THE THERMAL BREADTH OF NYLANDERIA FULVA (MAYR) (HYMENOPTERA: FORMICIDAE) IS NARROWER THAN THAT OF SOLENOPSIS INVICTA (BUREN) AT THREE THERMAL RAMPING RATES: 1.0, 0.12 AND 0.06 °C

Introduction

Determining the thermal limits of insects can provide valuable insight into how climate conditions like temperature influence the seasonality, daily activity, and distribution of species

(Chown and Terblanche 2007, Jumbam et al. 2008, Verble-Pearson et al. 2015). Investigating the thermal limits of invasive ants may improve estimates of their potential range expansion and quarantine areas. Furthermore, data on thermal limits can be applied to the timing of management tactics for ants, improving their control.

Thermal limits are often quantified using either static or dynamic methods. Static methods involve exposure of an organism to constant temperatures and assessing survival or time to loss of coordinated movement, whereas dynamic methods typically involve ramping the exposure temperature up or down from the starting temperature until the organism loses coordinated muscle movement or a change is observed in respiration (Lighton and Turner 2004,

Terblanche et al. 2007). Dynamic methods are widely used because they are thought to provide a more ecologically relevant estimation of upper (CTmax) and lower (CTmin) critical thermal limits by ramping organisms from permissive to limiting temperatures (Cerdá et al. 1998b, Somero

2005, Rezende et al. 2011). The rate at which organisms are cooled or heated can affect the estimation of critical thermal limits for insects (Terblanche et al. 2007, Rezende et al. 2011). For example, slower cooling rates resulted in lower CTmin values for the fruit fly Drosophila melanogaster (Meigen) (Kelty and Lee 1999) and the aphid Sitobion avenae (F.) (Powell and

Bale 2004), while faster heating rates resulted in higher CTmax values for the tsetse fly Glossina

pallidipes (Austen) (Terblanche et al. 2007) and the Argentine ant, Linepithema humile (Mayr)

(Chown et al. 2009).

Body size can also impact thermal limits. Many ant species have a wide range of body sizes within and between colonies. Smaller-bodied workers of several different ant species were found to have a narrower thermal breadth than their larger-bodied nestmates (Kay and Whitford

1978, Cerdá et al. 1998b, Clémencet et al. 2010). Although body size and thermal ramping rate are known to impact estimates of thermal tolerance, few studies have evaluated the effects of both these variables and the interactions between them in insects.

Solenopsis invicta (Buren), the red imported fire ant, is a polymorphic ant species native to South America that is established throughout the southern U.S. In North Florida, S. invicta colonies experience both air and soil surface temperatures that are cooler than temperatures experienced in their native South American range. Prior research investigating the critical thermal limits of S. invicta evaluated only minor caste workers (Cokendolpher and Phillips

1990). Furthermore, this study did not control for heating and cooling rates. A more comprehensive evaluation of the CTmin and CTmax of minor and major S. invicta workers at several rates of temperature change is needed to understand whether thermal ramping rates impact the critical thermal limits of this species.

Recently, another South American pest ant, Nylanderia fulva (Mayr), has invaded the southern U.S. Nylanderia fulva generates large polydomous and polygynous colonies that have damaged electrical equipment, reduced arthropod abundance, and displaced existing ant species in the U.S. (Meyers 2008, LeBrun et al. 2013). Like S. invicta, N. fulva is established in North

Florida where seasonal air and soil surface temperatures are lower than temperatures in its native range. Nylanderia fulva is a monomorphic ant species, yet workers do exhibit variance in body-

size that could affect thermal tolerance. Currently, there are no studies on the thermal limits of N. fulva. Therefore, the objectives of this study were to evaluate the impacts of ramping rate and body size on the critical thermal limits of N. fulva and S. invicta.

Materials and Methods

Collection and Maintenance of Field Colonies

Nylanderia fulva and S. invicta colonies were field collected (September to November

2014) and maintained as described in Bentley et al. (2015). Briefly, N. fulva were collected in

Alachua and Duval Counties, FL by extracting workers, male alates, queens, and brood from leaf litter and other debris. Reference samples of all castes are maintained at the Florida Department of Agriculture and Consumer Service Division of Plant Industry, Gainesville Florida.

Polygynous S. invicta colonies were collected in Alachua County, FL using a modified drip- flotation method to extract ants from dirt mounds as described by Banks et al. (1981).

All colonies were maintained in the laboratory at approximately 55.0 ± 8.0% RH, 27.0 ±

3.0 °C, and 12:12 h L:D photoperiod. Polystyrene petri dishes (100.0 × 15.0 mm for N. fulva, and 150.0 × 15.0 mm for S. invicta) partially filled with dental plaster (Castone® Dental Stone,

Dentsply International, York, PA, USA) and moistened with deionized water were used as artificial nests. Live termites or housefly maggots were provided as a protein source every other day, and glass test tubes filled with deionized water or 20.0% sucrose solution and plugged with cotton were provided as moisture sources daily.

Critical Thermal Limits

Body size has been shown to impact the thermal tolerance of ants (Kay and Whitford

1978, Hahn et al. 2008). Therefore, ants from four colonies of N. fulva and four colonies of S. invicta were collected and separated into “small” and “large” body size categories for evaluation based on visual estimation. One large or one small ant from a single colony for each species was

placed into a PVC-pipe arena (2.54 cm tall, 2.50 cm I. D.) interiorly coated with Fluon® and was assigned to each rate of temperature change in a randomized complete block design for a total of

48 experimental units (4 colonies x 3 rates of temperature change x 2 body sizes x 2 species).

Arenas were arranged in a 2 x 2 pattern on an aluminum heat-exchange block attached to a

NesLab RTE 740 digital refrigerated bath (Thermo Fisher Scientific, Waltham, MA). Ants were held in arenas for 20 min. at 26 °C to acclimate before thermal plate surface temperatures were increased or decreased. Thermal plate surface temperatures were monitored using a digital type-t thermocouple fitted with two leads that were affixed to either end of the thermal plate. After ants reached critical thermal limits, the head capsule width of each ant was measured as a metric of body size in mm using a LEICA stereomicroscope (Leica GZ7, Leica Microsystems AG,

Germany) fitted with a calibrated eye-piece micrometer.

The starting temperature for all critical thermal limit experiments was 26.0 °C. Once ants were acclimated, temperatures were increased or decreased at 1.00, 0.12, or 0.06 °C min-1 until the CTmin or CTmax for all ants were observed. The CTmin was recorded as the temperature at which locomotion and response to probing ceased for each ant. The CTmax was recorded as the temperature at which muscle spasms occurred (Lutterschmidt and Hutchison 1997). The 1.0 °C min-1 rate of temperature change is a standard for thermal sensitivity research (Salt 1966).

However, it is widely recognized that 1.0 °C min-1 is too fast to be considered ecologically relevant. Therefore, the rates of 0.06 and 0.12 °C min-1 were added to evaluate more ecologically relevant rates of temperature change (Terblanche et al. 2007, Terblanche et al. 2011).

Data Analysis

Critical thermal limit data met the assumptions of analysis of variance (ANOVA). One

CTmin value (N. fulva, 8.9 °C) was identified as an outlier using boxplots and was removed.

Upper and lower critical thermal limit values were evaluated using a mixed model ANOVA

(JMP version 9.0.2; SAS Institute, 2014) where thermal ramping rate, species, and head capsule width were the fixed effects, and colony was used as a random blocking factor. Tukey’s Honest

Significant Difference (HSD) Test (α = 0.05) was used to separate means for critical thermal limits of each species at each rate of temperature change. Simple linear regression was used to test for a relationship between head capsule width and critical thermal limit values for each species by plotting the mean CTmin and CTmax against head capsule width by thermal ramping rate for N. fulva and S. invicta.

Results

Overall, N. fulva had a narrower thermal breadth than S. invicta (Table 2-1, Figure 2-1).

The CTmin for N. fulva averaged 3.12 °C ± 0.11 °C (± SE) higher than that for S. invicta (Table

2). The CTmax for N. fulva averaged 4.04 °C ± 0.92 °C lower than that for S. invicta. For both species, slower warming rates resulted in lower CTmax values (Table 1). The mean CTmax for N. fulva at 0.06 °C min-1 (38.01 ± 0.76 °C) was 7.87 °C lower than at 1.0 °C min-1 (45.88 ± 1.29

-1 °C). Similarly, the mean CTmax for S. invicta at 0.06 °C min (41.85 ± 0.51 °C) was 8.75 °C

-1 lower than at 1.0 °C min (50.61 ± 0.45 °C). Cooling rate had no effect on CTmin in either species.

Overall, workers of N. fulva had a narrower range of head capsule widths (5.00 to 7.55 mm) than workers of S. invicta (5.50 to 8.55 mm) (Fig. 2). For N. fulva, workers with larger head capsules had lower CTmin values than workers with smaller head capsules at the 0.06 and 1.0 °C min-1 thermal ramping rates while workers of S. invicta with larger head capsules had lower

CTmin values than S. invicta workers with smaller head capsules at all thermal ramping rates.

Conversely, workers of both species with larger head capsules had higher CTmax values than smaller workers of the same species at the 0.06 and 1.0 °C min-1 ramping rates. At the 1.0 °C

-1 min ramping rate, the positive effect of head capsule with on CTmax was greater for N. fulva than S. invicta.

Discussion

Nylanderia fulva had a narrower thermal breadth than S. invicta, with higher CTmin values and lower CTmax values than S. invicta at all three ramping rates (Fig. 1). The overall thermal breadth we recorded for S. invicta (CTmin = 4.06 °C, CTmax = 45.31 °C) was wider than the thermal breadth reported by Cokendolpher and Phillips (1990) (CTmin = 6.30 °C, CTmax = 41.6

°C) when ant colonies were maintained at similar temperatures. Although both studies estimated the critical thermal limits of S. invicta, Cokendolpher and Phillips (1990) focused on minor caste workers only and did not control thermal ramping rates during their experiments. Previous studies have demonstrated that both body size and ramping rate can affect estimates of critical thermal limits in insects (Terblanche et al. 2007, Terblanche et al. 2011). Thus, differences in our results from those reported by Cokendolpher and Phillips (1990) likely reflect differences in methodology between the two studies.

-1 Slower heating rates (0.06 and 0.12 °C min ) resulted in significantly lower CTmax values for both N. fulva and S. invicta compared to the fastest heating rate of 1°C min-1 (Figure 2-1).

The results of our CTmax studies appear contradictory to other study conclusions that increasing exposure time to higher temperatures as a result of slower heating rates should increase heat shock resistance and improve thermal tolerance (Hoffmann et al. 2003, Powell and Bale 2006).

Other ramping experiments have produced results similar to ours with insects including the tsetse fly G. pallidipes (Terblanche et al. 2007), the fruit fly D. melanogaster (Chown et al. 2009), and the Argentine ant, L. humile (Chown et al. 2009). These studies suggest that in some insects, the cost to reaching CTmax increases as the rate of temperature change decreases, and the CTmax decreases as animals accumulate heat damage (Chown et al. 2009, Rezende et al. 2011).

Additionally, other uncontrollable variables such as water loss, nutrition, and energy expenditure due to higher metabolism can have increasing negative effects on performance as warming rates decrease (Rezende et al. 2011). Our experiments ran from ~10 min. at the fastest ramping rate of

1.0 °C min-1 to >20 h for the slowest ramping rate of 0.06 min-1 and support the findings by

Chown et al. (2009) and Rezende et al. (2011) that faster ramping rates may minimize the impacts of uncontrolled variables. Thus, more consideration should be given to the duration of exposure in ramping experiments.

In general, ants with larger head capsules were more resistant to both heat and cold exposures than ants with smaller head capsules (Figure 2-2). Specifically, workers with larger head capsules had higher CTmax values and lower CTmin values overall than smaller headed workers in both species. Cerdá et al. (1997) similarly observed higher CTmax values for larger bodied workers of several Mediterranean ant species versus smaller bodied workers, using both body length and head capsule width as metrics of body size. Kay and Whitford (1978) also reported higher CTmax values and lower CTmin values for larger bodied workers than smaller bodied workers of five Myrmecocystus ant species. Our data agree with previous studies and may also suggest that the larger body size of S. invicta workers could contribute to its wider thermal breadth when compared to smaller bodied N. fulva workers.

By estimating the critical thermal limits of small and large N. fulva and S. invicta workers at several ramping rates, we have illustrated how both methodological and biological factors can affect the estimation of thermal tolerance in these ant species. Furthermore, we have provided the first estimation of critical thermal limits for N. fulva and S. invicta when the effects of body size and ramping rate are considered. These data may improve our understanding of how temperature affects the distribution and habitat availability of both species, thus improving the predictive

power of geographical distribution models for these invasive ants. Our estimations of N. fulva’s thermal tolerance may also contribute to our knowledge of the seasonal activity patterns of this ant species, including temperature thresholds in which this ant can become active during seasonal temperature changes. These data could then be used to improve the timing of seasonal inspection and treatment efforts of N. fulva, directly improving the seasonal monitoring and management of this invasive pest.

Table 2-1. A mixed-model analysis of variance (ANOVA) summary table evaluating the critical thermal minima (CTmin) and critical thermal maxima (CTmax) for Nylanderia fulva and Solenopsis invicta at three rates of temperature change: 0.06, 0.12, 1.0 °C min-1. Model Source df F P CTmin Whole model 17 16.759 <0.0001 Rate 2 1.639 0.2117 Species 1 429.660 <0.0001 Rate*Species 2 0.4073 0.6692 Head Width 1 22.026 <0.0001 Rate*Head Width 2 0.995 0.3822 Species*Head Width 1 2.363 0.1351 Rate*Species*Head Width 2 0.224 0.7931 Colony (Random) 6 0.456 0.8347 Error 29 Total 46

CTmax Whole model 17 8.427 <0.0001 Rate 2 44.836 <0.0001 Species 1 24.116 0.0031 Rate*Species 2 0.133 0.8748 Head Width 1 5.994 0.0204 Rate*Head Width 2 0.603 0.5538 Species*Head Width 1 2.616 0.1163 Rate*Species*Head Width 2 0.794 0.4612 Colony (Random) 6 1.070 0.4023 Error 30 Total 47

Table 2-2. Critical thermal minima (CTmin) and critical thermal maxima (CTmax) values (± SE) for Nylanderia fulva and Solenopsis invicta at three rates of temperature change: 0.06, -1 0.12, 1.0 °C min . Mean CTmin or CTmax values followed by the same letter are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3, SAS Institute, Cary, NC)]). Thermal Limit Species Rate (°C) min-1 N1 Temperature (°C) (± SE) CTmin Nylanderia fulva 0.06 8 7.13 (± 0.17) a 0.12 7 7.49 (± 0.12) a 1.00 8 6.98 (± 0.24) a Solenopsis invicta 0.06 8 4.48 (± 0.20) a 0.12 8 3.99 (± 0.45) a 1.00 8 3.71 (± 0.59) a

CTmax Nylanderia fulva 0.06 8 38.01 (± 0.76) d 0.12 8 39.95 (± 1.16) cd 1.00 8 45.88 (± 1.29) b Solenopsis invicta 0.06 8 41.85 (± 0.51) cd 0.12 8 43.50 (± 1.34) bc 1.00 8 50.61 (± 0.45) a 1n=the number of ants tested per thermal ramping rate

Figure 2-1. Mean critical thermal values (± SE) for Nylanderia fulva (NF) and Solenopsis invicta (SI) at three rates of temperature change per min. (A) Critical thermal minima (CTmin) values for N. fulva and S. invicta at three cooling rates (0.06, 0.12, 1.00 °C min-1). (B) Critical thermal maxima (CTmax) values for N. fulva and S. invicta at three warming rates (0.06, 0.12, 1.00 °C min-1). Mean values within each figure followed by the same letter are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3, SAS Institute, Cary, NC)]).

Figure 2-2. Regression of critical thermal minima (CTmin) and critical thermal maxima (CTmin) values for workers of Nylanderia fulva (n=8 per thermal ramping rate) and Solenopsis invicta (n=8 per thermal ramping rate) at 0.06 °C (A, B), 0.12 °C (C, D), and 1.0 °C min-1 (E, F) as a function of head capsule size.

CHAPTER 3 TUNNELING PERFORMANCE INCREASES AT LOWER TEMPERATURES FOR SOLENOPSIS INVICTA (BUREN) BUT NOT FOR NYLANDERIA FULVA (MAYR)

Introduction

Biological invasions by invasive organisms can dramatically change the composition and ecology of landscapes (Pimentel et al. 2005). Invasive ants are among the most damaging biological invaders because they can displace native species thus altering the function of entire ecosystems (Holway 1999). When moving into a new range, invasive ants may be challenged by temperatures well above or below those in their native range. Subterranean tunneling is one thermoregulatory strategy used to overcome temperature extremes that is exhibited by many invasive ant species including Solenopsis invicta (Buren).

Solenopsis invicta, the red imported fire ant, is an economically important pest native to

South America that is established throughout the southern U.S. This mound-building ant species constructs elaborate subterranean nests with tunnel systems that can extend horizontally up to 84 m as well as reach 1.5 m vertically below ground (Markin et al. 1975, Cassill et al. 2002, Penick and Tschinkel 2008). Horizontal subterranean tunnels provide S. invicta ready access to its foraging territory, even during poor climactic conditions, such as rain or unfavorable temperatures (Markin et al. 1975). Vertical subterranean shafts allow S. invicta to exploit the temperature-buffering properties of the soil, thus providing a nesting site that maintains relatively stable moisture and temperature conditions during seasonal or sudden weather changes

(Cassill et al. 2002, Penick and Tschinkel 2008).

Recently, another South American ant has invaded the southern U.S., Nylanderia fulva

(Mayr). Unlike S. invicta’s mound-based colony structure, N. fulva generate large polygynous and polydomous colonies that commonly nest in leaf litter and other shallow debris (Zenner-

Polania 1990a). In rural and urban areas of Columbia where N. fulva are established as invaders,

foraging populations have reportedly displaced native ant species and injured livestock (Zenner-

Polania 1994). In the southern U.S., N. fulva colonies have displaced ant species and other arthropods, as well as damaged electrical equipment, landscape goods, and commercial honey bee hives (Meyers 2008, Drees 2009, Harmon 2009, LeBrun et al. 2013).

In Florida, N. fulva has been documented to occur in 26 counties throughout the state.

While southern Florida’s seasonal temperatures are generally mild, Northern and Central Florida can experience seasonal cold, where overnight air temperatures drop below freezing. To avoid exposure to freezing air temperatures, ant species, such as S. invicta, rely on subterranean tunneling as a cold avoidance strategy (Penick and Tschinkel 2008). However, the ways in which

N. fulva may avoid cold stress are poorly understood. Existing literature indicates that N. fulva rely predominately rely on shallow leaf litter and other debris for protection and nest construction in native and non-native ranges (Zenner-Polania 1990a). However, leaf litter typically does not offer much insulation from ambient air temperatures, and thus brood and reproductives housed in these shallow nests may be exposed to dangerously cold temperatures during North Florida winters.

Field observations in North Florida suggest that N. fulva can tunnel below ground. It is known that subterranean tunneling is part of a thermoregulatory strategy for other ant species, particularly the successful invader S. invicta (Penick and Tschinkel 2008). Therefore, we hypothesize that subterranean tunneling may similarly help N. fulva to ameliorate cold stress by buffering colony members from thermal extremes. There are currently no published data regarding N. fulva’s tunneling performance or the potential impact of temperature on N. fulva’s tunneling behavior. Thus, our objective was to evaluate the extent to which temperature affected

N. fulva’s tunneling performance, and compare these thermal effects on tunneling to those in S. invicta, an invasive ant that tunnels extensively.

Materials and Methods

Collection of Field Colonies

Nylanderia fulva were field collected from April to November 2013 as needed, from

Alachua and Duval Counties, FL. Debris containing ants and brood were transported to the laboratory in plastic trays (43.0 × 13.0 × 56.0 cm) coated with Fluon® (Insect-A-Slip, BioQuip,

Rancho Dominguez, CA, USA) to prevent ant escape. Reference samples of male alates, queens, and workers are maintained at the Florida Department of Agriculture and Consumer Service

Division of Plant Industry, Gainesville, FL. Polygynous S. invicta colonies were collected from

Alachua County, FL, by transferring mounds containing ants into a plastic bucket (19.0 L) coated with talc powder to prevent ant escape. Ants were extracted from soil and other debris using a modified drip-floatation method as described by Banks et al. (1981). One Styrofoam cup

(0.7 L) partially filled with paper towel was turned upside down and placed atop the debris within the bucket. As the water level rose within the bucket, ants and brood moved into the cup.

Once the water level was above the debris, the cup containing the colony was placed in a plastic tray as described above.

Colonies were kept in the laboratory for ≤ 2 months at approximately 55.0% ± 8% RH,

27.0 ± 3.0 °C, and 12:12 h L:D photoperiod. Ants were provided with one or more artificial nests constructed of polystyrene Petri dishes (100 × 15 mm for N. fulva, and 150 × 15 mm for S. invicta) partially filled with dental plaster (Castone® Dental Stone, Dentsply International, York,

PA, USA) and moistened with deionized water. Colonies were provided a protein source of live termites, dead crickets, or housefly maggots as available, every other day, as well as ad libitum

access to deionized water and 20% sucrose solution via glass test tubes (16.0 x 150.0 mm) plugged with cotton. Water and sucrose levels were checked daily and replaced as needed.

Two-Dimensional Tunneling Assays and Experimental Design

Ant tunneling performance was evaluated at four temperatures using eight Plexiglas tunneling arenas filled with sand, similar to that described by Puche and Su (Puche and Su 2001).

Each arena was constructed of two sheets of transparent Plexiglas (61.0 × 61.0 cm) separated by four Plexiglas strips (61.0 × 2.5 × 0.3 cm) affixed to the outer margins with screws. One access hole (6.5 cm diameter) was drilled at the center of the top sheet and fitted with an ant release chamber (6.5 cm diameter, 177.0 mL, Delta Plastics, Hot Springs, AR, USA) that allowed ants to be introduced into tunneling arenas as well as provided access to deionized water and 20% sucrose solution via test tubes, and live termites (Figure 3-1).

Ants from four colonies of N. fulva and four colonies of S. invicta were used. Colony fragments (≈1000 workers, 2 queens, ≈1 mL of brood) from a single colony of each species were assigned in pairs to temperature-controlled chambers in a split plot design with temperature as the whole plot factor and species as the sub-plot factor for a total of 32 experimental units (4 colonies × 4 temperatures × 2 species). Colony fragments paired within one chamber were considered one chamber-set. In total, four seven-day trials were conducted.

Before each trial, colony fragments were held with 20% sucrose solution at 15.0, 18.0,

20.0, or 22.0 °C for 24 h to acclimate. This temperature range was selected based upon previous studies indicating S. invicta and N. fulva’s lower thresholds for activity to be approximately

15.0 °C, with increased activity observed at temperatures above 21.0 °C (Porter and Tschinkel

1987, Zenner-Polania 1990a, Hill 2013). Colony fragments were then introduced into tunneling arenas and allowed to tunnel for seven days in the dark. Tunnels were measured daily, ants were

provided termites as a protein source every other day, and test tubes containing deionized water or 20% sucrose solution were replaced as needed.

To record daily tunneling distance, tunnels were traced on the Plexiglas with a dry erase marker (Expo®, Sanford Corporation, Atlanta, GA, USA) using a different colored marker each day. Tunneling arenas were placed over a photographic light box to illuminate tunnels, which improved accuracy of tracing. During the measurement process, individual tunneling arenas were removed from incubators for ≤5 min to minimize thermal change. For each trial, daily tunneling distance per species was summed by temperature to obtain cumulative tunneling data per temperature per species.

Analysis

Cumulative tunneling data for each species and temperature were log-transformed to meet normality assumptions, and analyzed using mixed model analysis (Proc Mixed SAS version

9.3, SAS Institute, Cary, NC, USA) with trial, species, and temperature as fixed effects, and colony and chamber-set as random effects. The interaction term species*temperature was significant, therefore we analyzed the relationship of temperature and tunneling distance separately for each species using a two-way ANOVA where temperature was the fixed effect and colony was the blocking factor. Additionally, least significant differences tests (LSD) (α = 0.05) were used to compare treatment means between species. Data in all figures are presented as arithmetic means.

Results

Overall, N. fulva tunneled significantly less than S. invicta (Figure 3-2, Table 3-1).

Nylanderia fulva tunneled most at higher temperatures, tunneling 101.0% farther at 20.0 °C than at 15.0 °C (Figure 3-2, Table 3-2). In contrast, S. invicta tunneled most at lower temperatures, tunneling 76.0% further at 15.0 °C than at 20 °C.

Nylanderia fulva tunneled an average of 41.7 cm per day over all temperatures. Mean daily tunneling performance for N. fulva more than doubled from day one to day two, but remained nearly constant from day two through day seven (Figure 3-3). Solenopsis invicta tunneled an average of 149.0 cm per day overall. Mean daily tunneling performance for S. invicta decreased marginally from day one to day two, but decreased by more than 500% from day two to day seven with the greatest drop in tunneling performance occurring between days two and three. Overall, N. fulva’s daily tunneling performance was noticeably lower than S. invicta’s at all temperatures (Figures 3-2 and 3-4).

Conclusions

To our knowledge, this is the first study to evaluate the effect of temperature on the tunneling performance of N. fulva and S. invicta. These results support the hypothesis that N. fulva has the capacity to construct subterranean tunnels across a range of temperatures that would occur during seasonally cool periods. One study evaluating the thermoregulatory properties of subterranean tunnels demonstrated that tunnels only 10.0 cm deep could be up to 10.0 °C warmer than ambient air temperatures (Hoshikawa et al. 1988). Similarly, Frouz (Frouz 2000) reported that temperatures only 3.0 cm below soil surface could be up to 9.8 °C warmer than air temperatures. These results combined with our data suggest that N. fulva, like S. invicta, could rely on subterranean tunnels to aid in nest thermoregulation during the process of spreading into more temperate ranges. To adequately examine N. fulva’s use of subterranean tunneling as a means of thermoregulation, further research is needed to measure N. fulva’s capacity for tunneling depth. Furthermore, additional investigation of N. fulva’s tunneling performance at upper and lower thermal limits are required.

Solenopsis invicta’s strong tunneling performance was consistent with previous studies documenting S. invicta’s extensive use of tunneling as a means of foraging, nest construction,

and thermoregulation (Markin et al. 1975, Penick and Tschinkel 2008); however, the negative relationship between temperature and tunneling performance we documented for S. invicta contradicts what is often reported in ants (Mellanby 1939, Porter and Tschinkel 1987, Porter

1988, Sinclair et al. 2003). As ectotherms, ants rely on environmental heat to thermoregulate, thus a positive relationship between temperature and activity is typically observed (Mellanby

1939, Porter and Tschinkel 1987, Porter 1988, Sinclair et al. 2003). Solenopsis invicta’s greater tunneling performance at lower temperatures may be attributed to its use of subterranean tunneling as a thermoregulatory strategy. Deep subterranean nests allow S. invicta to exploit the temperature buffering properties of the soil, allowing ants to shift to regions within the nest with preferred temperatures during seasonal or sudden daily changes (Porter and Tschinkel 1987,

Porter 1988). S. invicta’s increased tunneling performance at 15.0 °C may have been a behavioral response to minimize exposure to an undesirable temperature. To test this hypothesis, further evaluation of S. invicta’s tunneling performance across a thermal gradient is needed.

Solenopsis invicta’s mean daily tunneling performance at day one was nearly 1250% greater than that of N. fulva. However, by day seven, mean daily tunneling performance was almost equal for both ant species. Solenopsis invicta’s considerable decline in daily tunneling performance may have been an effect of the finite amount of available space and sand for ants to excavate within the tunneling arenas. During the first two days of each study, S. invicta were observed excavating long, wide tunnels that quickly reached the edges of the arena and occupied most of the available tunneling space. Over the remaining five days, S. invicta workers were limited in the amount of unoccupied space to construct new tunnels thus excavating smaller, narrower connecting tunnels (Figure 4). These observations may suggest that S. invicta’s decline in daily tunneling activity was a result of the spatial limits of our tunneling arenas and not typical

of S. invicta’s daily tunneling performance in the field. To adequately evaluate this hypothesis, additional investigation of S. invicta’s tunneling performance using larger tunneling arenas would be necessary.

Our data demonstrating N. fulva’s tunneling performance at temperatures as low as

15.0 °C may also improve seasonal monitoring and treatment programs for this pest ant.

Nylanderia fulva populations in North Florida typically reach peak densities from early summer to mid fall, making management of this pest very difficult during this period. One study evaluating N. fulva’s seasonal activity in North Florida found that satellite nest sites and foraging activity were greatly reduced throughout seasonal cold months (Hill 2013). During periods of seasonal cold (≤15.0 °C) it is also likely that N. fulva colonies contract, becoming localized around more permanent nests as seen in other invasive ant species (Heller and Gordon 2006).

Management of N. fulva could be more effective during periods of seasonal cold when N. fulva densities are low, and are centralized in smaller areas where nest sites containing reproductives are present. Currently, management programs for N. fulva involve inspection and treatment methods that predominantly target shallow leaf litter and other debris commonly associated with the nesting habits of this species. Our results demonstrating N. fulva’s tunneling performance suggests that N. fulva can tunnel below leaf litter and possibly even into the soil, potentially sheltering this pest ant from current inspection and treatment methods. Therefore, monitoring and control programs occurring during periods of seasonal cold should incorporate sub-soil inspection and treatment methods to facilitate N. fulva nest site detection to improve N. fulva’s management.

Table 3-1. A mixed model analysis evaluating the tunneling performance of Nylanderia fulva and Solenopsis invicta at 15.0, 18.0, 20.0, and 22.0 °C. Source df Type III F p Trial 3, 9 2.25 0.1514 Species 1, 12 229.79 <0.0001 Temperature 3, 9 0.44 0.7310 Species*Temperature 3, 12 11.00 0.0009

Table 3-2. A two-way ANOVA summary table by species evaluating the tunneling performance of Nylanderia fulva and Solenopsis invicta at 15.0, 18.0, 20.0, and 22.0 °C. Species Source df F p Model 6 3.49 0.0459 Temperature 3 4.80 0.0290 Nylanderia fulva Colony (Random) 3 2.17 0.1619 Error 9 Total 15 Model 6 4.85 0.0177 Temperature 3 6.55 0.0122 Solenopsis invicta Colony (Random) 3 3.15 0.0794 Error 9 Total 15

Figure 3-1. Tunneling arena (61 × 61 cm) constructed of two sheets of transparent Plexiglas (61.0 × 61.0 cm) separated by four Plexiglas strips (61.0 × 2.5 × 0.3 cm) affixed to the outer margins with screws. One access hole (6.5 cm diameter) was drilled at the center of the top sheet and fitted with an ant release chamber.

Figure 3-2. Mean total (±SE) tunneling distances (cm) per temperature for Nylanderia fulva (NF) and Solenopsis invicta (SI). Mean values within species not sharing the same letter are significantly different (p <0.05, LSD standardized test (SAS version 9.3, SAS Institute, Cary, NC, USA)).

Figure 3-3. Mean (±SE) daily tunneling distances (cm) for Nylanderia fulva (Nf) and Solenopsis invicta (Si) over all temperatures.

Figure 3-4. Tunneling performance at 20.0 °C after three days for Nylanderia fulva (A) and Solenopsis invicta (B). Photo courtesy of Michael Bentley.

CHAPTER 4 COMPETITIVE SUCCESS FOR CARBOHYDRATE AND PROTEIN RESOURCES INCREASES AT LOWER TEMPERATURES FOR NYLANDERIA FULVA (MAYR) AND SOLENOPSIS INVICTA (BUREN) WHEN COLONIES ARE EQUAL-SIZED

Introduction

Temperature is among the most influential abiotic factors that affects ants (Porter and

Tschinkel 1987, Cerdá et al. 1998b, Pranschke and Hooper-Bùi 2003). Variations in thermal tolerance among ant species results in temporal partitioning that can shape foraging patterns, reproduction, and species distribution (Porter and Tschinkel 1987, Briere et al. 1999).

Temperature has even been shown to mediate competition between ants by influencing foraging behavior at upper thermal limits of foraging among competing ant species (Cerdá et al. 1998a,

Cerdá et al. 1998b, Fitzpatrick et al. 2014). While most studies have focused on temperatures near the upper thermal threshold of ant foraging, little attention has been given to the impact of lower temperatures on foraging by invasive ants.

Solenopsis invicta (Buren) is an invasive ant native to South America that is established throughout the southern U.S. Since its entry into Mobile, AL (Wilson 1958), S. invicta has become a major ecological pest, displacing native species throughout its non-native range

(Tschinkel 2006). Recently, another pest ant has invaded the southern U.S., Nylanderia fulva

(Mayr). This polydomous and polygynous species generates large populations that have damaged electrical equipment, decimated native ant fauna, and even displaced other invasive ants, such as

S. invicta (Zenner-Polania 1990a, McDonald 2012, LeBrun et al. 2013).

Temperature influences the foraging behavior of N. fulva and S. invicta. Foraging activity for both species increases as temperatures warm in early spring (≥ 15.0 °C), and subside as temperatures cool in late fall (Tschinkel 2006, Hill 2013). However, little is known of the thermal ecology of these species in the context of competition. Temperature may influence

aspects of competition-related performance, such as resource defense and nestmate recruitment.

Thus, we investigated the impact of temperature on the resource occupancy and number of ants at resources for N. fulva and S. invicta in the presence and absence of competition to improve our understanding of the mechanisms associated with the competition between these invasive species.

Materials and Methods

Collection of Field Colonies

Nylanderia fulva and S. invicta were field collected from September to November 2014, and maintained in the laboratory as described in Bentley et al. (2015). Briefly, N. fulva were extracted from organic debris (leaves, sticks, logs) containing ants and brood collected from

Alachua and Duval Counties, FL. Reference samples of all castes are maintained at the Florida

Department of Agriculture and Consumer Service Division of Plant Industry, Gainesville, FL.

Polygynous S. invicta colonies were collected from Alachua County, FL, by extracting ants and brood from excavated mounds using a drip-floatation method modified from Banks et al. (1981).

In the laboratory, colonies were kept in fluoned (Fluon®, Insect-A-Slip, BioQuip, Rancho

Dominguez, CA) plastic trays for ≤ 4 months where they were maintained at approximately 55.0

± 8% RH, 27.0 ± 3.0 °C, and 12:12 h L:D photoperiod. Ants were provided with artificial nests constructed of polystyrene Petri dishes (100 x 15 mm for N. fulva, and 150 x 15 mm for S. invicta) partially filled with dental plaster (Castone® Dental Stone, Dentsply International, York,

PA) and moistened with deionized water. Colonies were fed a protein source of frozen housefly maggots every other day, and were provided ad libitum access to deionized water and 20% sucrose solution via glass test tubes (16.0 x 150.0 mm) plugged with cotton. Tubes were checked daily and replaced as needed.

All Assays

The behavior of N. fulva and S. invicta were evaluated in temperature controlled rooms at three temperatures using one plastic arena (65.0 x 80.0 x 12.0 cm) with the interior sides coated with Fluon® (Figure 4-1). For each assay, ants were given access to sixteen food resources consisting of 0.25 ml of 20% sucrose solution and 0.05 g of water-packed tuna (Bumble Bee

Chunk Light, Bumble Bee Foods, Toronto, ON, Canada) that were arranged in a 4 x 4 grid, with resources of the same type placed in alternating horizontal rows of 4. In the event that any resource was depleted during the study, that resource was replenished with the same amount that was provided at the beginning of the trial. Tuna and 20% sucrose solution were selected because preliminary evaluations indicated these foods were readily consumed by both ant species, and were used simultaneously to enhance feeding during experiments.

Colony fragments containing ≈ 1,000 workers, 2 queens, ≈ 1 ml of brood from a single colony of each species were first starved for 24 h, then held in a temperature controlled room at

15.0, 25.0, or 35.0 °C for one hour prior to evaluation in order to acclimate. Previous studies indicated that workers of N. fulva and S. invicta foraged between 15.0 to 37.0 °C and 15.0 to

42.0 °C, respectively (Porter and Tschinkel 1987, McDonald 2012). Temperatures were selected to bracket upper and lower thermal thresholds for foraging by both species.

In both experiments described below, colony fragments from the same five parent colonies of N. fulva and five parent colonies of S. invicta were assigned to each of the temperatures (15.0, 25.0, and 35.0 °C) in a randomized complete block design, blocking on colony, for a total of 30 experimental units (5 colonies x 3 temperatures x 2 species).

Impact of Temperature on Foraging Behavior without Competition

Either one colony fragment of N. fulva or S. invicta was placed at one end of the foraging arena and allowed access to the food resources. An occupied food resource was defined as any

resource at which ants were contacting the food. The number of resources occupied by ants and the total number of ants on all food resources were counted at 1 min intervals for 20 min. Counts were then averaged every 5 min. That is, for each 5 min interval, an average point was calculated and labeled as 5, 10, 15, and 20 min averages. Categories identified as “5” included counts taken from 1 to 5 min, “10” for counts take from 6 to 10 min, “15” for counts taken from 11 to 15 min, and “20” for counts taken from 16 to 20 min.

Impact of Temperature on Foraging when Forced to Compete for Resources

This experiment was set up to measure N. fulva’s response to a competitor, S. invicta, at resources. Solenopsis invicta was designated as the treatment. First, a colony fragment of N. fulva was placed at one end of the foraging arena and allowed access to food resources for 10 min without S. invicta present, allowing ants to acclimate. During the acclimation period, the number of resources occupied and the number of ants at resources were recorded at one min intervals for 10 min. A colony fragment of S. invicta was then introduced at the opposite end of the arena and data collection continued at one min intervals for 20 more min. Data were placed into 5 categories as described above: 0, 5, 10, 15, and 20 min averages. Point “0” is defined as when the competitor was introduced into the arena. Data at point “0” is an average of all counts for N. fulva during the 10 min before the competitor S. invicta was introduced.

To measure S. invicta’s response to a competitor, the experiment was repeated with S. invicta given first access to food resources for 10 min before the competitor N. fulva was introduced. Nylanderia fulva was designated as the treatment.

Analysis

Repeated measures analyses of variance (ANOVA) were conducted to examine the effects of temperature and species on numbers of resources occupied and numbers of ants at resources in both no-competition and competition experiments using mixed model procedures

(Proc Mixed; SAS version 9.3, SAS Institute, Cary, NC, USA). For the no-competition study, temperature, species, and resource type were the fixed effects, and the random effect of colony was the blocking factor.

For the competition study, two separate analyses were conducted. For each analysis, temperature, species, and resource type were the fixed effects, , and the random effect of colony was the blocking factor. The competitor was designated as the treatment. In one analysis, N. fulva was the competitor and in the second analysis S. invicta was the competitor.

All repeated measures analyses were adjusted for potential correlation due to sampling the same experimental unit over time by fitting a first-order autoregressive heterogeneous covariance structure (AR(1)) and using the Kenward-Roger correction for degrees of freedom.

The AR(1) covariance structure is used in mixed model analyses because correlations between repeated measurements decrease as measurements are more distantly separated by time. To account for non-independence of the data, the AR(1) covariance structure calculates a different error term for each time point. The Kenward-Roger correction is applied to the denominator degrees of freedom, and is preferred in mixed model analyses when an experimental design is balanced, a covariance structure is specified, and sample size is moderate to small (Kenward and

Roger 2009). As part of the repeated measures analyses, Tukey’s Honest Significant Difference

(HSD) Test (α = 0.05) was used in conjunction with simple effects tests (slice statement in SAS) when interactions were significant (Table 4-1). The “slice” statement was used to compare means between species, and holds one factor constant while comparing means of another factor.

SAS code for this chapter is included in the Appendix.

Results and Discussion

Without Competition

The species*temperature*time interaction was significant for the number of resources occupied but not number of ants on resources (Table 4-1). Figure 4-2 illustrates the relationship of the interactions. The bait type*time and temperature*time interactions combined data from both species, so they did not yield useful information for each species, and thus they are not addressed. Hereafter, any interactions that did not yield information for each species is not addressed.

In the absence of competition, N. fulva generally occupied more of both sucrose and tuna resources within five min of being introduced into the arena than they did in the final five min of the assay (Figures 4-2). The number of sucrose resources occupied by N. fulva declined significantly over the duration of the study at 15.0 and 25.0 °C, but the biological relevance is equivocal due to the small change in the absolute numbers of resources occupied (~2 resources occupied at 15.0 and 25.0 °C). There was no significant difference in tuna resources occupied by

N. fulva over time at any temperature (Figure 4-2A-C). The number of resources occupied by S. invicta remained unchanged over the duration of the study for all temperatures.

There was no significant difference between sucrose and tuna resources occupied by N. fulva at any temperature (Figures 4-2A, 4-2B, 4-2C). Lack of significance may be because the tuna also contained water. At various times, S. invicta occupied significantly more sucrose resources compared with tuna at both 15.0 and 25.0 °C, while at 35.0 °C, there was no significant difference (Figures 4-2A, 4-2B, 4-2C).

The species*time and species*temperature interactions were significant for numbers of ants on resources (Table 4-1). The number of N. fulva at tuna resources declined significantly at the beginning of the trial at 15.0 °C (Figure 4-2D), but there were no other significant differences

in the numbers of N. fulva at resources for any of the other treatments. The number of S. invicta at both sucrose and tuna resources did not significantly change over the duration of the study for all temperatures.

There was no significant difference between the number of N. fulva at sucrose or tuna resources at any temperatures, and no significant difference in the number of S. invicta at either resource type except for the 15 °C treatment (Figure 4-2D). Again, the lack of significance could be partially due to the confounding factor of water in the tuna resource. It is interesting to note that the number of N. fulva at resources at the 5 min average is 3 to 4 times more at 35 °C compared with 25 °C, but the number of resources occupied at 5 min appears to be similar for all temperatures (Figure 4-2 A-C).

Nylanderia fulva is a fast-moving ant species that dominate resources in non-native ranges by quickly locating and heavily recruiting nestmates to resources (Zenner-Polania 1990a,

McDonald 2012). Solenopsis invicta move slower than N. fulva, but are also capable of resource dominance (Porter and Savignano 1990, Tschinkel 2006). Our data support previous research that N. fulva’s quick dispersal throughout the foraging arena led to faster resource occupation and a greater number of ants at resources than S. invicta in the first 5 min of being introduced without competition at 25.0 and 35.0 °C. Additionally, these data also agree with previous research suggesting that S. invicta may be slower to locate and recruit to resources than N. fulva, but remain at resources longer.

In the Presence of Competition

The introduction of a competitor into the foraging arena resulted in a greater decrease in sucrose and tuna resources occupied and number of ants per occupied resources for N. fulva than for S. invicta (Figures 4-3, 4-4). When N. fulva was introduced into the foraging arena first, the species*temperature*time interaction was significant for resources occupied (Table 4-2, Figure

4-3). At both 25.0 and 35.0 °C, the number of sucrose and tuna resources occupied by N. fulva declined by >74% within five min after the competitor was introduced (Figure 4-3B, 4-3C). Not unexpectedly, at 25.0 and 35.0 °C, N. fulva per occupied resources also declined by almost 100% five min after the competitor was introduced (Figure 4-3E, 4-3F). The species*temperature*bait type*time interaction for ants per occupied resource was significant, but can be largely attributed to the response of ants at the 15 °C treatment where the number of occupied baits remained stable at ≥5 min (Figures 4-3D-F).

When S. invicta was introduced into the foraging arena first, with N. fulva as the competitor, the species*temperature*time interaction also was significant for resources occupied

(Table 4-3, Figure 4-4). There was no significant difference in resources occupied by S. invicta across time categories at 15.0 or 25.0 °C (Figures 4-4A, 4-4B), but there was a significant decrease in resources occupied and number of S. invicta at resources at 35 °C. Within 10 min of the competitor (N. fulva) being introduced, resources occupied by S. invicta declined by >62.0%

(Figure 4-4C). There were also significant differences between sucrose and tuna resources occupied and number of S. invicta at resources across various times and temperatures (Figure 4-4

A-C).

Considerable evidence suggests that interference competition plays an important role in determining the success of ant species whose resource requirements overlap (Adams and

Traniello 1981, Human and Gordon 1996, Lach 2005). Human and Gordon (1995) and

Carpintero and Reyes-López (2008) demonstrated that the presence of L. humile at baits reduced the foraging success of native ant species competing for resources. Our data similarly demonstrated that the presence of a competitor (S. invicta) at resources suppressed the number of resources occupied and the number of N. fulva per occupied resource at warmer temperatures

(25.0 and 35.0 °C), but when S. invicta was first to resources at cooler temperatures (15 °C) introducing N. fulva as a competitor had little impact. These findings may suggest that interference competition could favor S. invicta in establishing populations earlier in a season when temperatures are lower (~15.0 °C). Results of cold tolerance studies (Chapter 2) support S. invicta’s ability to withstand lower temperatures than N. fulva, supporting this hypothesis.

Other studies investigating the impact of temperature on interspecific competition within ant communities have reported an increase in aggressive behavior between ants as ambient temperatures approached the upper thermal limits of foraging among competing ant species

(Cerdá et al. 1998a, Cerdá et al. 1998b, Fitzpatrick et al. 2014). However, these studies focused little attention on resource competition at lower thermal thresholds. We similarly observed an increase in aggression (biting and/or producing defensive chemicals) between N. fulva and S. invicta at warmer temperatures that led to a greater number of ants fighting rather than feeding.

With more ants of both species engaged in aggressive interactions, fewer ants were left to forage, likely explaining the lower number of resources occupied and lower number of ants at resources recorded at 25.0 and 35.0 °C. In contrast at 15.0 °C, we observed a reduction in aggression between N. fulva and S. invicta where workers foraged with little interaction between the species, likely explaining the overall greater number of resources occupied and number of ants at resources recorded at cooler temperatures.

In no-competition and competition assays for both ant species, when counts of resources occupied and ants at resources differed significantly between resource types, counts were always greater for sucrose than tuna (Figures 4-2, 4-3, 4-4). These results were most often observed for

N. fulva at 15.0 °C and for S. invicta at 15.0 and 25.0 °C. Previous experiments investigating diet preference for N. fulva reported that workers preferred mint jelly (carbohydrate) to hotdog

(protein) baits in spring and fall when baits were shaded from the sun, but preferred shaded hotdog to mint jelly baits in the summer (McDonald 2012). Similar experiments evaluating diet preference for S. invicta reported a greater number of workers recruited to grape jelly

(carbohydrate) baits during winter and spring months when ground temperatures were cooler

(approximately 17.0 °C) while more workers recruited to cat food (protein) baits in summer and fall months when ground temperatures were warmer (approximately 25.0 °C) (Stein et al. 1990).

Our data for N. fulva and S. invicta agree with previous findings, further suggesting that both species may prefer carbohydrates during cooler months when seasonal temperatures are lower, and proteins during warmer seasons likely due to the greater presence of brood which require protein for development (Cassill and Tschinkel 1999).

In several locations, N. fulva has displaced S. invicta (LeBrun et al. 2013). The exact mechanisms of this displacement are still under investigation, but recent research suggests that

N. fulva’s production of chemicals that can detoxify S. invicta’s venom provides N. fulva an advantage (Chen et al. 2013, LeBrun et al. 2014). Collectively, our research demonstrating the effects of temperature and competition on the resources occupied and number of ants at resources for N. fulva and S. invicta illustrate a fundamental difference in the foraging strategies of both species that could also contribute to N. fulva’s displacement of, and the eventual return of

S. invicta. These data suggest that, during warmer periods, N. fulva may rely on faster resource location and strong nestmate recruitment to exploit and retrieve resources before the arrival of competitors, including S. invicta, rather than aggressively displacing other ants already present at resources. In contrast, S. invicta’s ability to persist on resources at lower temperatures in the presence of competition may give it a seasonal competitive advantage over other ants. While our laboratory data support this hypothesis, assays were conducted for only 20 min and may not have

allowed enough time to observe if prolonged competitive dominance were possible for either species. At the end competition assays, N. fulva and S. invicta were still engaged in aggressive interactions, with neither species clearly dominating the other.

Lastly, our data illustrating the effect of temperature on the number of ants per occupied resource for N. fulva and S. invicta may also improve our understanding of how temperatures can influence the foraging success of both species. Previous studies have demonstrated that ant species with greater nestmate recruitment than their competitors were more successful when foraging and competing for resources (Fellers 1987, Adams 1990). The higher number of resources occupied by N. fulva and greater number of N. fulva at resources at warmer temperatures suggests that N. fulva may locate and recruit nestmates to resources more quickly than S. invicta during warm seasons. Conversely, S. invicta’s greater number of resources occupied and higher number of S. invicta at resources at lower temperatures may indicate S. invicta’s ability to exploit resources more effectively than N. fulva during cooler seasonal periods. Improving our understanding how seasonal temperatures can influence the foraging success of N. fulva and S. invicta could identify ideal periods to implement inspection and baiting programs, thereby improving our management efforts of these invasive ants.

Table 4-1. Repeated measures analyses of variance (ANOVA) table for the main effect of temperature (15.0, 25.0, and 35.0 °C) and resource on the number of resources occupied and the number of ants at resources for Nylanderia fulva and Solenopsis invicta without competition. Response Variable Source df F P Resources Occupied Time 3 0.47 0.7055 Colony 4 0.73 0.5781 Species 1 2.08 0.1562 Temperature 2 1.34 0.2719 Bait Type 1 7.56 0.0088 Species*Temperature 2 0.21 0.8096 Species*Bait 1 3.22 0.0801 Colony*Time 12 1.54 0.1186 Species*Time 3 28.46 <0.0001 Temperature*Time 6 7.86 <0.0001 Bait Type*Time 3 6.80 0.0003 Species*Temperature*Time 6 3.80 0.0018 Species*Bait Type*Time 4 1.56 0.1950 Species*Temperature*Bait Type*Time 16 1.10 0.3607

Ants at Resources Time 3 2.89 0.0407 Colony 4 2.34 0.0704 Species 1 1.03 0.3159 Temperature 2 3.06 0.0576 Bait Type 1 1.14 0.2926 Species*Temperature 2 4.81 0.0132 Species*Bait 1 2.00 0.1650 Colony*Time 12 0.34 0.9791 Species*Time 3 13.78 <0.0001 Temperature*Time 6 0.47 0.8297 Bait Type*Time 3 6.52 0.0005 Species*Temperature*Time 6 1.82 0.1032 Species*Bait Type*Time 4 0.97 0.4301 Species*Temperature*Bait Type*Time 16 0.86 0.6105

Table 4-2. Repeated measures analyses of variance (ANOVA) table for the main effect of temperature (15.0, 25.0, and 35.0 °C) and bait type on the number of resources occupied and the number of ants at resources when the competitor was Solenopsis invicta. Response Variable Source df F P Resources Occupied Time 4 6.55 0.0001 Colony 4 8.40 <0.0001 Species 1 7.68 0.0080 Temperature 2 12.96 <0.0001 Bait Type 1 6.54 0.0138 Species*Temperature 2 1.15 0.3251 Species*Bait 1 0.53 0.7110 Colony*Time 16 2.28 0.0055 Species*Time 4 54.11 <0.0001 Temperature*Time 8 16.68 <0.0001 Bait Type*Time 4 0.53 0.7110 Species*Temperature*Time 8 2.73 0.0085 Species*Bait Type*Time 5 1.73 0.1361 Species*Temperature*Bait Type*Time 20 1.21 0.2580

Ants at Resources Time 4 24.20 <0.0001 Colony 4 5.21 0.0016 Species 1 5.79 0.0205 Temperature 2 7.92 0.0012 Bait Type 1 5.26 0.0268 Species*Temperature 2 1.36 0.2676 Species*Bait 1 0.03 0.8645 Colony*Time 16 2.72 0.0009 Species*Time 4 74.39 <0.0001 Temperature*Time 8 23.17 <0.0001 Bait Type*Time 4 0.63 0.6437 Species*Temperature*Time 8 6.38 <0.0001 Species*Bait Type*Time 5 4.50 0.0012 Species*Temperature*Bait Type*Time 20 2.52 0.0010

Table 4-3. Repeated measures analyses of variance (ANOVA) table for the main effect of temperature (15.0, 25.0, and 35.0 °C) and bait type on the number of resources occupied and the number of ants at resources when Nylanderia fulva was the competitor. Model Source df F P Resources Occupied Time 4 5.58 0.0004 Colony 4 1.80 0.1451 Species 1 47.07 <0.0001 Temperature 2 14.51 <0.0001 Bait Type 1 8.57 0.0053 Species*Temperature 2 6.49 0.0033 Species*Bait Type 1 3.49 0.0682 Colony*Time 16 1.44 0.1316 Species*Time 4 8.31 <0.0001 Temperature*Time 8 2.89 0.0056 Bait Type*Time 4 0.70 0.5954 Species*Temperature*Time 8 2.07 0.0441 Species*Bait Type*Time 5 1.06 0.3881 Species*Temperature*Bait Type*Time 20 1.28 0.2039

Ants at Resources Time 4 2.91 0.0255 Colony 4 1.40 0.2522 Species 1 23.68 <0.0001 Temperature 2 3.59 0.0368 Bait Type 1 4.21 0.0468 Species*Temperature 2 2.60 0.0868 Species*Bait Type 1 3.61 0.0647 Colony*Time 16 1.05 0.4064 Species*Time 4 2.94 0.0242 Temperature*Time 8 1.20 0.3080 Bait Type*Time 4 0.58 0.6768 Species*Temperature*Time 8 0.92 0.5054 Species*Bait Type*Time 5 0.85 0.5184 Species*Temperature*Bait Type*Time 20 0.59 0.9149

Figure 4-1. Diagram of the foraging arena used to evaluate the number of resources occupied and number of ants at resources for Nylanderia fulva and Solenopsis invicta at 15.0, 25.0 and 35.0 °C with or without competition with the opposite species. Triangles represent the placement locations of resources (0.25 ml of 20% sucrose solution or 0.05 g of water-packed tuna) that were arranged in alternating horizontal rows by resource type. Ant placement locations were sites where ants were introduced into foraging arenas.

Figure 4-2. In the absence of competition, mean number (±SE) of resources occupied by ants (A- C) and mean number of ants at resources (D-F) when Nylanderia fulva (n = 5 per temperature) and Solenopsis invicta (n = 5 per temperature) were placed separately in an area with two resources at 15.0, 25.0 and 35.0 °C. Means for each species followed by the same letter in figures A, B, and D are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3, SAS Institute, Cary, NC)]). An asterisks (*) indicates significant differences between resource types for a species at a given time point (P ≤ 0.05, ANOVA [SAS version 9.3, SAS Institute, Cary, NC)]).

Figure 4-3. Mean number (±SE) of resources occupied by ants (A-C) and of ants at resources (D- F) when Nylanderia fulva (n = 5 per temperature) and competitor S. invicta were placed in an arena with two resources at 15.0, 25.0 and 35.0 °C. Time “0” is defined as the time at which the competitor was introduced into the arena Means followed by the same letter for each bait type (upper case letters = sucrose, lower case letters = tuna) and species over time are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3, SAS Institute, Cary, NC)]). An asterisks (*) indicates significant differences between resource types per species at a given time point (P ≤ 0.05, ANOVA [SAS version 9.3, SAS Institute, Cary, NC)]).

Figure 4-4. Mean number (±SE) of resources occupied by ants (A-C) and of ants at resources (D- F) when Solenopsis invicta (n = 5 per temperature) and competitor Nylanderia fulva (n = 5 per temperature) were placed in an arena with two resources at 15.0, 25.0 and 35.0 °C. Time “0” is defined as the time at which the competitor was introduced into the arena. Means followed by the same letter for each bait type (upper case letters = sucrose, lower case letters = tuna) and species over time are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3, SAS Institute, Cary, NC)]). An asterisks (*) indicates significant differences between resource types per species at a given time point (P ≤ 0.05, ANOVA [SAS version 9.3, SAS Institute, Cary, NC)]).

CHAPTER 5 NUMERICAL ADVANTAGE IMPROVES COMPETITIVE SUCCESS FOR RESOURCES AT LOWER TEMPERATURE FOR NYLANDERIA FULVA (MAYR) AND AT HIGHER TEMPERATURES FOR SOLENOPSIS INVICTA (BUREN)

Introduction

In previous experiments where N. fulva and S. invicta were forced to compete for resources at different temperatures at a 1:1 ratio, N. fulva occupied more resources at higher temperatures while S. invicta occupied more resources at lower temperatures (Chapter 4).

However, the size of field colonies of N. fulva and S. invicta can be highly variable over time and may rarely be equal to one another. S. invicta colonies typically increase as air temperatures warm in early spring (≥17.0 °C), and decrease as temperatures cool again in late fall (Porter

1988). There is ample evidence that N. fulva colonies also increase as temperatures warm, but there are no quantitative studies published at this time.

Fluctuations in colony size can influence the numerical advantage one species may have over another in resource defense (Holway 1999, Holway and Case 2001, Barbieri et al. 2014).

Currently, there is no research that measures the impact of worker ratios and temperatures on competition for resources. Therefore, our objective was to quantify the number of workers of N. fulva and S. invicta at sucrose (carbohydrate) and protein (tuna) resources at three different temperatures as a measure of resource dominance.

Materials and Methods

Collection of Field Colonies

Nylanderia fulva and S. invicta were collected from September to November 2014 and maintained as described in Bentley et al. (2015). Briefly, N. fulva were extracted from leaf litter, fallen branches, and other debris collected from Alachua and Duval Counties, FL. Reference samples of male alates, queens, and workers are maintained at the Florida Department of

Agriculture and Consumer Service Division of Plant Industry, Gainesville, FL. Polygynous S. invicta colonies were collected from Alachua County, FL, by extracting ants and brood from soil and other debris using a modified drip-floatation method as described by Banks et al. (1981).

Colonies were kept in plastic trays with interior sides coated with Fluon® (Insect-A-Slip,

BioQuip, Rancho Dominguez, CA), and maintained in the laboratory for ≤4 months at approximately 55.0 ± 8% RH, 27.0 ± 3.0 °C, and 12:12 h L:D photoperiod. Ants were provided with artificial nests (described in Chapter 2), a protein source of frozen housefly maggots every other day, and ad libitum access to deionized water and 20% sucrose solution.

Impact of Temperature and Worker Ratio When Competing for Resources

This study was designed to investigate the response of N. fulva and S. invicta at resources when worker ratios of both species varied at 15.0, 25.0, and 35.0 °C. For each assay, ants were given access to sixteen food resources consisting of 0.25 ml of 20% sucrose solution or 0.05 g of water-packed tuna (Bumble Bee Chunk Light, Bumble Bee Foods, Toronto, ON, Canada) as described in detail in Chapter 4 (Figure 5-1). In the event that a resource was depleted during the study, that resource was replenished with the same amount that was provided at the beginning of the trial.

Colony fragments of N. fulva contained 250, 500, or 1,000 workers, and 2 queens with

~1.0 ml of brood. All colony fragments of S. invicta contained 500 workers, and 2 queens with

~1.0 ml of brood. Colony fragments were combined in the following ratios of N. fulva to S. invicta: 250:500, 500:500, and 1,000:500. Colony fragments collected from a different colony of each species were first starved for 24 h, then held for one hour in a temperature controlled room at 15.0, 25.0, or 35.0 °C to acclimate. Temperatures were selected to bracket upper and lower thermal thresholds for foraging by both species (Porter and Tschinkel 1987, McDonald 2012).

The first assay measured N. fulva’s response to a competitor, S. invicta, at resources when worker ratios were manipulated. Solenopsis invicta was designated as the treatment. First, a colony fragment of N. fulva containing 250, 500, or 1000 workers was placed at one end of the foraging arena and allowed access to food resources for 10 min without S. invicta present, allowing ants to discover baits. The number of resources occupied and the total number of N. fulva at resources were recorded at one min intervals for 10 min. Next, the S. invicta colony fragment was introduced at the opposite end of the foraging arena and data collection continued at one min intervals for 20 more min. For graphing purposes, data were placed into 5 categories

(0, 5, 10, 15, and 20 min average) as described in detail in Chapter 4. Briefly, point “0” is defined as when the competitor S. invicta was introduced into the arena. Data at point “0” is an average of all counts for N. fulva during the 10 min before the competitor was introduced. For the remaining categories, data for each species were averaged over 5 min intervals.

To measure S. invicta’s response to a competitor, the experiment was repeated with 500

S. invicta given access to food resources for 10 min before a colony fragment of N. fulva containing either 250, 500, or 1000 workers was introduced. Nylanderia fulva was designated as the treatment. Data were collected and categorized as described above. Colony fragments from the same six parent colonies of N. fulva and six parent colonies of S. invicta were assigned to each temperature and each worker ratio treatment in a randomized complete block design, blocking on colony, for a total of 108 experimental units (6 colonies x 3 temperatures x 3 worker ratios x 2 species as treatments).

Analysis

Two separate analyses of variance (ANOVA) were conducted on cumulative numbers of resources occupied and cumulative numbers of ants over the 30 minute test using general linear models procedures (Proc GLM SAS version 9.3, SAS Institute, Cary, NC, USA). For each

analysis, species, temperature, worker ratio, and resource type were fixed effects. The competitor species was considered treatment. In one analysis, N. fulva was the competitor and in the second analysis S. invicta was the competitor. Colony was a random effect and blocking factor.

Appropriate simple effects tests were done when interactions were significant. When appropriate, means were separated used Tukey’s Honest Significant Difference (HSD) Test (α =

0.05) (SAS version 9.3, SAS Institute, Cary, NC, USA).

Results and Discussion

Our findings show that worker ratio and temperature influence the cumulative number of resources occupied and the cumulative number of ants per resource (Tables 5-1 to 5-6). I observed that these factors also affect the aggression of N. fulva and S. invicta. The order in which the competitor is introduced impacted the outcome of competition trials. The first species to resources usually had the advantage over the second (Tables 5-1, 5-4).

250, 500, or 1000 N. fulva Vs a Competitor (500 S. invicta)

When N. fulva was introduced first, the Species*Ratio*Bait interaction for the cumulative number of resources occupied, and Species*Ratio and Species*Temperature interactions for cumulative number of ants at resources were significant (Table 5-1; Figures 5-2 to 5-4).

Within 10 min of the competitor being introduced at all worker ratios, resources occupied by N. fulva declined by as much as 97.0% at 25.0 and 35.0 °C, but counts were relatively unchanged at 15.0 °C (Figures 5-2 to 5-4). Similar results were also observed for N. fulva at 25.0 and 35.0°C during competition experiments described in Chapter 4.

When N. fulva were first to resources at 15.0 °C, the number of resources occupied by N. fulva and the total number of ants at resources were significantly higher than those counts at 25.0 or 35.0 °C (Table 5-2; Figures 5-2B-C, 5-2E-F). At 15.0 °C, fewer N. fulva engaged in fighting which likely explained their persistence at resources. While N. fulva appeared most aggressive

towards the competitor when the N. fulva: S. invicta worker ratio was 1:1 or 2:1 at 25.0 and 35.0

°C, there was a significant difference in sucrose resources occupied at all temperatures and when worker ratios were 1:2 versus 2:1 (Tables 5-2, 5-3; Figures 5-3, 5-4). With more N. fulva engaged in fighting rather than feeding at warmer temperatures, fewer ants remained on resources.

We did observe N. fulva’s chemical detoxification of S. invicta’s venom, however at the end of 20 min there was no clear victor at resources. Recent research suggests that N. fulva can detoxify S. invicta’s venom, and likely plays a role in N. fulva’s displacement of S. invicta in serval locations in Texas (LeBrun et al. 2013, Chen et al. 2013). The short duration of our assays may have limited N. fulva’s ability to recover from exposure to S. invicta venom enough to outcompete S. invicta for resources.

When significant differences were observed between bait types for N. fulva, counts were always higher on sucrose compared to tuna baits at every temperature and worker ratio (Tables

5-1 to 5-6; Figures 5-2 to 5-4). Foraging behavior of ants is strongly influenced by the nutritional needs of individual ants as well as the needs of the colony (Hölldobler and Wilson 1990). While colonies may require protein to nourish brood and reproductives; the immediate energy needs of individual ants are met through the intake of carbohydrates. In our experiment, ants were starved for 24h prior to analysis and therefore may have required carbohydrates as an immediate energy source, thus explaining this observation.

500 S. invicta vs a Competitor (250, 500, or 1000 N. fulva)

When S. invicta was introduced first, the Temperature*Species*Bait Type interaction and the Species*Ratio*Bait Type interaction were significant for cumulative resources occupied and cumulative ants at resources (Table 5-4; Figures 5-5 to 5-7). In order to address significant interactions, data were analyzed by species and bait type so that the main effects of temperature

and ratio could be investigated. Over all ratios, sucrose resources occupied and total ants at sucrose were significantly higher for S. invicta at 35.0 °C than at 15.0 and 25.0 °C (Table 5-5;

Figures 5-5, 5-6). Over all temperatures, there was a significant difference between worker ratio

1:2 versus 2:1 in sucrose resources occupied and ants at sucrose (Table 5-6; Figure 5-7).

When S. invicta competed at a 1:1 or a 1:2 ratio with N. fulva at 35.0 °C, the introduction of the N. fulva competitor resulted in the greatest reduction in sucrose resources occupied or ants at sucrose resources by S. invicta (Figures 5-7B-C, 5-7E-F). Within 5 min of the competitor being introduced, sucrose resources occupied by S. invicta declined by >45.0% while the number of S. invicta at sucrose resources declined by >46.0%.

Similar to those observations for N. fulva at 35.0 °C, S. invicta appeared most aggressive towards the competitor when the S. invicta: N. fulva worker ratio was 2:1 or 1:1 at 35.0 °C, and least aggressive at 15.0 °C. There was a significant difference in resources occupied by S. invicta and ants at resources at 35.0 °C (Table 5-6; Figure 5-7), and when worker ratio favored S. invicta at 1:2 versus 2:1 (Tables 5-5; Figures 5-5 to 5-7). More S. invicta were engaged in fighting rather than feeding at the higher temperature, leaving fewer ants to forage and likely explained S. invicta’s significant decline in resources occupied and total number of workers at resources.

Many species of ants rely on interference competition to reduce the foraging success of competitors whose resource requirements overlap (Savolainen and Vepsäläinen 1988, Human and Gordon 1996, Lach 2005, Carpintero and Reyes-López 2008). Previous studies investigating the effect of temperature on interference competition between L. humile and subordinate ant species reported increased aggression between ants at temperatures approaching the upper thermal limits of activity (Cerdá et al. 1998a, Cerdá et al. 1998b). Our results demonstrate that the presence of a competitor suppressed the number of resources occupied and the total number

of ants at resources for N. fulva and S. invicta at warmer temperatures. Additionally, the presence of a competitor had the greatest negative effect on both species at warmer temperatures when N. fulva or S. invicta competed at a numerical disadvantage. Collectively, these findings suggest that gaining first access to resources may give one species the competitive advantage during warmer periods, especially when one species has the numerical advantage.

Our findings have shown that worker ratio in addition to temperature influence the number of resources occupied, the total number of ants per resource, and aggression of N. fulva and S. invicta. Understanding of how seasonal temperatures impact the competitive interactions between these ant species at resources could improve our timing of monitoring and baiting programs for N. fulva and S. invicta.

Table 5-1. Analyses of variance (ANOVA) table for main effects of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500), temperature (15.0, 25.0, and 35.0 °C), and resource on the cumulative number of resources occupied and the cumulative number of ants at resources when the competitor was S. invicta. Response Variable Source df F P Resources Occupied Colony 5 6.34 <0.0001 Species 1 12.83 0.0004 Temperature 2 18.26 <0.0001 Ratio 2 0.27 0.7613 Bait Type 1 50.11 <0.0001 Temperature*Species 2 23.85 <0.0001 Temperature*Ratio 4 1.28 0.2808 Temperature*Bait 2 0.76 0.4680 Species*Ratio 2 17.99 <0.0001 Species*Bait 1 0.00 0.9579 Ratio*Bait 2 0.94 0.3907 Temperature*Species*Ratio 4 1.11 0.3548 Temperature*Species*Bait 2 0.47 0.6241 Temperature*Ratio*Bait 4 0.40 0.8094 Species*Ratio*Bait 2 3.54 0.0311 Temperature*Species*Ratio*Bait 4 0.81 0.5192

Ants at Resources Colony 5 3.93 0.0021 Species 1 10.43 0.0051 Temperature 2 11.91 <0.0001 Ratio 2 0.38 0.6825 Bait Type 1 19.60 <0.0001 Temperature*Species 2 17.01 <0.0001 Temperature*Ratio 4 2.19 0.0725 Temperature*Bait 2 1.36 0.2603 Species*Ratio 2 9.91 <0.0001 Species*Bait 1 0.25 0.6198 Ratio*Bait 2 0.74 0.4796 Temperature*Species*Ratio 4 1.90 0.1130 Temperature*Species*Bait 2 0.09 0.9123 Temperature*Ratio*Bait 4 0.24 0.9146 Species*Ratio*Bait 2 1.45 0.2374 Temperature*Species*Ratio*Bait 4 1.11 0.3533

Table 5-2. Comparison of the effect of temperature by resource over all worker ratios (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) when the competitor was S. invicta. Response Variable Species Bait Type Temperature N Mean (±SD) (°C) Resources N. fulva Sucrose 15 18 39.83 (±27.78) a Occupied 25 18 21.33 (±16.42) b 35 18 13.94 (±10.69) b

Tuna 15 18 29.94 (±22.50) a 25 18 3.61 (±2.79) b 35 18 4.61 (±4.15) b

Competitor Sucrose 15 18 18.00 (±14.44) 25 18 20.00 (±16.46) 35 18 18.56 (±17.99)

Tuna 15 18 5.11 (±6.77) 25 18 6.89 (±5.66) 35 18 7.06 (±5.76)

Ants at Resources N. fulva Sucrose 15 18 81.00 (±82.13) a 25 18 37.05 (±51.97) ab 35 18 23.94 (±24.06) b

Tuna 15 18 71.44 (±77.46) a 25 18 3.89 (±3.18) b 35 18 4.67 (±4.61) b

Competitor Sucrose 15 18 26.28 (±28.00) 25 18 45.61 (±56.55) 35 18 27.00 (±31.95)

Tuna 15 18 5.54 (±7.58) 25 18 8.22 (±6.35) 35 18 7.44 (±6.15) Footnote: Means of cumulative resources occupied and cumulative ants at resources were compared between temperatures by resource for each species. Means followed by the same letter are not significantly different (P ≤ 0.05, Tukey-Kramer HSD test [Proc GLM, SAS version 9.3, SAS Institute, Cary, NC)]).

Table 5-3. Comparison of the effect of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) by resource over all temperatures when the competitor was S. invicta. Response Variable Species Bait Type Ratio N Mean (±SD) Resources Occupied N. fulva Sucrose 1:2 18 15.61 (±10.85) b 1:1 18 25.40 (±26.35) ab 2:1 18 34.11 (±23.21) a

Tuna 1:2 18 7.94 (±7.49) 1:1 18 14.94 (±22.85) 2:1 18 15.28 (±19.71)

Competitor Sucrose 1:2 18 30.22 (±18.64) a 1:1 18 13.39 (±7.85) b 2:1 18 12.94 (±13.82) b

Tuna 1:2 18 9.72 (±6.50) a 1:1 18 5.83 (±5.23) ab 2:1 18 3.5 (±4.81) b

Ants at Resources N. fulva Sucrose 1:2 18 25.22 (±25.84) 1:1 18 53.50 (±84.24) 2:1 18 62.28 (±57.71)

Tuna 1:2 18 9.28 (±9.80) 1:1 18 37.55 (±75.60) 2:1 18 33.17 (±53.98)

Competitor Sucrose 1:2 18 56.11 (±52.75) a 1:1 18 21.61 (±27.23) b 2:1 18 21.17 (±29.71) b

Tuna 1:2 18 10.43 (±7.32) a 1:1 18 7.17 (±6.19) ab 2:1 18 3.61 (±4.80) b Footnote: Means of cumulative resources occupied and cumulative ants at resources were compared between ratios by resource for each species. Means followed by the same letter for each ratio at one temperature are not significantly different (P ≤ 0.05, Tukey-Kramer HSD test [Proc GLM, SAS version 9.3, SAS Institute, Cary, NC)]).

Table 5-4. Analyses of variance (ANOVA) table for main effects of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500), temperature (15.0, 25.0, and 35.0 °C), and resource on the cumulative number of resources occupied and the cumulative number of ants at resources when the competitor was N. fulva. Response Variable Source df F P Resources Occupied Colony 5 2.00 0.0809 Species 1 22.67 <0.0001 Temperature 2 9.02 0.0002 Ratio 2 3.19 0.0438 Bait Type 1 87.27 <0.0001 Temperature*Species 2 33.99 <0.0001 Temperature*Ratio 4 1.72 0.1484 Temperature*Bait 2 2.06 0.1300 Species*Ratio 2 25.59 <0.0001 Species*Bait 1 4.34 0.0387 Ratio*Bait 2 1.66 0.1924 Temperature*Species*Ratio 4 1.52 0.1988 Temperature*Species*Bait 2 10.49 <0.0001 Temperature*Ratio*Bait 4 1.28 0.2794 Species*Ratio*Bait 2 4.03 0.0194 Temperature*Species*Ratio*Bait 4 0.84 0.4993

Ants at Resources Colony 5 0.72 0.6093 Species 1 2.41 0.1225 Temperature 2 1.12 0.3279 Ratio 2 0.66 0.5174 Bait Type 1 40.47 <0.0001 Temperature*Species 2 22.16 <0.0001 Temperature*Ratio 4 2.65 0.0348 Temperature*Bait 2 5.29 0.0059 Species*Ratio 2 14.33 <0.0001 Species*Bait 1 5.04 0.0260 Ratio*Bait 2 1.09 0.3387 Temperature*Species*Ratio 4 1.08 0.3687 Temperature*Species*Bait 2 11.66 <0.0001 Temperature*Ratio*Bait 4 1.77 0.1362 Species*Ratio*Bait 2 3.71 0.0264 Temperature*Species*Ratio*Bait 4 1.35 0.2525

Table 5-5. Comparison of the effect of temperature by resource over all worker ratios (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) when the competitor was N. fulva. Response Variable Species Bait Type Temperature (°C) N Mean (±SD) Resources Occupied S. invicta Sucrose 15 18 20.05 (±15.02) b 25 18 10.16 (±8.94) b 35 18 39.05 (±25.54) a

Tuna 15 18 3.94 (±5.17) 25 18 0.88 (±1.97) 35 18 3.055 (±5.56)

Competitor Sucrose 15 18 38.22 (±22.20) a 25 18 32.83 (±24.37) a 35 18 12.61 (±9.67) b

Tuna 15 18 27.66 (±21.27) a 25 18 11.94 (±11.10) b 35 18 5.05 (±5.49) b

Ants at Resources S. invicta Sucrose 15 18 29.72 (±27.13) b 25 18 15.27 (±13.31) b 35 18 115.61 (±117.91)a

Tuna 15 18 4.77 (±7.72) 25 18 0.944 (±1.95) 35 18 3.27 (±6.23)

Competitor Sucrose 15 18 57.38 (±42.14) 25 18 71.44 (±83.75) 35 18 19.55 (±20.43)

Tuna 15 18 50.05 (±47.13) a 25 18 18.83 (±21.66) b 35 18 7.00 (±9.09) b Footnote: Means of cumulative resources occupied and cumulative ants at resources were compared between temperatures by resource for each species. Means followed by the same letter are not significantly different (P ≤ 0.05, Tukey-Kramer HSD test [Proc GLM, SAS version 9.3, SAS Institute, Cary, NC)]).

Table 5-6. Means comparison of the effect of worker ratio (Nylanderia fulva: Solenopsis invicta, 250:500, 500:500, 1000:500) over all temperatures when the competitor was N. fulva. Response Variable Species Bait Type Ratio N Mean (±SD) Resources Occupied S. invicta Sucrose 1:2 18 32.38 (±23.71) a 1:1 18 24.50 (±23.88) a 2:1 18 12.38 (±8.19) b

Tuna 1:2 18 2.72 (±4.36) 1:1 18 2.61 (±5.30) 2:1 18 2.55 (±4.42)

Competitor Sucrose 1:2 18 17.83 (±13.84) b 1:1 18 24.94 (±19.50) b 2:1 18 40.88 (±26.45) a

Tuna 1:2 18 5.44 (±6.44) c 1:1 18 14.72 (±15.59) b 2:1 18 24.50 (±20.33) a

Ants at Resources S. invicta Sucrose 1:2 18 80.94 (±106.46) a 1:1 18 58.38 (±86.51) ab 2:1 18 21.27 (±17.49) b

Tuna 1:2 18 2.83 (±5.17) 1:1 18 3.44 (±7.93) 2:1 18 2.72 (±4.50)

Competitor Sucrose 1:2 18 25.11 (±20.87) 1:1 18 43.27 (±49.83) 2:1 18 80.00 (±78.47)

Tuna 1:2 18 7.00 (±9.43) b 1:1 18 24.11 (±29.49) b 2:1 18 44.77 (±45.87) a Footnote: Means of cumulative resources occupied and cumulative ants at resources were compared between ratios by resource for each species. Means followed by the same letter for each ratio at one temperature are not significantly different (P ≤ 0.05, Tukey-Kramer HSD test [Proc GLM, SAS version 9.3, SAS Institute, Cary, NC)]).

Figure 5-1. Diagram of the foraging arena used to evaluate the foraging behavior of Nylanderia fulva and Solenopsis invicta at 15.0, 25.0, and 35.0 °C. Triangles represent the placement locations of resources (0.25 ml of 20% sucrose solution or 0.05 g of tuna). Ant placement locations were sites where ants were introduced into foraging arenas.

Figure 5-2. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 15.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources at the following ratios: 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

Figure 5-3. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 25.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources at the following ratios: 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

Figure 5-4. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 35.0 °C when N. fulva and competitor S. invicta were placed in arenas with two resources at the following ratios: 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

Figure 5-5. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 15.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources at the following ratios: (N. fulva: S. invicta, 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

Figure 5-6. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 25.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources at the following ratios: (N. fulva: S. invicta, 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

Figure 5-7. Mean number (±SE) of resources occupied by ants (A-C) and mean number of ants at resources (D-F) at 35.0 °C when S. invicta and competitor N. fulva were placed in arenas with two resources at the following ratios: (N. fulva: S. invicta, 1:2 (A, D), 1:1 (B, E), and 1:2 (C, F) (n = 6 per ratio). Time “0” is defined as the time at which the competitor was introduced into the arena.

CHAPTER 6 SOLENOPSIS INVICTA (BUREN) ARE DISPLACED FROM SUCROSE AND PROTEIN RESOURCES BY LOW AND HIGH NUMBERS OF NYLANDERIA FULVA (MAYR)

Introduction

Numerous ant species including the big headed ant, Pheidole megacephala (Fabricius)

(Smith 1933 in Deyrup et al. 2000); the Argentine ant, Linepithema humile (Mayr) (Wheeler

1932 in Deyrup et al. 2000); and the red imported fire ant, Solenopsis invicta (Buren) (Tschinkel

2006) have entered through the southern U.S. Solenopsis invicta was introduced in the 1930’s and was considered to be among the most ecologically dominant ants in the southern U.S.

(Tschinkel 2006). Recently, the tawny crazy ant, Nylanderia fulva (Mayr) has been identified from several southern states. It is a polydomous and polygynous species that can overwhelm native ecosystems and displace multiple ant species including S. invicta (LeBrun et al. 2013).

Competition for resources is known to play an important role in structuring ant communities (Fellers 1987, Savolainen and Vepsäläinen 1988, Human and Gordon 1996,

Holway 1999). Several factors including colony size and forager recruitment can influence the outcome of resource competition (Holway 1999, Holway and Case 2001, Barbieri et al. 2014).

For example, in L. humile, larger colony size has been attributed to higher nestmate recruitment, faster resource location, and increased aggression towards other ants improving their competitive success against native ant species (Human and Gordon 1996, Holway and Case 2001).

Diet also has been documented to affect the competitive behavior of multiple ant species including L. humile and N. fulva (Grover et al. 2007, Horn et al. 2013). When sucrose was restricted in laboratory colonies, it resulted in lower aggression among of L. humile when workers were placed together in an arena; however, there were no differences in aggression among colonies deprived of protein compared with controls (Grover et al. 2007). Conversely,

Horn et al. (2013) found that N. fulva workers on low-sugar diets were more aggressive and less

likely to be killed in encounters with S. invicta. These findings suggest the effect of the availability of sucrose on behavior is species dependent and may be influenced by other factors that were not investigated.

The displacement of S. invicta by N. fulva during the initial stages of invasion has been documented (LeBrun et al. 2013); however, the mechanisms associated with that displacement are still being defined. Our objective was to measure the response of S. invicta to low and high levels of N. fulva at oil-protein and sucrose resources. Unlike the laboratory experiments described above, it was not possible to control what N. fulva field colonies fed on prior to this study. However, based on our previous laboratory studies, we expect that S. invicta will be displaced from both resource types regardless of whether N. fulva foraging intensity is high or low.

Materials and Methods

Collection of Field Colonies

Solenopsis invicta colonies were collected in June 2015 from Alachua County, and were extracted from mounds using a modified drip-floatation method as described by Banks et al.

(1981). Colonies were kept in the laboratory in plastic trays with sides coated with fluon (Insect-

A-Slip®, BioQuip, Rancho Dominguez, CA) and maintained at approximately 55.0 ± 8% RH,

27.0 ± 3.0 °C, and 12:12 h L:D photoperiod. Ants were provided with one artificial nest made of a polystyrene Petri dish (150 x 15 mm) partially filled with dental plaster (Castone® Dental

Stone, Dentsply International, York, PA, USA) and moistened with deionized water. Every other day a protein source of housefly maggots was provided, as well as access to deionized water and

20% sucrose solution in glass test tubes (16.0 x 150.0 mm) plugged with cotton and replaced as needed.

Solenopsis invicta Introduction Experiments

All field assays were conducted in June 2015 in Nassau County, FL. The landscape surrounding the field site is characteristic of a typical Florida scrub zone, with mixed vegetation consisting of oak and pine trees, scrub palmettos, and native grasses. During the time of the field assays, only three small S. invicta colonies existed at the site. A residential area is located southwest of the field site, and a small creek (Alligator Creek) runs through the north most corner. Nylanderia fulva were first reported at the site and the immediate surrounding areas around 2009.

To investigate how low and high foraging intensities of N. fulva affected encounters with

S. invicta at food resources, assays were conducted in the morning (7:00 – 9:00) and evening

(19:00 – 21:00) at two locations. Each location was spaced approximately 100 m apart with similar ground cover consisting of low-growing scrub palmettos and native grasses on damp clay and sandy soils. Before each assay, counts of N. fulva foragers were taken at each location to estimate foraging intensity by placing sliced Vienna sausage (ca. 1.0 g) (Original Vienna

Sausage, Cherry Hill, NJ; 2.7g of fat and 1.7g of protein per 16.0g sausage) every 2.0 m in a 6 x

6 grid pattern that measured 10.0 x 10.0 m (Figure 6-1). After approximately 30 min, the number of N. fulva foragers on resources was counted. The Vienna sausage with the fewest N. fulva foragers (<30 ants) was identified as the area of low forager intensity while the Vienna sausage with the most foragers (>70 ants) was identified as an area of high forager intensity. “Low” and

“high” designations were made based upon forager counts taken at a single sausage slice. One randomly selected area was identified as the control site where N. fulva were excluded from resources during the entire study. Thus, the treatments for this experiment included a control site, one site with a high N. fulva forager intensity, and one site with a low N. fulva forager intensity.

All three sites were within the 10 x 10 m grid, and the same three treatments were established at each field location (Figure 6-1).

A plastic excluder (47.0 x 38.0 x 13.3 cm) for each treatment was installed at each field location to regulate N. fulva’s access to resources (ca. 1.0 g Vienna sausage and 0.5 ml honey).

Excluders consisted of plastic cat litter trays (#CP2, Van Ness Plastic Molding Co., Clifton, NJ) with the bottom removed and the walls coated interiorly and exteriorly with fluon. Additionally, excluders were fitted with a cover to provide shade (Figure 6-2). One S. invicta colony fragment

(2,000 workers, 2 queens, ~1.0 ml brood) was placed inside each excluder and left for 10 min to acclimate. Afterwards, resources were arranged on an index card (7.6 x 12.7 cm) and placed inside each plastic excluder, allowing S. invicta immediate access to resources, but not N. fulva.

When >10 S. invicta were present at resources (~10 min), excluders were raised at sites with high and low forager level to allow N. fulva access to resources, but were not raised at control sites to allow S. invicta to interact with resources in the absence of competition with N. fulva. The number of ants at each resource was counted for both species at 2 min intervals for 60 min.

Counts were averaged by species over 60 min resulting in one mean count per food resource per trial. Experiments were repeated in the morning at a temperature of (25.9 ± 1.69 °C) and evening at a temperature of (27.9 ± 1.25 °C) at both sites. The design of this experiment was completely randomized with 12 replicates where sampling occurred either in the morning or evening at two sites on three consecutive dates.

Data Analysis

Mean counts of S. invicta and N. fulva at honey and Vienna sausage resources were graphed to visualize resource occupancy over time for each species and resource type (Figs. 5 and 6). Data were analyzed by species. Ranks of cumulative counts for each species were used because they did not meet assumptions of ANOVA. First, to examine the effect of N. fulva

100

forager intensity on S. invicta at resources, I used a general linear model analysis of variance

(ANOVA) on ranks (Conover and Iman 1981) where forager intensity and resource type were treatments and S. invicta counts was the response variable (Proc GLM; SAS version 9.3, SAS

Institute, Cary, NC, USA). Second, to further investigate the food preference of field populations of N. fulva, I used a general linear model ANOVA where the number of N. fulva was the response variable with forager intensity and resource types as treatments. For both analyses, resource type was designated as a fixed effect and forager intensity of N. fulva was designated a random effect. Tukey’s Honest Significant Difference (HSD) Test (α = 0.05) was used to separate means between forager count treatments for each analysis (SAS version 9.3, SAS

Institute, Cary, NC, USA).

Results and Discussion

Regardless of forager intensity, N. fulva significantly decreased the number of S. invicta at resources compared to the control (F = 104.89; df = 2, 71; P <0.0001) (Table 6-1, Figure 6-3).

At sites of both low and high forager intensity, counts of S. invicta declined to an average of <1 within 10 min of N. fulva’s access to sites (Figure 6-4) and remained <1 for the duration of the assay. Simultaneously, counts of N. fulva increased over the duration of the experiment regardless of forager intensity (Figure 6-5). It was only at control sites where N. fulva were excluded that S. invicta persisted on resources over the duration of the experiment (Figure 6-4).

Other field studies examining interactions between L. humile and native ants found that larger colony size contributed to higher nestmate recruitment and increased aggression, thus improving L. humile’s competitive success against native ants (Human and Gordon 1996,

Holway and Case 2001). While I did not quantify recruitment or aggression in this study, field colonies of N. fulva were considerably larger than S. invicta colony fragments of ~2,000 workers.

In Chapter 5, we demonstrated in the laboratory that increasing N. fulva from 250 to 1,000

101

resulted in significant reductions of S. invicta at resources when temperatures were 35.0 °C, but at lower temperatures (15.0 and 25.0 °C), differences in worker ratio had little effect on the number of resources occupied. Thus, strong evidence points to a combination of larger colony size and warm temperatures as factors in N. fulva’s competitive success.

Nylanderia fulva’s displacement of S. invicta at low forager intensity sites may demonstrate N. fulva’s ability to defend the edges of territorial boundaries where forager numbers are lowest as was found in L. humile (Human and Gordon 1996, Holway and Case

2001). A limitation of this study may be our definition of low and high N. fulva forager intensities because they were numerically not very different (30 vs 70 N. fulva on sausage) when taken in context of the potential size of N. fulva colonies. Also, forager counts were taken geographically close together for study feasibility and lack of existing information on defining distinct N. fulva colonies. Additional research is needed on colony size and defining territories.

Overall, significantly more S. invicta were counted at sausage (oil-protein) resources than at honey (carbohydrate) resources (F = 41.67; df = 1, 71; P <0.0001) (Table 6-1, Figure 6-6).

Vienna sausage is high in fat, and the preference of S. invicta for fats and oils is well documented

(Lofgren et al. 1954, Glunn et al. 1981). In contrast, significantly more N. fulva workers were counted at honey resources than at sausage resources (F = 13.06; df = 1, 47; P = 0.0008) (Table

6-2, Figure 6-7). Honey is a rich carbohydrate source and polygynous ant species are known to exhibit a strong preference for carbohydrates (Abbott 2005, Rowles and Silverman 2009).

Solenopsis invicta were slow to occupy resources compared to N. fulva (Figures 6-4 and

6-5). These findings were supported by previous laboratory results in Chapter 4 where S. invicta were slower to occupy resources even in the absence of competition at 25.0 and 35.0 °C. It appeared that S. invicta spent the initial minutes after introduction into a foraging arena re-

102

gathering brood; whereas, N. fulva spent initial minutes occupying resources. Together, these findings may suggest that S. invicta’s initial response to disruption, in this case competition, is to protect brood; whereas, N. fulva’s response is to dominate resources.

(LeBrun et al. 2014) demonstrated N. fulva’s ability to detoxify S. invicta venom as part of the explanation for N. fulva’s ability to outcompete S. invicta during individual-level interactions. I also observed N. fulva biting and/or using defensive chemicals against S. invicta workers and reproductives. My field observations align with those findings by Lebrun et al.

(2014) and suggest that N. fulva’s use of defensive chemicals plays an important role in N. fulva’s initial displacement of S. invicta.

LeBrun et al. (2013) concluded that S. invicta was displaced by N. fulva via sampling areas that were uninvaded, invaded and adjacent to invaded areas with pitfall traps, then defining the prevalence of species in an area. Here, we experimentally confirmed 1) that N. fulva can displace S. invicta within 10 min of initial interactions, 2) that warm temperature and colony size are significant factors in N. fulva’s competitive success, and 3) that a difference in food preference combined with what appears to be different strategies in reacting to colony disruption

(i.e., S. invicta collecting brood versus N. fulva dominating resources) may be important factors in N. fulva’s ability to initially displace S. invicta from an area.

103

Table 6-1. Analysis of variance table showing the effects of forager intensity and resource type on the number of Solenopsis invicta counted at Vienna sausage and honey resources when N. fulva were present in high numbers (>70 N. fulva at Vienna sausage resources), when N. fulva were present in low numbers (<30 N. fulva at sausage resources), or when N. fulva were excluded from resources (control). Source df F P Whole Model 5 50.75 <0.0001 N. fulva forager intensity 2 104.89 <0.0001 Resource type 1 41.67 <0.0001 Resource type*N. fulva forager intensity 2 1.13 0.3281 Error 66 Total 71

104

Table 6-2. Analysis of variance table showing the effects of forager intensity and resource type on the number of Nylanderia fulva counted at Vienna sausage and honey resources when N. fulva were present in high numbers (>70 N. fulva at Vienna sausage resources) or when N. fulva were present in low numbers (<30 N. fulva at sausage resources). Source df F P Whole Model 3 9.53 <0.0001 N. fulva forager intensity 1 15.37 0.0003 Resource type 1 13.06 0.0008 Resource type*N. fulva forager intensity 1 0.17 0.6838 Error 44 Total 47

105

Figure 6-1. A diagram showing the placement of sliced Vienna sausage (ca. 1.0 g) (Original Vienna Sausage, Cherry Hill, NJ) baits that were used to take counts of N. fulva foragers at each location. Sausage slices were placed every 2.0 m in a 6 x 6 grid pattern that measured 10.0 x 10.0 m. After approximately 30 min, the number of N. fulva foragers on sausage slices was counted. The Vienna sausage with the fewest N. fulva foragers (always <30 ants) was identified as a low forager intensity site while the Vienna sausage with the most foragers (always >70 ants) was identified as a high forager intensity site. The red box represents the hypothetical placement of excluders (47.0 x 38.0 x 13.3 cm) at one low, high, and control site within the 10 x 10 m grid.

106

Figure 6-2. Plastic excluder (47.0 x 38.0 x 13.3 cm) constructed of a modified plastic cat litter tray (#CP2, Van Ness Plastic Molding Co., Clifton, NJ) fitted with a cover to provide shade, and with walls interiorly and exteriorly coated with Fluon® (Insect-A-Slip, BioQuip, Rancho Dominguez, CA) to prevent ants from entering or escaping. Photo courtesy of Michael Bentley.

107

Figure 6-3. Cumulative count of Solenopsis invicta (± SE) (n = 6 per treatment) on Vienna sausage and honey resources over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage bait), with a low number of N. fulva foragers (≤30 N. fulva foragers per bait), or when N. fulva were excluded from resources (control). Mean values sharing the same letter are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3]).

108

Figure 6-4. Mean number of Solenopsis invicta (± SE) (n = 6 per time point for each treatment or control) on Vienna sausage and honey resources counted at 2 min intervals over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage), with a low number of N. fulva forager (≤30 N. fulva foragers per sausage), and when N. fulva foragers were excluded from resources (Control).

109

Figure 6-5. Mean number of Nylanderia fulva (± SE) (n = 6 per time point for each treatment) on Vienna sausage and honey resources counted at 2 min intervals over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage) and with a low number of N. fulva forager (≤30 N. fulva foragers per sausage).

110

Figure 6-6. Cumulative count of Solenopsis invicta (± SE) (n = 6 per treatment) on honey and Vienna sausage and honey resources over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage bait), with a low number of N. fulva foragers (≤30 N. fulva foragers per bait), or when N. fulva were excluded from resources (control). Mean values sharing the same letter are not significantly different (P >0.05, Tukey- Kramer HSD test [SAS version 9.3]).

111

Figure 6-7. Cumulative count of Nylanderia fulva (± SE) (n = 6 per treatment) on honey and Vienna sausage resources over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage bait) and with a low number of N. fulva foragers (≤30 N. fulva foragers per bait). Mean values sharing the same letter are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3]).

112

Figure 6-8. Cumulative count of Nylanderia fulva (± SE) (n = 6 per treatment) on Vienna sausage and honey resources over 60 min when S. invicta colonies (2,000 workers, 2 queens, ~1.0 ml brood) interacted with a high number of N. fulva foragers (≥70 N. fulva foragers per sausage bait) and with a low number of N. fulva foragers (≤30 N. fulva foragers per bait). Mean values sharing the same letter are not significantly different (P >0.05, Tukey-Kramer HSD test [SAS version 9.3]).

113

CHAPTER 7 FIELD POPULATIONS OF NYLANDERIA FULVA (MAYR) READILY CONSUME A DILUTED FORMULATION OF MAXFORCE™ QUANTUM LIQUID ANT BAIT IN CHOICE AND NO-CHOICE ASSAYS

Introduction

Insecticidal baits at current label rates have had limited success in managing invasive N. fulva populations. The carbohydrate-based bait, Maxforce™ Quantum (0.03% imidacloprid,

Bayer Environmental Science, Montvale, NJ), is acceptance by workers of N. fulva, but is viscous and seemingly difficult to manipulate. This bait can eventually harden. When compared to other carbohydrate sources consumed by N. fulva in the field, the bait may have limited consumption and transfer throughout the colony. Preliminary data that diluting Maxforce™

Quantum with a 25.0% sucrose solution lowers viscosity and increases consumption by N. fulva.

Previous data demonstrate that a single control method is ineffective in controlling this ant, and that management is most successful when integrated pest management (IPM) methods are started in early spring (F. M. Oi, personal communication). Thus, the purpose of this study was not to measure the efficacy of diluted Maxforce™ Quantum on decreasing N. fulva, but to investigate the consumption of diluted Maxforce™ Quantum in choice and no-choice field assays during the summer when N. fulva populations were abundant and active.

Materials and Methods

All field assays were conducted in Nassau County, FL. No-choice bait consumption assays were conducted from July to November 2014, and choice assays were conducted from

June to August 2015. As described in Chapter 6, the landscape surrounding the Nassau County field site is characteristic of a typical FL scrub zone, with mixed vegetation consisting of oak and pine trees, scrub palmettos, and native grasses. Nylanderia fulva were first reported in the area in

2009.

114

Each bait station used in no-choice (2014) and choice (2015) assays consisted of one plastic bottle (500.0 ml) fitted with two wicks cut from synthetic chamois cloth (1.5 x 64.8 cm,

Aquadry Towel, Calderon Textiles, Indianapolis, IN). The bottle was placed into a covered PVC- pipe housing (25.0 x 7.6 cm I.D.) which had four access holes (1.0 cm I.D.) located equidistantly around the circumference of the pipe at approximately 5.0 cm from the bottom. Diluted

Maxforce™ Quantum was prepared by dissolving 226.8 g of table sugar in 946.0 ml of water and 120.0 g of Maxforce™ Quantum, resulting in a final solution of 0.003% imidacloprid in the bait formulation.

2014 No-Choice Bait Assay

Nineteen experimental bait stations containing only diluted Maxforce™ Quantum were installed approximately 3.0 m apart along the perimeter of the field site (Figure 7-1), and the volume of diluted Maxforce™ Quantum in each bait station was measured every 14 days over 16 weeks (n=8 observation dates). When bait stations were first serviced on July 10th, which approximately coincides with the population height of these ants, all 150.0 ml of diluted

Maxforce™ Quantum had been depleted. To ensure an adequate volume of diluted Maxforce™

Quantum was provided, the volume of diluted Maxforce™ Quantum of each replacement container was increased to 300.0 ml for the remainder of 2014 and 2015 studies.

Bait containers were entirely replaced at each visit. Consumption of diluted Maxforce™

Quantum was calculated by subtracting the volume recorded at each inspection from the starting volume of 300 ml (150 ml for July 10th). We then used various descriptive statistics to illustrate the amount of bait consumed per station, site, and field season.

In addition, N. fulva were sampled at each bait station using sausage lures (0.5 cm sliced

Vienna sausage, Original Vienna Sausage, Cherry Hill, NJ). Sausage lures were placed at around

115

8:00 am each morning and left for approximately 30 min before the number of foragers per sausage lure (maximum of 100 ants) was recorded.

2015 Choice Bait Assay

In 2015, we measured N. fulva’s consumption of diluted Maxforce™ Quantum when paired with a 25.0% sucrose solution to examine the effect of a competing food source. Sixteen pairs of bait stations were installed along the perimeter of the field site (Figure 7-3). Paired bait and sucrose solution stations were installed approximately 10.0 cm from each other, while pairs of bait stations were installed approximately 3.0 m apart. Each bait station contained either 300.0 ml of diluted Maxforce™ Quantum (treatment) or 300.0 ml of 25.0% sucrose solution (control), and was serviced every 14 days for 8 weeks (n=4 observation dates) as described above in 2014 experiments. Similarly, N. fulva were sampled as described above in 2014 experiments.

Results and Discussion

Nylanderia fulva’s mean depletion of diluted Maxforce™ Quantum was lower when a competing food source was present. When colonies were only provided diluted bait in 2014, N. fulva depleted 3,237.0 ± 384.6 ml per 14 day sampling period (Figure 7-3). After the volume of diluted bait per station was increased to 300.0 ml, depletion of diluted bait averaged 173.3 ± 6.72 ml per station and ranged from 15.0 to 300.0 ml. In 2015, when N. fulva were provided both diluted Maxforce™ Quantum and 25.0% sucrose solution, ants depleted 36.1% less diluted bait per 14 day sample period (2,069.0 ± 191.4 ml) while depleting 100.0% of the 25.0% sucrose solution (4,200.0 ± 0.0 ml) (Figure 7-4). Additionally, the mean number of N. fulva per sausage lure averaged over all dates increased from 64.7 ± 2.5 in 2014 (Figure 7-5) to 76.2 ± 5.8 in 2015

(Figure 7-6) , suggesting the reduction in diluted bait consumption from 2014 to 2015 was not likely a result of a decline in foraging populations for N. fulva. These observations combined with our results suggest that the presence of competing food sources could reduce N. fulva’s

116

consumption of this diluted bait. Management programs for N. fulva using bait stations equipped with diluted Maxforce™ Quantum may be enhanced if alternative carbohydrate sources such as honeydew producing insects are reduced or eliminated.

Nylanderia fulva depleted 25,896.0 ml of diluted Maxforce™ Quantum over 16 weeks in

2014 (Figure 7-3), and 8,275.0 ± 191.4 ml of diluted bait and 4,200.0 ml of sucrose solution over eight weeks in 2015 (Figure 7-5). Our findings indicate that N. fulva depleted an average of nearly 1.5 liters of diluted bait per week, providing novel information regarding the volume of liquid bait that large colonies can consume. Collectively, these data provide an initial prospective baseline of diluted bait needed for high density populations of N. fulva which could have cost implications for pest control operators planning short-term or season long management of larger

N. fulva infestations.

Our findings have demonstrated the palatability of a dilute formulation of Maxforce™

Quantum liquid ant bait to foragers of N. fulva in the field. Furthermore, the results of this field assay illustrate the importance of mitigating other carbohydrate sources to improve consumption of this diluted bait. Ultimately, these data contribute to our understanding of the response of pest ants to liquid baits (Klotz et al. 1997, Rust et al. 2002, Klotz et al. 2003), and may improve baiting strategies for this or other invasive pest ants.

117

Figure 7-1. Diagram of 19 bait stations installed in 2014 along the perimeter of the field site in Nassau County, FL to evaluate Nylanderia fulva’s consumption of diluted Maxforce™ Quantum. Each bait station contained 300 ml of the diluted bait, and was serviced every 14 days from July – November. In addition, N. fulva were sampled at each bait station using sausage lures (0.5 cm sliced Vienna sausage, Original Vienna Sausage, Cherry Hill, NJ).

118

Figure 7-2. Diagram of 16 pairs of bait stations installed in 2015 along the perimeter of the field site in Nassau County, FL to evaluate Nylanderia fulva’s consumption of diluted Maxforce™ Quantum when ants were provided the diluted bait in addition to an alternative carbohydrate source (25.0% sucrose solution). Each bait station contained 300 ml of the diluted bait or 300 ml of 25.0% sucrose solution, and was serviced every 14 days from June – August. In addition, N. fulva were sampled at each bait station using sausage lures (0.5 cm sliced Vienna sausage, Original Vienna Sausage, Cherry Hill, NJ).

119

Figure 7-3. Cumulative mean (± SE) volume of diluted Maxforce™ Quantum liquid ant bait depleted from 19 bait stations at a field site heavily invested with Nylanderia fulva in Nassau County, FL. Stations installed on 7/10/2014 were filled with 150.0 ml of diluted bait, while bait stations for all other service dates were filled with 300.0 ml.

120

Figure 7-4. Cumulative mean (± SE) volume of diluted Maxforce™ Quantum liquid ant bait and 25.0% sucrose solution depleted from 16 pairs of bait stations at a field site heavily invested with Nylanderia fulva in Nassau County, FL. Each pair of bait stations was filled with 300 ml of the diluted bait and 300 ml of 25.0% sucrose per service date.

121

Figure 7-5. Mean (± SE) number of Nylanderia fulva (n = 19 per service date) sampled at each bait station in 2014 using sausage lures (0.5 cm sliced Vienna sausage, Original Vienna Sausage, Cherry Hill, NJ) that were deployed around 8:00 am each morning and left for approximately 30 min before the number of foragers per sausage lure (maximum of 100 ants) was recorded.

122

Figure 7-6. Mean (± SE) number of Nylanderia fulva (n = 16 per service date) sampled at each bait station in 2015 using sausage lures (0.5 cm sliced Vienna sausage, Original Vienna Sausage, Cherry Hill, NJ) that were deployed around 8:00 am each morning and left for approximately 30 min before the number of foragers per sausage lure (maximum of 100 ants) was recorded.

123

CHAPTER 8 CONCLUSIONS

Nylanderia fulva (Mayr) is a relatively new invasive ant that has ecologically dominated landscapes in the southern U.S., even displacing the aggressive red imported fire ant, Solenopsis invicta (Buren). Existing efforts to manage Nylanderia fulva using conventional bait and insecticide treatments have failed to provide any long-term suppression. In North Florida, seasonal cold air temperatures (≤ 15.0 °C) lead to a dramatic decrease in N. fulva’s foraging activity, thus providing a source of consistent relief for homeowners and pest management professionals. Although the effects of seasonal cold on N. fulva’s foraging behavior have been observed, the thermal biology and ecology of N. fulva are poorly understood. Therefore, the overarching goal of my research was to better understand how seasonal temperatures in north

Florida may impact N. fulva’s success as an invader, thus providing information that may improve its management.

In summary, my research has 1) identified N. fulva’s thermal limits for activity which can improve our understanding of how temperature may limit its distribution and seasonal activity, 2) confirmed a behavior (subterranean tunneling) used by many ant species as a thermoregulatory strategy that provides further insight into how N. fulva may survive sub-freezing temperatures, and thus can improve seasonal monitoring and management, 3) demonstrated the significance of numerical dominance and temperature for N. fulva’s ability to compete for resources with another historically successful invader, S. invicta, which contributes to our understanding of how

N. fulva may outcompete S. invicta for resources seasonally and 4) demonstrated the reduced depletion of diluted formulation of Maxforce™ Quantum liquid ant bait when a another sucrose source was present illustrating the importance of reducing alternative carbohydrate sources when utilizing this bait.

124

Little data existed on the thermal sensitivity of N. fulva. Therefore, the first objective of this dissertation was to examine the upper and lower thermal limits for activity of N. fulva and the widespread invasive ant, S. invicta (Chapter 2). I hypothesized that N. fulva, a comparatively smaller bodied ant, would have a narrower thermal breadth than S. invicta. Additionally, I postulated that faster rates of thermal change would result in lower thermal tolerance for both species. To test these hypotheses I evaluated the effects of body size and thermal ramping rate on the upper (CTmax) and lower (CTmin) critical thermal limits of N. fulva and S. invicta. Overall, N. fulva had a narrower thermal breadth than S. invicta. For both species, smaller ants had a narrower thermal breadth than larger ants. Furthermore, slower thermal ramping rates resulted in lower CTmax values for both species. These data not only provide the first estimation of critical thermal limits of N. fulva, but they also refine our understanding of the critical thermal limits of both species. This knowledge can be used to develop predictive models that estimate the future spread of N. fulva and S. invicta, thereby directly improving our quarantine and monitoring efforts of both species.

Nylanderia fulva can overwinter in similar ranges as S. invicta throughout the southeastern U.S., despite N. fulva’s narrower thermal breadth (Chapter 1). However, nothing is known of possible thermoregulatory strategies exhibited by N. fulva. Thus, the second goal of this dissertation was to investigate subterranean tunneling as a possible thermoregulatory strategy for N. fulva based upon behavior I observed of N. fulva tunneling in the field (Chapter

3). Other ants, such as S. invicta, rely on subterranean tunneling as a means to avoid thermal extremes. I hypothesized that N. fulva, a tropical species, has the capacity to tunnel below ground at lower thermal thresholds in order to successfully overwinter in North Florida. To test this hypothesis, tunneling performance for N. fulva and S. invicta, another ant that tunnels

125

extensively, were compared at four temperatures (15.0, 18.0, 20.0, and 22.0°C). Nylanderia fulva tunneled furthest at warmer temperatures whereas S. invicta tunneled furthest at cooler temperatures. Overall, N. fulva tunneled significantly less than S. invicta. However, N. fulva constructed subterranean tunnels at all temperatures evaluated. These data support my hypothesis that N. fulva is capable of tunneling in temperatures as low as 15.0 °C, confirming that this ant can also perform a behavior that is used by other ants for cold avoidance.

Competition for resources is known to play an important role in structuring ant communities. Although N. fulva has been shown to displace S. invicta, the mechanisms associated with that displacement are still being evaluated. Temperature has even been shown to influence competition between ants, but most studies have focused on temperatures near the upper thermal thresholds of foraging. Therefore, the third goal of this dissertation was to evaluate the impact of a range of temperatures (15.0, 25.0, 35.0 °C) on competition between N. fulva and

S. invicta at sucrose and tuna resources in short-term laboratory studies (Chapter 4). I hypothesized that N. fulva would occupy more resources with a greater number of ants than S. invicta at warmer temperatures, but S. invicta’s lower thermal tolerance would give it the competitive edge over N. fulva at 15.0 °C. To test these hypotheses I compared the number of resources occupied by ants and the total number of ants at resources for 1000 workers of each species in the absence and presence of competition at 15.0, 25.0, and 35.0°C. When no competitor was present, N. fulva were quicker to resources in the first five min of being introduced at 25.0 and 35.0 °C, leading to a higher number of resources occupied and a greater number of ants a resources than S. invicta. These data suggest that, during warmer periods, N. fulva could rely on faster resource location to exploit resources before competitors such as S. invicta arrive, possibly contributing to N. fulva’s initial displacement of S. invicta.

126

Data from the previous competition studies in Chapter 4 indicated that temperature had a significant effect on resources occupied and total ants at resources for both N. fulva and S. invicta in the presence of competition when colony numbers were equal (1,000 N. fulva to 1,000

S. invicta workers). However, the size of field colonies of N. fulva and S. invicta can be highly variable over time and may rarely be equal to one another. Numerical dominance at resources has been among the factors contributed to the competitive success of other unicolonial invasive ant species such as L. humile (Holway et al. 2002). Therefore, the fourth goal of this dissertation

(Chapter 5) was to investigate worker ratio in addition to temperature as factors that could influence competition between N. fulva and S. invicta. I hypothesized that N. fulva colony fragments competing with a numerical advantage would occupy more resources with a greater number of ants than the competitor, even at lower temperatures. To test this hypothesis, competition experiments were repeated as described above for Chapter 4, but worker ratio was manipulated to provide three ratios of N. fulva to S. invicta (1:2, 1:1, 2:1). When N. fulva was introduced first, the number of sucrose resources occupied by N. fulva and the total number of ants at sucrose were greatest at 15.0 °C and when worker ratios were 2:1 vs 1:2. When S. invicta was introduced first, S. invicta occupied more sucrose resources with a greater number of ants at

35.0 °C and when worker ratios were at 1:1 or a 1:2 ratio with N. fulva. Additionally, the presence of a competitor had the greatest negative effect on both species at warmer temperatures when N. fulva or S. invicta competed at a numerical disadvantage. Together, these findings suggest that interference competition may favor N. fulva when competing at a numerical advantage during colder temperatures, while the same may be true for S. invicta during warmer periods. For both N. fulva and S. invicta, gaining first access to resources may give one species

127

the competitive advantage during warmer periods, especially when one species has the numerical advantage.

The previous competition assay (Chapter 5) indicated that numerical advantage and diet could influence competition between N. fulva and S. invicta. However, this study was completed under laboratory conditions that could influence the behavior of either species. The fifth goal of this dissertation was to field validate the results of Chapter 5 by measuring the effect of numerical dominance on competition between these species (Chapter 6). In this study I evaluated the response of S. invicta to low and high levels of N. fulva at oil-protein (Vienna sausage) and sucrose (honey) resources. I hypothesized that, even at low numbers, N. fulva foragers would displace S. invicta foragers from resources. Regardless of forager intensity, N. fulva displaced S. invicta from both resources within 10 min of N. fulva’s access to resources. Additionally, nearly two times more N. fulva recruited to honey than to sausage. These findings suggest that N. fulva’s displacement of S. invicta, even at low forager intensity sites, may demonstrate N. fulva’s ability to defend the edges of territorial boundaries where forager numbers are lowest.

Furthermore, N. fulva’s recruitment to honey over sausage confirms this ant’s strong affinity for carbohydrates.

The sixth and final goal of this dissertation was to evaluate the depletion of a diluted formulation of Maxforce™ Quantum (MQD) liquid ant bait by large populations of N. fulva in a no-choice assay (2014) and in a choice assay with a competing sucrose solution (25.0% sucrose)

(2015) (Chapter 6). I hypothesized that depletion of MQD by foragers of N. fulva would remain high, even in the presence of a competing carbohydrate bait. Overall, N. fulva depleted 25,896.0 ml of diluted bait over 16 weeks in 2016, and 8275.0 over 8 weeks in 2015. Nylanderia fulva’s mean depletion of diluted bait per 14 day sample period was 36.1% lower when a competing

128

sucrose solution was present in 2015 (2,069.0 ± 191.4 ml) compared to when ants were only provided diluted Maxforce™ Quantum in 2014 (3,237.0 ± 384.6 ml). These data demonstrated the reduction in depletion of diluted Maxforce™ Quantum when a competing carbohydrate source is present. This suggests that baiting strategies utilizing liquid, sugar bait formulations may enhance their effectiveness if alternative carbohydrate sources such as honeydew-producing

Homoptera are also managed.

Collectively, results from this research have improved our understanding of the seasonal behavior and competitive performance of N. fulva, and have provided further insight into factors that contribute to N. fulva’s success as an invasive species.

129

APPENDIX SAS CODE FOR REPEATED MEASURES ANALYSES OF VARIANCE (ANOVA) EXAMINING THE EFFECTS OF TEMPERATURE AND SPECIES ON NUMBERS OF RESOURCES OCCUPIED AND NUMBERS OF ANTS AT RESOURCES IN BOTH NO- COMPETITION AND COMPETITION EXPERIMENTS WHEN WORKER-RATIO WAS NOT MANIPULATED

SAS Code for the Analysis of Data for Nylanderia Fulva and Solenopsis Invicta when Ants Foraged without Competition proc mixed data=NOCOMPETITION; class Subject Colony Species Temp Bait Time; model Resources = Time Colony Species Temp Bait Species*Temp Species*bait Time*Colony Time*Species Time*Temp Time*Bait Time*Species*Temp Time*Species*Bait Time*Species*Temp*Bait/ ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temp / slice=Time adjust=tukey; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Bait / Slice=bait slice=species adjust=tukey; lsmeans Time*Species*Temp / slice=Time*Temp slice=Species*Time adjust=tukey; lsmeans Time*Species*Temp*Bait / slice=Time*Temp slice=Time*Species slice=Temp*Species slice=Time*Temp*Species adjust=tukey; run; proc mixed data=NOCOMPETITION; class Subject Colony Species Temp Bait Time; model Ants = Time Colony Species Temp Bait Species*Temp species*bait Time*Colony Time*Species Time*Temp Time*Bait Time*Species*Temp Time*Species*Bait Time*Species*Temp*Bait/ ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temp / slice=Time adjust=tukey; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Bait / Slice=bait slice=species adjust=tukey; lsmeans Time*Species*Temp / slice=Time*Temp slice=Species*Time adjust=tukey; lsmeans Time*Species*Temp*Bait / slice=Time*Temp slice=Time*Species slice=Temp*Species slice=Time*Temp*Species adjust=tukey; run;

SAS Code for the Analysis of Data for N. Fulva and a Competitor (S. Invicta) when N. Fulva was Given First Access to Resources proc mixed data=NFvsCompetitor; class Subject Colony Species Temperature Time Bait;

130

model Resources = Time Colony Species Temperature Bait Species*Temperature Time*Colony Time*Species Time*Temperature Time*Bait Time*Species*Bait Time*Species*Temperature Time*Species*Temperature*Bait/ ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temperature / slice=Time adjust=tukey ; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Temperature / slice=Temperature slice=species adjust=tukey; lsmeans Time*Species*Temperature / slice=Time*Temperature adjust=tukey; lsmeans Time*Species*Temperature*Bait/ slice=Time*Temperature slice=Time*Species slice=Temperature*Species slice=Time*Temperature*Species adjust=tukey; run; proc mixed data=NFvsCompetitor; class Subject Colony Species Temperature Time Bait; model ants = Time Colony Species Temperature Bait Species*Temperature Time*Colony Time*Species Time*Temperature Time*Bait Time*Species*Bait Time*Species*Temperature Time*Species*Temperature*Bait / ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temperature / slice=Time adjust=tukey ; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Temperature / slice=Temperature slice=species adjust=tukey; lsmeans Time*Species*Temperature / slice=Time*Temperature adjust=tukey; lsmeans Time*Species*Temperature*Bait/ slice=Time*Temperature slice=Time*Species slice=Temperature*Species slice=Time*Temperature*Species adjust=tukey; run;

SAS Code for the Analysis of Data for S. Invicta and a Competitor (N. Fulva) when S. Invicta was Given First Access to Resources proc mixed data=SIvsCompetitior; class Subject Colony Species Temperature Time Bait; model Resources = Time Colony Species Temperature Bait Species*Temperature Time*Colony Time*Species Time*Temperature Time*Bait Time*Species*Bait Time*Species*Temperature Time*Species*Temperature*Bait/ ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temperature / slice=Time adjust=tukey ; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Temperature / slice=Temperature slice=species adjust=tukey; lsmeans Time*Species*Temperature / slice=Time*Temperature adjust=tukey; lsmeans Time*Species*Temperature*Bait/ slice=Time*Temperature slice=Time*Species slice=Temperature*Species slice=Time*Temperature*Species adjust=tukey; run; proc mixed data=SIvsCompetitior;

131

class Subject Colony Species Temperature Time Bait; model ants = Time Colony Species Temperature Bait Species*Temperature Time*Colony Time*Species Time*Temperature Time*Bait Time*Species*Bait Time*Species*Temperature Time*Species*Temperature*Bait / ddfm=kr residual outpred=Residuals; repeated / subject=Subject type=ARH(1); lsmeans Time*Temperature / slice=Time adjust=tukey ; lsmeans Time*Species / slice=Time slice=species adjust=tukey; lsmeans Time*Bait / Slice=Time adjust=tukey; lsmeans Species*Temperature / slice=Temperature slice=species adjust=tukey; lsmeans Time*Species*Temperature / slice=Time*Temperature adjust=tukey; lsmeans Time*Species*Temperature*Bait/ slice=Time*Temperature slice=Time*Species slice=Temperature*Species slice=Time*Temperature*Species adjust=tukey; run;

132

LIST OF REFERENCES

Abbott, K. L. 2005. Supercolonies of the invasive yellow crazy ant, Anoplolepis gracilipes, on an oceanic island: forager activity patterns, density and biomass. Insectes Soc. 52:266-273.

Adams, E. S. 1990. Boundary disputes in the territorial ant Azteca trigona: effects of asymmetries in colony size. Anim. Behav. 39:321-328.

Adams, E. S., and J. F. A. Traniello. 1981. Chemical interference compeition by Monomorium minum (Hymenoptera: Formicidae). Oceologia 51:265-270.

Arcila, A. M., L. A. Gomez, and P. Ulloa-Chacon. 2002. Immature development and colony growht of crazy ant Paratrechina fulva under laboratory conditions. Sociobiol. 39:1-15.

Banks, W., D. Jouvenaz, C. Lofgren, and D. Hicks. 1973. Evaluation of coatings for corncob grit-soybean oil bait used to control imported fire ants. J. Econ. Entomol. 66:241-244.

Barbieri, R. F., J. Grangier, and P. J. Lester. 2014. Synergistic effects of temperature, diet and colony size on the competitive ability of two ant species. Austral Ecol. 40:90-99.

Bolton, B., G. Alpert, P. S. Ward, and P. Naskrecki. 2006. Bolton's catalogue of ants of the world. Harvard University Press, Cambridge, Mass.

Briere, J., P. Pracros, A. Le Roux, and J. Pierre. 1999. A novel rate model of temperature- dependent dvelopment for arthropods. Environ. Entomol. 28:22-29.

Calibeo, D., and F. Oi. 2011. Caribbean crazy ants: A new pest in the Southeast, pp. 11-12. In C. L. Harmon, S. D. Stocks and A. Hodges (eds.), First Detector Networks News, vol. 6.

Carpintero, S., and J. Reyes-López. 2008. The role of competitive dominance in the invasive ability of the Argentine ant (Linepithema humile). Biol. Inv. 10:25-35.

Cassill, D., W. R. Tschinkel, and S. B. Vinson. 2002. Nest complexity, group size and brood rearing in the fire ant, Solenopsis invicta. Insect. Soc. 49:158-163.

Cassill, D. L., and W. R. Tschinkel. 1999. Regulation of diet in the fire ant, Solenopsis invicta. J. Insect Behav. 12:307-328.

Cerdá, X., J. Retana, and A. Manzaneda. 1998a. The role of competition by dominants and temperature in the foragin of subordinate species in Mediterranean ant communities. Oecologia 117:404-412.

Cerdá, X., J. Retana, and S. Cros. 1998b. Critical thermal limits in Mediterranean ant species: trade-off between mortality risk and foraging performance. Funct. Ecol. 12:45-55.

133

Chen, J., T. Rashid, G. Feng, L. Zhao, D. Oi, and B. M. Dees. 2013. Defensive chemicasl of tawny crazy ants, Nylanderia fulva (Hymenoptera: Formicidae) and their toxicity to red imported fire ants, Solenopsis invicta (Hymenoptera: Formicidae). Toxicon 76:160-166.

Chown, S. L., K. R. Jumbam, A. A. Sorensen, and J. S. Terblanche. 2009. Phenotypic variance, plasticity and heritability estimates of critical thermal limits depend on methodological context. Functional Ecol. 23:133-140.

Chown, S. L., and J. S. Terblanche. 2007. Physiological diversity in insects: ecological and evolutionary contexts. Adv. Insect Physiol. 33:50-152.

Clémencet, J., L. Cournault, A. Odent, and C. Doums. 2010. Worker thermal tolerance in the thermophilic ant Cataglyphis cursor (Hymenoptera, Formicidae). Insect. Soc. 57:11-15.

Cokendolpher, J. C., and S. A. Phillips. 1990. Critical thermal limits and locomotor activity of the red imported fire ant (Hymenopera: Formicidae). Environ. Entomol. 19:878-881.

Conover, W. J., and R. L. Iman. 1981. Rank transformations as a bridge between parametric and nonparametric statistics. Am. Stat. 35:124-129.

Cook, S. C., M. D. Eubanks, R. E. Gold, and S. T. Behmer. 2011. Seasonality Directs Contrasting Food Collection Behavior and Nutrient Regulation Strategies in Ants. PLoS ONE 6:1-8.

Deyrup, M., L. Davis, and S. Cover. 2000. Exotic Ants in Florida. Transactions of the American Entomological Society (1890-) 126:293-326.

Dowell, R. V., and C. J. Krass. 1992. Exotic pests pose growing problem for California. Caloifornia Agriculture 46:6-10.

Drees, B. M. 2009. Rasberry crazy ant - a new threat to nurseries. Am. Nurseryman 6-8.

Fellers, J. H. 1987. Interference and Exploitation in a Guild of Woodland Ants. Ecology 68:1466-1478.

Fitzpatrick, G., M. C. Lanan, and J. L. Bronstein. 2014. Thermal tolerance affects mutualisit attendance in an ant-plant protection mutualism. Oecologia 176:129-138.

Frank, J. H., and E. D. McCoy. 1995. Precinctive Insect Species in Florida. Florida Entomol 78:21-35.

Friedman, L. 2013. 'Crazy ants' invade Southern states, altering ecosystem, vol. News, July, 2, 2013 ed. USA Today, http://www.usatoday.com/story/news/health/2013/06/30/crazy-ants- south/2446941/.

134

Frouz, J. 2000. The effect of nest moisture on daily temperature regime in the nests of Formica polyctena wood ants. Insect. Soc. 47:229-235.

Furman, B. D., and R. E. Gold. 2006. Determination of the most effective chemical form and concentrations of indoxacarb, as well as the most appropriate grit size, for use in Advion™. Sociobiology 48:309-334.

Global Invasive Species Database. 2013. Global Invasive Species Database. http://www.issg.org/database.

Glunn, F., D. F. Howard, and W. R. Tschinkel. 1981. Food preference in colonies of the fire antSolenopsis invicta. Insectes Soc. 28:217-222.

Gotelli, N. J. 1996. Ant Community Structure: Effects of Predatory Ant Lions. Ecology 77:630- 638.

Gotzek, D., S. G. Brady, R. J. Kallal, and J. S. LaPolla. 2012. The importance of using multiple approaches for identifying emerging invasive species: the case of the Rasberry Crazy Ant in the United States. PLOS One 7:1-10.

Grover, C. D., A. D. Kay, J. A. Monson, T. C. Marsh, and D. A. Holway. 2007. Linking nutiriion and behavioral dominance: carbohydrate scarcity limits aggression and activity in Argentine ants. Proc. R. S. Biol. Sci. 274:2951-2957.

Hahn, D. A., A. R. Martin, and S. D. Porter. 2008. Body size, but not cooling rate, affects supercooling points in the Red Imported Fire Ant, Solenopsis invicta. Physiol. Ecol. 37:1074-1080.

Harmon, K. 2009. Honeybees face new threat in Texas: "Cracy" ants. Scientific American, http://www.scientificamerican.com/blog/post.cfm?id=honeybees-face-new-threat-in-texas- 2009-08-07.

Heller, N. E., and D. M. Gordon. 2006. Seasonal spatial dynamics and causes of nest movement in colonies of the invasive Argentine ant (Linepithema humile). Ecol. Entomol. 31:499-510.

Hill, K. S. 2013. Foraging strategies and aggression patterns of Nylanderia fulva (Mayr) (Hymenoptera: Formicidae) in North Central Florida, Doctor of Philosophy ed. University of Florida, Gainesville, Florida.

Hoffmann, A. A., J. G. Sørensen, and V. Loeschcke. 2003. Adaptation of Drosophila to temperature extremes: bringing together quantitative and molecular approaches. J. Therm. Biol. 28:175-216.

Hölldobler, B., and E. O. Wilson. 1990. The Ants. Harvard University Press, Cambridge, Massachusetts.

135

Holway, D. A., and T. J. Case. 2001. Effects of colony-level variation on competitive abiliyt in the invasive Argentine ant. Anim. Behav. 61:1181-1192.

Holway, D. A. 1998. Effect of Argentine ant invasions on ground-dwelling arthropods in northern California riparian woodlands`. Oecologia 116:252-258.

Holway, D. A. 1999. Competitive mechanisms underlying the displacement of native ants by the invasive Argentine ant. Ecology 80:238-251.

Holway, D. A., L. Lach, A. V. Suarez, N. D. Tsutsui, and T. J. Case. 2002. The causes and consequences of ant invasions. Annu. Rev. Ecol. Syst. 33:181-233.

Holway, D. A., and A. V. Suarez. 1999. Animal behavior: an essential component of invasion biology. Trends in Ecology & Evolution 14:328-330.

Hooper-Bùi, L. M., R. Strecker, X. Chen, D. Aguillard, and A. Miller. 2010. Super colonies of crazy ants in Louisiana, pp. 13-16. In Anonymous 2010 Imported Fire Ant and Invasive Ant Conference, April 19-22, 2010, Little Rock Arkansas.

Hooper-Bui, L. M., and M. K. Rust. 2000. Oral toxicity of abamectin, boric acid, fipronil, and hydramethylnon to laboratory colonies of Argentine ants (Hymenoptera: Formicidae). J. Econ. Entomol. 93:858-864.

Horn, K. C., M. D. Eubanks, and E. Siemann. 2013. The effect of diet and opponent size on aggressive interactions involving Caribbean crazy ants (Nylanderia fulva). PloS one 8:1-7.

Hoshikawa, K., H. Tsutsui, K. Honma, and S. F. Sakagami. 1988. Cold resistance in four species of beetles overiwntering in the soil, with notes on the overwintering strategies of some soil insects. Appl. Ent. Zool 23:273-281.

Howard, D. F., and W. R. Tschinkel. 1980. The effect of colony size and starvation on food flow in the fire ant, Solenopsis invicta (Hymenoptera: Formicidae). Behav. Ecol. Sociobiol. 7:293-300.

Hu, G. Y., and J. H. Frank. 1996. Effect of the red imported fire ant (Hymenoptera: Formicidae on dung-inhabiting arthropods in Florida. Environ. Entomol 25:1290-1296.

Human, K. G., and D. M. Gordon. 1996. Exploitation and interference competition between the invasive Argentine ant, Linepithema humile, and native ant species. Oecologia 105:405-412.

Jumbam, K. R., S. Jackson, J. S. Terblanche, M. A. McGeoch, and S. L. Chown. 2008. Acclimation effects on critical and lethal thermal limits of workers of the Argentine ant, Linepithema humile. J. Insec. Physiol. 54:1008-1014.

Kay, C. R., and W. G. Whitford. 1978. Critical thermal limits of desert honey ants: possible ecological implications. Physiol. Zool. 51:206-213.

136

Kelty, J. D., and R. E. Lee. 1999. Induction of rapid cold hardening by cooling at ecologically relevant rates in Drosophila melanogaster. J. Insect Physiol. 45:719-726.

Kenward, M. G., and J. H. Roger. 2009. An Improved Approximation to the Precision of Fixed Effects from Restricted Maximum Likelihood. Comput. Stat. and Data Analysis 53:2583- 2595.

Klotz, J. H., L. Greenberg, and E. C. Venn. 1998a. Liquid boric acid bait for control of the Argentine ant (Hymenoptera: Formicidae). J. Econ. Entomol. 91:910-914.

Klotz, J. H., J. R. Mangold, K. M. Vail, J. L. R. Davis, and R. S. Patterson. 1995. A survey of the urban pest ants (Hymenoptera: Formicidae) of peninsular Florida. Florida Entomologist 78:109-118.

Klotz, J. H., M. K. Rust, H. S. Costa, D. A. Reierson, and K. Kido. 2002. Strategies for controlling Argentine ants (Hymenoptera: Formicidae) with sprays and baits. J. Agric. Urban Entomol. 19:85-94.

Klotz, J. H., M. K. Rust, D. Gonzalez, L. Greenberg, H. Costa, P. Phillips, C. Gispert, D. A. Reierson, and K. Kido. 2003. Directed sprays and liquid baits to manage ants in vineyards and citrus groves. J. Agric. Urban Entomol. 20:31-40.

Klotz, J. H., K. M. Vail, and D. F. Willams. 1997. Toxicity of a boric acid-sucrose water bait to Solenopsis invicta (Hymenoptera: Formicidae). J. Econ. Entomol. 90:488-491.

Klotz, J., L. Greenberg, and E. C. Venn. 1998b. Liquid boric acid bait for control of the Argentine ant (Hymenoptera: Formicidae). J. Econ. Entomol. 91:910-914.

Knight, R. L., and M. K. Rust. 1991. Efficacy of formulated baits for control of Argentine ant (Hymenoptera: Formicidae). J. Econ. Entomol. 84:510-514.

Lach, L. 2005. Interference and exploitation competition of three nectar-thieving invasive ant species. Insect. Soc. 3:257-262.

LaPolla, J. S., S. G. Brady, and S. O. Shattuck. 2010. Phylogeny and taxonomy of the Prenolepis genus-group of ants (Hymenoptera: Formicidae). System. Entomol. 35:118- 131.

LaPolla, J. S., S. G. Brady, and S. O. Shattuck. 2011. Monograph of Nylanderia (Hymenoptera: Formicidae) of the World: An introduction to the systematics and biology of the genus. Zootaxa 3110:1-9.

LeBrun, E. G., J. Abbot, and L. E. Gilbert. 2013. Imported crazy ant displaces imported fire ant, reduces and homogenizes grassland ant and arthropod assemblages. Biol. Inv. 15:2429- 2442.

137

LeBrun, E. G., N. T. Jones, and L. E. Gilbert. 2014. Chemical warfare among invaders: a detoxification interaction facilitates an ant invasion. Sci 343:1017-1017.

Lester, P. J., and A. Tavite. 2004. Long-Legged Ants, Anoplolepis gracilipes (Hymenoptera: Formicidae), Have Invaded Tokelau, Changing Composition and Dynamics of Ant and Invertebrate Communities. Pac. Sci. 58:391-401.

Lighton, J. R., and R. J. Turner. 2004. Thermolimit respirometry: an objective assessment of criticial thermal maxima in two sympatric desert harvester ants, Pogonomyrmex rugosus and P. californicus. J. Exp. Biol. 207:1903-1913.

Lofgren, C. 1986. History of imported fire ants in the United States, pp. 36–47. In. CS Lofgren and RK Vander Meer [eds.], Fire ants and leaf-cutting ants: biology and management.

Lofgren, C., W. Banks, and B. Glancey. 1975. Biology and control of imported fire ants. Annu. Rev. Entomol. 20:1-30.

Lofgren, C. S., B. R. Bartlett, and C. E. Stringer. 1954. The acceptability of some fats and oils as food to imported fire ants. J. Econ. Entomol. 67:601-602.

Lutterschmidt, W. I., and V. H. Hutchison. 1997. The critical thermal maximum: data to support the onset of spasms as the definitive end point. Can . J. Zool. 75:1553-1560.

MacGown, J. A. 2013. Ants (Formicidae) of the Southeastern United States. http://www.mississippientomologicalmuseum.org.msstate.edu/Researchtaxapages/Formicid aepages/genericpages/Nylanderia_fulva.htm.

MacGown, J. A., and B. Layton. 2010. The invasive Rasberry crazy ant, Nylanderia sp near pubens (Hymenotpera: Formicidae). Midsouth Entomol. 3:44-47.

Macom, T. E., and S. D. Porter. 1990. Comparison of polygyne and monogyne imported red fire ant (Hymenoptera: Formicidae) population densities. Entomol. Soc. Am. 71:2095-2106.

Mailleux, A. C., C. Detrain, and J. L. Deneubourg. 2006. Starvation drives a threshold triggering communication. J. Exp. Biol. 209:4224-4229.

Markin, G. P., J. O'Neal, and J. Dillier. 1975. Foraging tunnels of the red imported fire ant, Solenopsis invicta (Hymenoptera: Formicidae). USDA, APHIS 48:83-89.

Markl, H. 1973. The evolution of stridulatory communication in ants, pp. 258-265. In Anonymous Proceedigns of the Seventh Congress of the International Union for the Study of Social Insects, London.

McDonald, D. L. 2012. Investigation of an invasive ant species: Nylanderia fulva colony extraction, management, diet preference, fecundity, and mechanical vector potential, Doctor of Philosophy ed. Texas A&M University, College Station, TX.

138

McGlynn, T. P. 1999. The Worldwide Transfer of Ants: Geographical Distribution and Ecological Invasions. J. Biogeogr. 26:535-548.

Mellanby, K. 1939. Low temperature and insect activity. Proceedings of the Royal Society of London.Series B, Biological Sciences 127:473-487.

Meyers, J. M. 2008. Identification, distribution and control of an invasive pest ant, Paratrechina SP. (Hymenoptera: Formicidae), in Texas, Dissertation ed. Texas A&M University, Texas.

Meyers, J. M., and R. E. Gold. 2008. Identification of an exotic pest ant, Paratrechina sp. nr. pubens (Hymenoptera: Formicidae), in Texas. Sociobiol. 52:589-604.

Miyata, H., T. Shimamura, H. Hirosawa, and S. Higashi. 2003. Morphology and phenology of the primitive ponerine army ant Onychomyrmex hedleyi (Hymenoptera: Formicidae: Ponerinae) in a highland rainforest of Australia. J. Nat. Hist. 37:115-125.

Moller, H. 1996. Lessons for invasion theory from social insects. Biol. Conservation 78:125-142.

Morales, M. A., J. L. Barone, and C. S. Henry. 2008. Acoustic alarm signalling facilitates predator protection of treehoppers by mutualist ant bodyguards. Proc. Biol. Sci. 275:1935- 1941.

Ness, J. H., and J. L. Bronstein. 2004. The Effects of Invasive Ants on Prospective Ant Mutualists. Biol. Invasions 6:445 461.

Ness, J. H., W. F. Morris, and J. L. Bronstein. 2009. For ant-protected plants, the best defense is a hungry offense. Ecology 90:2823-2831.

Oi, D. 2015. United States Department of Agriculture - Agricultural Research Service Center for Medical, Agricultural and Veterinary Entomology.

Oi, D. H., and F. M. Oi. 2006. Speed of efficacy and delayed toxicity characteristics of fast- acting fire ant (Hymenoptera: Formicidae) baits. J. Econ. Entomol. 99:1739-1748.

Passera, L. 1994. Characteristics of tramp species, pp. 23-43. In D. F. Williams (ed.), Exotic ants: biology, impact, and control of introduced species. Westview Press, Boulder, Colorado.

Penick, C. A., and W. R. Tschinkel. 2008. Thermoregulatory brood transport in the fire ant, Solenopsis invicta . Insect. Soc. 55:176-182.

Pimentel, D. 1997. Techniques for Reducing Pesticides: Enviornmental and Economic Benefits. John Wiley and Sons, Chichester, UK.

139

Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecol. Econ. 52:273-288.

Plowes, R. M., J. J. Becnel, E. G. LeBrun, D. H. Oi, S. M. Valles, N. T. Jones, and L. E. Gilbert. 2015. Myrmecomorba nylanderiae gen. et sp. nov., a microsporidian parasite of the tawny crazy ant Nylanderia fulva. J. Invertebr. Pathol. 129:45-56.

Porter, S. D. 1988. Impact of the temperature on colony growth and developmental rates of the ant, Solenopsis invicta. J. Insect Physiol 34:1127-1133.

Porter, S. D., and D. A. Savignano. 1990. Invasion of polygyne fire ants decimates native ants and disrupts arthropod community. Ecol. 71:2095-2106.

Porter, S. D., and W. R. Tschinkel. 1987. Foraging in Solenopsis invicta (Hymenoptera: Formicidae): Effects of weather and season. Environ. Entomol. 16:802-808.

Porter, S. D., and W. R. Tschinkel. 1993. Fire ant thermal preferences: behavioral control of growth and metabolism. Behav. Ecol. Sociobiol. 32:321-329.

Powell, S. J., and J. S. Bale. 2006. Effect of long-term and rapid cold hardening on the cold torpor temperature of an aphid. Physiol. Entomol. 31:348-352.

Powell, S. J., and J. S. Bale. 2004. Cold shock injury and ecological costs of rapid cold hardening in the grain aphid Sitobion avenae (Hemiptera: Aphididae). J. Insect Physiol. 50:277-284.

Pranschke, A. M., and L. M. Hooper-Bùi. 2003. Influence of abiotic factors on red imported fire ant (Hymenoptera: Formicidae) mound population ratings in Louisiana. Environ. Entomol. 32:204-207.

Puche, H., and N. Y. Su. 2001. Application of fractural analysis for tunnel systems of subterranean termites (Isoptera: Rhinotermitidae) under laboratory conditions. Environ. Entomol. 30:545-549.

Redford, K. H., and J. G. Dorea. 1984. The nutritional value of invertebrates with emphasis on ants and termites as food for mammals. J. Zool. 203:385-395.

Retana, J., and X. Cerdá. 2000. Patterns of Diversity and Composition of Mediterranean Ground Ant Communities Tracking Spatial and Temporal Variability in the Thermal Environment. Oecologia 123:436-444.

Rezende, E. L., M. Tejedo, and M. Santos. 2011. Estimating the adaptive potential of critical thermal limits: methodological problems and evolutionary implications. Functional Ecol. 25:111-121.

140

Ross, K. G., E. L. Vargo, and L. Keller. 1996. Social evolution in a new environment: the case of introduced fire ants. Proc. Natl. Acad. Sci 93:3021-3025.

Rowles, A. D., and J. Silverman. 2009. Carbohydrate supply limits invasion of natural communities by Argentine Ants. Oecologia 161:161-171.

Rust, M. K., D. A. Reierson, and J. H. Klotz. 2002. Factors affecting the performance of bait toxicants for Argentine ants (Hymenoptera: Formicidae). pp. 115-120. In Anonymous , http://www.icup.org.uk/reports/ICUP211.pdf. Susan C. Jones, Jing Azai, and Wm H. Robinson [eds.].

Salt, R. W. 1966. Effect of cooling rate on the freezing tmepratures of supercooled insects. Can. J. Zool. 44:655-659.

Savolainen, R., and K. Vepsäläinen. 1988. A competition hierarchy among boreal ants: impact on resource partitioning and community structure. Oikos 51:135-155.

Sharma, S., D. H. Oi, and E. A. Buss. 2013. Honeydew-producing hemipterans in Florida associated with Nylanderia fulva (Hymenoptera: Formicidae), an invasive crazy ant. Fla. Entomol. 96:538-547.

Sinclair, B. J., P. Vernon, C. J. Klok, and S. L. Chown. 2003. Insects at low temperature: an ecological perspective. Trends in Ecology & Evolution 18:257-262.

Somero, G. N. 2005. Linking biogeography to physiology: evolutionary and acclimatory adjustments of thermal limits. Front. Zool 2:1-9.

Sorvari, J., P. Theodora, S. Turillazzi, H. Hakkarainen, and L. Sundstrom. 2008. Food resources, chemical signaling, and nest mate recognition in the ant Formica aquilonia. Behav. Ecol. 19:441-447.

Stein, M. B., H. G. Thorvilson, and J. W. Johnson. 1990. Seasonal changes in bait preference by red imported fire ant, Solenopsis invicta (Hymenoptera: Formicidae). Fla. Entomol. 117- 123.

Styrsky, J. D., and M. D. Eubanks. 2007. Ecological Consequences of Interactions between Ants and Honeydew-Producing Insects. Proceedings: Biological Sciences 274:151-164.

Suarez, A. V., D. A. Holway, and T. J. Case. 2001. Patterns of spread in biological invasions dominated by long-distance jump dispersal: Insights from Argentine ants. Proc. Natl. Acad. Sci. U. S. A. 98:1095-1100.

Terblanche, J. S., J. A. Deere, S. Clusella-Trullas, C. Janion, and S. L. Chown. 2007. Critical thermal limits depend on methodological context. Proc. R. Soc. Bio. 274:2935-2942.

141

Terblanche, J. S., A. A. Hoffman, K. A. Mitchell, L. Rako, P. C. Roux, and S. L. Chown. 2011. Ecologically relevant measures of tolerance to potentially lethal temperatures. J. Exp. Bio. 214:3713-3725.

Tollerup, K., M. Rust, K. Dorschner, P. Phillips, and J. Klotz. 2004. Low-toxicity baits control ants in citrus orchards and grape vineyards. Calif. Agric. 58:213-217.

Trager, J. C. 1984. A revision of the genus Paratrechina (Hymenoptera: Formicidae) of the contintental United States. Sociobiol. 9:51-162.

Travassos, M. A., and N. E. Pierce. 2000. Acoustics, context and function of vibrational signalling in a lycaenid butterfly–ant mutualism. Anim. Behav. 60:13-26.

Tschinkel, W. R. 1998. The reproductive biology of fire ant societies. Bioscience 48:593-605.

Tschinkel, W. R. 2006. The Fire Ants. Harvard University Press, Cambridge, MA.

Tsutsui, N. D., and T. J. Case. 2001. Population genetics and colony structure of the argentine ant (Linepithema humile) in its native and introduced range. Evol. 55:976-985.

Tsutsui, N. D., and A. V. Suarez. 2003. The colony structure and population biology of invasive ants; estructura de una colonia y biología poblacional de hormigas invasoras. Conserv. Biol. 17:48-58.

Tsutsui, N. D., A. V. Suarez, Holway D. A., and T. J. Case. 2000. Reduced genetic variation and success of invasive species. PNAS 97:5948-5953.

Valles, S. M., and D. H. Oi. 2014. Novel Nylanderia Fulva Virus, vol. 20140234262Gainesville, FL, US.

Verble-Pearson, R. M., M. E. Gifford, and S. P. Yanoviak. 2015. Variation in thermal tolerance of North American ants. J. Thermal Biol. 48:65-68.

Vinson, S. B. 1999. The economic impact of the inported firee ant infestation on the State of Texas. Texas A&M University, College Station, TX.

Ward, P. S. 2014. The phylogeny and evolution of ants. Annual Review of Ecology, Evolution, and Systematics 45:23-43.

Warner, J., and R. H. Scheffrahn. 2004. Caribbean Crazy ant, Paratrechina pubensi Forel. http://creatures.ifas.ufl.edu/urban/ants/caribbean.crazy.ant.htm.

Warner, J. R. 2005. Efficacy of selected insecticides using arboreal bioassays and rubidium marking of white-footed ants, Technomyrmex albipes (Hymenoptera: Formicidae), Ph.D. ed. University of Florida, United States -- Florida.

142

Wetterer, J. K. 2007. Paratrechina pubens (Forel, 1893) (Hymenoptera: Formicidae), a candidate for the plague ant of the 19th century Bermuda. Myrmecol. News 10:39-40.

Wetterer, J. K., and J. W. Keularts. 2008. Population explosion of the hairy crazy ant, Paratrechina pubens (Forel), on St. Croiz, US Virgin Islands (Hymenoptera: Formicidae). Fla. Entomol. 91:423-427.

Wilson, E. O. 1958. The fire ant. Sci. Am. 3:36-41.

Wilson, E. O. 1971. The Insect Societies. Harvard University Press, Cambridge, Massachusetts.

Zenner-Polania, I. 1990a. Biological aspects of the ‘’Hormiga Loca,’’ Paratrechina (Nylanderia) fulva (Mayr), in Colombia, pp. 290-297. In R. K. Vander Meer, K. Jaffe and A. Cedeno (eds.), Applied Myrmecology, a World Perspective. Westview Press.

Zenner-Polania, I. 1990b. Management of the "Hormiga Loca" Paratrechina (Nylanderia) fulva (Mayr) in Columbia, pp. 701-707. In R. K. Vander Meer, K. Jaffe and A. Cedeno (eds.), Applied Myrmecology: A world perspective. Westview Press, Boulder, CO.

Zenner-Polania, I. 1994. Impact of Paratrechina fulva on other ant species, pp. 120-131. In D. F. Williams (ed.), Exotic ants. Biology, impact, and control of introduced species. Westview Press, Boulder, CO.

Zhang, Q., D. L. McDonald, D. R. Hoover, J. R. Aldrich, and R. G. Schneidmiller. 2015. North American Invasion of the Tawny Crazy Ant (Nylanderia fulva) Is Enabled by Pheromonal Synergism from Two Separate Glands. J. Chem. Ecol. 41:853-858.

143

BIOGRAPHICAL SKETCH

Michael Thomas Bentley was born in Noblesville, Indiana but grew up in Vero Beach,

Florida. He earned his bachelor’s degree in criminology in 2005 and Master of Science degree in entomology and nematology in 2008 from the University of Florida before accepting a job in the pest management industry. After completing his Doctor of Philosophy Michael plans to continue his career in the research and development sector of the pest management industry.

144