water management in termites: a ......4-1 initial mass (mg), percentage of total body water (%tbw),...

WATER MANAGEMENT IN TERMITES: A COMPREHENSIVE LOOK AT THE BEHAVIOR AND PHYSIOLOGY OF RESISTANCE TO DESICCATION

By

JOHN G. ZUKOWSKI

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

2015

To Mr. David Bassett, whose dedication to his students was a mainstay and whose dry wit and enthusiasm for science were contagious. And to Douglas Greig, who taught me

so much, but not nearly enough, and with whom I wish I had more time.

4

ACKNOWLEDGMENTS

I wish to thank my family and friends, without whose support and understanding I

would not have made it this far. I wish to thank my advisor, Dr. Nan-Yao Su, for the

opportunity to work, learn, and grow as a scientist under his tutelage at this fine

institution. I wish to thank my lab-mates and the rest of the wonderful people I have

interacted with in the FLREC and UF communities. All of the conversations, inspirations,

patience, and advice are truly appreciated. I especially thank my beautiful wife, Arisha,

who has remained my dedicated counterpart and best friend, despite the time and

distance. And lastly, Scooby…good dog!

5

TABLE OF CONTENTS page

ACKNOWLEDGMENTS .................................................................................................. 4

LIST OF TABLES ............................................................................................................ 7

LIST OF FIGURES .......................................................................................................... 9

ABSTRACT ................................................................................................................... 11

CHAPTER

1 GENERAL INTRODUCTION .................................................................................. 13

Termite Biology ....................................................................................................... 13 Termite Species Used............................................................................................. 14

Coptotermes formosanus Shiraki ..................................................................... 15

Cryptotermes brevis (Walker) ........................................................................... 16 Cryptotermes cavifrons Banks .......................................................................... 18

Neotermes jouteli (Banks) ................................................................................ 19

Termite Water Relations ......................................................................................... 20

Objectives ............................................................................................................... 27

2 HISTOLOGICAL APPROACH TO EXAMINING PHYSIOLOGICAL RESISTANCE TO DESICCATION ......................................................................... 32

Introduction ............................................................................................................. 32 Materials and Methods............................................................................................ 34

Results .................................................................................................................... 36 Discussion .............................................................................................................. 37

3 WATER LOSS TOLERANCE OF FOUR TERMITE SPECIES EXPOSED TO VARIOUS RELATIVE HUMIDITIES ........................................................................ 50

Introduction ............................................................................................................. 50

Materials and Methods............................................................................................ 51


4 CUTICULAR PERMEABILITY, BODY WATER LOSS, AND RELATIVE HUMIDITY EQUILIBRIA OF FOUR TERMITE SPECIES ....................................... 60

Introduction ............................................................................................................. 60 Materials and Methods............................................................................................ 62

Cuticular Permeability and Body Water Loss ................................................... 62 RH Equilbria ..................................................................................................... 63

6


5 RELATIVE HUMIDITY PREFERENCE OF FOUR TERMITE SPECIES IN A MULTIPLE-CHOICE ARENA .................................................................................. 85

Introduction ............................................................................................................. 85 Materials and Methods............................................................................................ 86 Results .................................................................................................................... 88 Discussion .............................................................................................................. 89

6 UTILIZATION OF WATER SOURCES BY FOUR SPECIES OF TERMITE ........... 96

Introduction ............................................................................................................. 96

Materials and Methods............................................................................................ 97 Results .................................................................................................................... 99 Discussion ............................................................................................................ 102

7 SURVIVORSHIP, PREFERENCE, AND BEHAVIOR IN RESPONSE TO A RH AND MOISTURE AVAILABILITY SHIFT ............................................................... 112

Introduction ........................................................................................................... 112

Materials and Methods.......................................................................................... 114 Results .................................................................................................................. 117 Discussion ............................................................................................................ 119

8 CONCLUSIONS ................................................................................................... 130

LIST OF REFERENCES ............................................................................................. 138

BIOGRAPHICAL SKETCH .......................................................................................... 147

7

LIST OF TABLES

Table page 2-1 Cuticle thicknesses (mean±SEM) of four termite species ................................... 42

2-2 Rectal pad widths (mean±SEM) of four termite species ..................................... 43

3-1 Analysis of variance for a factorial experiment to evaluate the effects of humidity on worker survival of four termite species over 12 days ....................... 57

3-2 Effects of five RH levels on worker survival (%) of four termite species over 12 days (mean±SEM) ......................................................................................... 58

4-1 Initial mass (mg), percentage of total body water (%TBW), and cuticular permeability (CP) value of an average live individual worker of four termite species exposed to 0.3-3.3% RH and ≈26.1°C and an average dead individual worker of the same four species exposed to 4.5-17.8% RH and ≈26.2°C .............................................................................................................. 75

4-2 Regression equations of cumulative percentage of total body water content (%TBW) lost over time (hours) for live termites of four species exposed to 0-3.3% RH and 26°C and dead termites of the same species exposed to 4.5-17.8% RH and ≈26.2°C ...................................................................................... 76

4-3 Species-specific RH equilibriums produced from evaporation of body water from groups of 25 live workers and groups of 25 dead workers of four species of termite after 16 hours ........................................................................ 77

5-1 Analysis of variance for a factorial experiment to evaluate preference of workers of four species of termite exposed to an arena with five different RH levels for 12-16 hours ......................................................................................... 93

5-2 Effects of species and relative humidity on preference of termites in a multiple-choice arena (mean±SEM) ................................................................... 94

6-1 Analysis of variance for a factorial experiment to evaluate the survival of termites exposed to various types of water sources from 1 to 4 weeks ............ 106

6-2 Effects of water source and species on worker survival (%) of four species of termite after 1 week (mean±SEM) .................................................................... 107

6-3 Effects of water source and species on worker survival (%) of four species of termite after 2 weeks (mean±SEM) .................................................................. 108


8


7-1 Shelter study data compilation from 6 replicates with 300 termites total .......... 123

7-2 Termite survival (%) in a dual chamber arena (mean±SEM) ............................ 124

7-3 Analysis of variance for a factorial experiment to evaluate preference of surviving termites after 4 days in a dual chamber arena .................................. 125

7-4 Preference of live individuals (%) of four termite species for two refuge chambers (mean±SEM) .................................................................................... 126

9

LIST OF FIGURES

Figure page 1-1 Termite species used for experimentation, with pseudergate (left), soldier

(center) and winged imago (right) shown. .......................................................... 30

1-2 Diagram of a general termite water budget. ....................................................... 31

2-1 Transverse sections of stained abdominal cuticle at 200x magnification. ........... 44

2-2 Transverse sections of stained rectal pads at 200x magnification. ..................... 45

2-3 Abdominal spiracles on the tergites and near the pleural membrane of N. jouteli. ................................................................................................................. 46

2-4 Dissected spiracles at various magnifications. ................................................... 47

2-5 Comparison of artist interpretation and photograph of the attachment of the spiracle structure and the trachea of C. formosanus. ......................................... 48

2-6 Comparison of artist interpretation and photograph of the attachment of the spiracle structure and the trachea in the kalotermitid species used (Cr. brevis, Cr. cavifrons, and N. jouteli). ................................................................... 49

3-1 Setup for examining termite survival when exposed to various relative humidities. .......................................................................................................... 59

4-1 Setup for determining relative humidity equilibria in a confined space for four termite species. .................................................................................................. 78

4-2 Species-specific average %RH equilibria curves with SEM bars for live termites. .............................................................................................................. 79

4-3 Relationship of cumulative percentage of total body water content (%TBW) lost over time for live individuals of four termite species exposed to 0-3.3% RH and 26°C. ..................................................................................................... 80

4-4 Relationship between cuticular permeability (CP) values of an average live individual from 10 replicates of 10 live individuals and mean RH equilibria (RH-EQ) of 4 replicates of 25 live individuals for four species of termites. ......... 81

4-5 Species-specific average %RH equilibria curves with SEM bars for dead termites. .............................................................................................................. 82

4-6 Relationship of cumulative percentage of total body water content (%TBW) lost over time for dead individuals of four termite species exposed to 4.5-17.8% RH and ≈26.2°C. ..................................................................................... 83

10

4-7 Relationship between cuticular permeability (CP) values of an average dead individual from 10 replicates of 10 dead individuals and mean RH equilibria (RH-EQ) of 4 replicates of 25 dead individuals for four species of termites. ....... 84

5-1 Setup for determining RH level preferences of four termite species. .................. 95

6-1 Setup for examining utilization of various water sources (or lack thereof) by four termite species. ......................................................................................... 111

7-1 Setup for examining the use of refuge and shelter, as well as associated behaviors by four species of termite when exposed to environmental change.. ............................................................................................................ 127

7-2 Diagram illustrating dual chamber shelter study protocols for C. formosanus/N. jouteli couplet and Cr. brevis/Cr. cavifrons couplet. .................. 128

7-3 Examples of shelter and chamber sealing. ....................................................... 129

11

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

WATER MANAGEMENT IN TERMITES: A COMPREHENSIVE LOOK AT THE

BEHAVIOR AND PHYSIOLOGY OF RESISTANCE TO DESICCATION

By

John G. Zukowski

December 2015

Chair: Nan-Yao Su Cochair: William Kern Jr. Major: Entomology and Nematology

Although some organisms are more reliant on water, it is vital to the survival and

reproduction of every organism on Earth. The size of an organism and its habitat are

two important components in determining to what degree water plays a role in the life of

an animal. As body size decreases, the surface area to volume ratio increases. Thus,

for smaller organisms such as insects, the volume available for retaining water

resources within the body is low compared with the area through which they can be lost.

This disadvantage in terms of water balance (homeostasis) must be compensated for,

and makes regulation of water loss difficult in many environments. Termites are one of

the most desiccation prone insects due to their small, soft bodies. Therefore, they must

avoid desiccation by efficiently locating and utilizing water resources (as well as other

resources) as individuals and as a group. Comprehensive studies on water

management in termites from different habitats are lacking. This study focuses on the

physiology and behavior of water management in four termite species from two families.

These include Rhinotermitidae: Coptotermes formosanus Shiraki, and Kalotermitidae:

Neotermes jouteli (Banks), Cryptotermes cavifrons (Banks), and Cr. brevis (Walker).

12

Survivorship and preferences at various relative humidity (RH) levels and environmental

conditions revealed that C. formosanus and N. jouteli exhibited a preference as well as

the requirement for high RH conditions and readily available water resources.

Cryptotermes brevis, while not exhibiting a significant difference in survival in terms of

RH level, did exhibit a preference for and survived better in drier conditions and

situations in which direct contact with water could be avoided. Cryptotermes cavifrons,

however, did not exhibit a significant difference in survival nor a major preference for

certain conditions in terms of RH, but also needed to avoid direct contact with water.

Species-specific RH equilibria resulting from the evaporation of body water through the

cuticle into a confined space and its relationship to cuticular permeability (CP) and total

percent body water (%TBW) were also determined. Each species also responded (as a

group) to a quick and drastic change in environmental RH and water availability

conditions by utilizing and/or creating a refuge to buffer against a less favorable

environment. If a group was unable to utilize or create such a refuge quickly enough,

they suffered collapse. Histological observations indicated that differences in water loss

and retention come primarily from cuticular and rectal pad differences. They may also

be affected by spiracular morphology and physiology as well. Ultimately, the ability of

termites to establish themselves in such a multitude of habitats is due to the inherent

plasticity and modulation in behavior and physiology of both the individual and the

colony.

13

CHAPTER 1 GENERAL INTRODUCTION

Termite Biology

Termites are an economically and ecologically important group of insects (Su

2002). Belonging to the Class Insecta, and the Order Isoptera, they are distributed on

six of seven continents, across the tropical, subtropical, and temperate regions of the

Earth (Bignell and Eggleton 2000). There are more than 3,100 living and fossil termite

species, in nine families, with just over 360 being significant pest species (11.7%)

(Krishna et al. 2013). They are sometimes placed in the Order Blattodea because of

evidence that they are a sister taxa to wood roaches, family Cryptocercidae (Inward et

al. 2007, Klass et al. 2008). All termites are eusocial, having overlapping generations,

cooperative brood care, and a caste system in place to divide colony labor. Such a

widely distributed group of insects certainly encounters a diverse range of habitats,

where, being so small, microhabitats are often of importance (Bignell and Eggleton

2000, Abe et al. 1997). This has led to a wide range of behaviors, physiology, caste-

specific morphology, and intimate interactions with diverse microbial ecosystems (Abe

et al. 2000, Bignell 2000). Termites can be found living within the dry wood of a

structure, as well as the damp wood of a rotting log. They have soldiers that use

piercing mandibles, squirt glue from a nasus, block tunnels with phragmotic heads, and

others even lack a soldier caste entirely. Nests are located in trees, underground in

tunnels and galleries, within intricately constructed epigeal mounds, or inside wooden

man-made structures. Foraging may take place underground or above-ground,

exposed, within shelter tubes, or completely within a food source. Termites feed on

wood, leaf-litter, soil, grass, or fungus. Their association with such cellulosic material

14

has developed a dependence on endosymbionts that aid in digestion of cellulose. While

lower termites do have their own endogenous cellulases, they are not expressed at a

level that efficiently digests cellulose on their own, as they are in the higher termites that

make up the majority of the Isoptera (Slaytor 2000).These protozoan and bacterial

endosymbionts are gut fauna, whereas wood-decaying fungi are the exosymbiotic nest

fauna (Bignell and Eggleton 2000). Termites are generally small and soft-bodied, which

results in water loss (cuticular permeability coupled with evaporation from cuticle,

mouth, anus, and spiracles) being a major problem that must be managed. They

address desiccation concerns physiologically and behaviorally, both at the individual

and group (or colony) level. This study focuses on the behavior and physiology of water

management in four species of termite that represent a gradient of different water

requirements in the niches they occupy. This includes a subterranean termite in the

family Rhinotermitidae, as well as two drywood species and one dampwood species

from the family Kalotermitidae.

Termite Species Used

The four termite species selected for this study were chosen because of their

different habitats and requirements for handling water availability or a lack thereof. The

subterranean termite selected was from the genus Coptotermes and happens to be one

of the most economically important species in the U.S. and Florida (Su 2002, Su and

Scheffrahn 1990, 1998). C. formosanus was chosen because of their requirement for

high RH and water availability. The habitat of this termite is higher in humidity compared

with that of the drywoods, due to the moisture found in the soil they use for their

foraging tunnels and nest. The dampwood species selected was from the genus

Neotermes, a commonly occurring, but minor (or nuisance) pest. Both C. formosanus

15

and N. jouteli require and are known to inhabit environments with high RH and water

availability. Whereas the former is found living and foraging within soil, trees or damp

voids, the latter lives within its damp wood food sources. The drywood termites selected

were Cr. cavifrons and Cr. brevis, two species readily found in Florida in natural

occurring wood and structural wood, respectively. As such, the habitats of Cr. cavifrons

are more humid than those of Cr. brevis, due to the former being exposed to free water,

whereas the latter is not. In addition, direct exposure to water has been seen to be toxic

to Cr. brevis. While Cr. brevis and C. formosanus are important economic wood-feeding

pests and require control measures, Cr. cavifrons and N. jouteli are only minor or

nuisance pests. This is mainly due to the latter two species having higher moisture

requirements than pest species such as Cr. brevis or Incisitermes minor (Hagen)

(another kalotermitid) (Su and Scheffrahn 1998).

Coptotermes formosanus Shiraki

The Formosan subterranean termite (FST), C. formosanus, is one of the most

widely distributed (Africa, Asia, Australasia-Pacific, and North America) and

economically important of the over 60 species in the genus Coptotermes (Rust and Su

2012, Krishna et al. 2013). Characterized as a cryptic species with colonies composed

of several million individuals and a foraging range of up to approximately 100 horizontal

meters within soil, the presence of this termite poses serious threats to manmade

structures located near a colony (Su 2003, Su and Scheffrahn 1990, 1998, 2013). This

pest termite species is found worldwide in the subtropical and temperate regions. In the

US it can been found in Alabama, Florida, Georgia, Hawaii, Louisiana, Mississippi,

North and South Carolina, Tennessee, and Texas (Messenger et al. 2002, Su 2003, Su

and Scheffrahn 2013). Underground and interconnected tunnels or galleries are shared

16

by the members of a colony to move and forage. Above-ground foraging tubes connect

the subterranean tunnels to a food source (such as a house) while keeping the micro-

habitat favorable in terms of humidity for the termites, which are highly susceptible to

desiccation. Aerial colonies also occur on occasion if conditions in terms of food and

water are favorable above ground (e.g. flat-topped high rise buildings). Colonies of the

FST are primarily made up of three castes: reproductive, soldiers, and workers.

Workers make up the majority of the population and are responsible for finding food and

caring for the structure (e.g. - tunneling and repair) and development (e.g. - brood care

and retinue) of the colony. Reproductives include the king and queen as well as mature

and immature alates (swarmers and nymphs, respectively), and are useful for

identification. Soldiers are the colony’s main line of defense from predators and are also

useful for identification (Su and Scheffrahn 2013). Figure 1-1A shows the major castes

of this species.

Cryptotermes brevis (Walker)

The West Indian drywood termite, Cryptotermes brevis (Walker), is the most

economically important pest species of family Kalotermitidae. They are often found

living within structural timber and in furniture such as chairs, cabinets, and picture

frames (at one time known as the “furniture termite”). Characteristics of the alates and

soldiers are used to differentiate these species. The origin of Cr. brevis was determined

to be coastal Peru and Chile and it is thought to have spread and been introduced (non-

indigenous) to Florida and the continental US through the dissemination of wooden

goods and wooden ships (Scheffrahn et al. 2009). This species is the most widespread

tropical drywood termite in the world and can be found throughout Florida, Hawaii, and

coastal regions of the southeastern US, and the tropical Americas, but is interestingly

17

absent from Southeast Asia (replaced by Cr. domesticus). Colonies of Cr. brevis

develop within their wood food source and are made up of reproductives (king, queen,

and alates), pseudergate workers, and soldiers. As kalotermitids, the pseudergates are

not reproductively sterile (undifferentiated immatures) and can develop into

reproductives or soldiers if needed, but are normally the labor caste of the colony

responsible for gallery excavation, construction, and for feeding other caste members of

the colony. As with Cr. cavifrons, soldiers exhibit a phragmotic head for blocking tunnels

as defense, but make up a small percentage of the colony (≈1-2%). The reproductives

disperse to mate and start new colonies (swarming). Colony maturation of Cr. brevis

and Cr. cavifrons takes several years due to a low inherent reproductive rate while the

colony is confined to their food source (single-piece feeders) and slow maturation of

immatures (Abe 1987). Due to its preference for extremely dry wood, such as that used

in manmade structures, Cr. brevis is an important structural pest. As in Cr. cavifrons,

water resources are conserved by removing water from feces to create frass pellets,

and by relying on metabolically produced water and moisture adsorbed to wood. A

telltale sign of an infestation of drywood termites including Cr. brevis, is the presence of

piles of this frass found in a structure. Multiple colonies may infest a single structure as

colony size is relatively small and damaging population levels are reached over a matter

of years. Locating a drywood colony can be difficult because they live and feed within

the wood and do not forage outside of it (leaving only for dispersal swarming). Cr. brevis

is known to be of great economic importance in the US and abroad, with many millions

of dollars spent on control of and damage caused by this species (Su and Scheffrahn

1990). Fumigation (or tenting) is the primary method of control, but other options have

18

been used to eliminate an infestation of Cr. brevis (e.g. - spot injections of pesticide,

liquid nitrogen, heat treatment, and electrocution) (Scheffrahn and Su 2007b). Figure 1-

1(D) shows the major castes of this species.

Cryptotermes cavifrons Banks

Cryptotermes cavifrons Banks, classified in the family Kalotermitidae, lives within

naturally found wood, such as logs, stumps, and branches. In Florida, Cr. cavifrons is

most often misidentified as Cr. brevis, a closely related drywood termite. Characteristics

of the alates and soldiers are used to differentiate these species. Cr. cavifrons is

endemic to peninsular Florida and exhibits an unusually broad distribution for a species

of Cryptotermes. This species can be found through the Florida peninsula, Cayman

Islands, Cuba, the Bahamas, Jamaica, and Turks and Caicos Islands (Brammer and

Scheffrahn 2007). Colonies of Cr. cavifrons develop within their wood food source and

are made up of reproductives (king, queen, and alates), nymphs and pseudergate

workers, and soldiers. The pseudergates are immatures that can develop into

reproductives or soldiers as needed, but are normally the labor caste of the colony.

Soldiers exhibit a phragmotic head for blocking tunnels as defense, but only make up

around 1-2% of the colony. The reproductives disperse to mate and start new colonies

(dispersal flights). Colony maturation takes several years due to a low inherent

reproductive rate while the colony is confined to their food source (single-piece feeders).

Cr. cavifrons can be found inhabiting dry word as well as relatively damp wood. In order

to conserve their water resources, they use three pairs of rectal pads to extract as much

water as possible from feces before elimination as frass pellets (Noirot and Noirot-

Timothee 1977, Collins 1969). They also rely heavily on metabolically produced water

and water adsorbed to wood particles (Abe et al. 2000, Arquette 2013, Collins 1958,

19

1969, Nation 2002). Cr. cavifrons is quite common in wood from natural settings, but

rarely makes its way into homes, buildings, or other structures due to its moisture

requirements. Because of this, it is known as a minor (or nuisance) pest and control is

generally not a problem (Brammer and Scheffrahn 2007). Figure 1-1C shows the major

castes of this species.

Neotermes jouteli (Banks)

Neotermes jouteli (Banks) is a member of a diverse genus within the

Kalotermitidae family with approximately 100 species living primarily in tropical regions

of the world. Members of this genus are known as dampwood termites. This Neotermes

species occurs in southeastern Florida, the Bahamas, Cuba, and Turks and Caicos

Islands (Scheffrahn et al. 2006). It is characterized by having a large body size (largest

in Eastern US) and requires high humidity and regular contact with free water, but does

not forage within soil. Colonies are primarily composed of three castes: reproductives,

soldiers, and pseudergates. Reproductives include the king and queen, as well as

unmated winged forms (alates). Pseudergates, or false workers, are immature termites

that do not exhibit external signs of wings. Nymphs are immatures that exhibit wing

buds. Both pseudergates and nymphs are responsible for excavating and feeding on

wood to provide themselves and the colony with nourishment. The soldiers are

responsible for colony defense. Dampwood termite colonies may reach several

thousand individuals, and are characterized as one-piece feeders, staying within a

solitary food source for many years. In terms of pest status, this termite species is

considered a minor (or nuisance) pest, due mostly to its moisture requirements, habitat

preferences, and slow colony growth rates (Scheffrahn and Su 2007a). Figure 1-1B

shows the major castes of this species.

20

Termite Water Relations

All life on the planet requires water. The size of an organism and its habitat are

two vital components in determining to what degree water plays a role in an animal’s life

cycle and natural history. As an organism’s body size increases, its surface area to

volume ratio decreases. The converse is also true. As an organism’s body size

decreases, the surface area to volume ratio increases. This means that for smaller

organisms such as insects, the volume available to retain water resources within the

body in relation to the area available to lose such resources to evaporative mechanisms

is low and must be compensated for. This is a disadvantage in terms of water balance

(homeostasis) interactions for an organism and makes regulation of water loss difficult

in certain environments.

Termites are not exempted from this rule. In terms of insects, termites are

considered exceptionally prone to desiccation due in large part to their small size and

generally thin integument. The soft body of termites is due, at least in part, to their mode

of living. Termites live within nests and their adjoining foraging galleries are generally

found relatively close to or within a food source. Termites generally do not forage above

ground and exposed, and even then, not for an appreciable amount of time (Bignell and

Eggleton 2000). These nests can be found above and below ground depending on the

species. Such a lifestyle generally leads to fewer encounters with predators than more

exposed insects. Evidence exists that termites can create an environment with

antimicrobial properties that are effective when the colony is unstressed (Chouvenc et

al. 2013). Evolutionarily, a eusocial life cycle, a habitat (or microhabitat) of nests, and

behavior that tended to avoid or protect against major predators, is likely to have

21

enabled termites to establish individuals with softer bodies. Instead of developing

heavily sclerotized integuments in all individuals, as in ants (Order Hymenoptera, family

Formicidae), only a subset of individuals such as soldiers or alates that interact with the

outside world are endowed with sclerotized bodies or body parts. The tradeoff would be

less nitrogenous resources being allocated to developing hardened cuticles, as well as

ease of movement in tunnels and around nestmates with softer bodies. Colony behavior

(i.e. foraging and colony defense) is the likely reason different termite species have

different proportions of soldiers in an unstressed colony.

A more heavily layered cuticle may also help to keep or regulate the loss of water

held within an insect’s body. This is due to the generally impermeable and protective

nature of the insect cuticle (Wigglesworth 1945). The permeability of chitin (a major

component of the insect cuticle) to water makes the integument of insects the major

evaporative surface involved in water loss and its regulation (cuticular permeability). In

addition to the body surface, the oral, anal, and respiratory openings (spiracles) are also

areas important in water evaporation. Insects are known to have physiological

deterrents to water loss built into their cuticle: the cement and wax layers. The

properties of these layers are variable, however, and their roles in water regulation for

termites are largely unknown. Behaviorally, many termites address the problem of water

loss by modifying the microhabitat of the nest and adjoining foraging tunnels, creating

and regulating the microclimate within their nests and galleries (Collins 1991, Grube and

Rudolph 1999a, 1999b, Noirot 1970). The methods of this regulation are also not well

known for most termite species, but presumably involve location and movement of

water within the colony’s foraging area. This involves the detection of water resources

22

(by receptors housed in sensillae) that are then collected (labial glands) and relocated

to an area in need of increased humidity and/or moisture (Grube and Rudolph 1999a,

1999b). This also translates to the manipulation of food as seen in a study by Gallagher

and Jones (2010).

The term “water resource” is slightly misleading, however. The connotation with

this term implies that termites require free liquid water at all times. This is not an

accurate picture, however. Water in this case can refer to gaseous water (humidity),

moisture accumulation due to dew point depression, metabolic water obtained through

the breakdown of food, free liquid water (if encountered), as well as water bound to

various substrates (e.g. soil, food, cadavers). Termites are found in both arid and humid

habitats. There are desert dwelling species of termites that live deep underground,

termites that live and feed within extremely dry wood, and termites that live in other

xeric habitats where water resources may be difficult to procure. Termites also live in

mesic or hydric environments where moisture levels are moderate to high (humid).

Termite species (esp. subterranean termites) that live in damp or moist environments

must work to keep humidity levels even higher than normal. Even within a genus, the

water requirements of a given species of termites may differ. This begs the questions of

how and why some termite species (especially those more closely related) can have

such varied requirements with regard to water resources and how certain behaviors

address these requirements.

The goal of this project, then, was to examine the behavioral, physiological, and

morphological adaptations in water management among termite species found in

varying habitats. Indeed, water management concerns for termites (and other insects)

23

are paramount, especially since most species establish colonies within or near food

sources. A water source(s) must be relatively close, or the colony may have to relocate

or risk collapse. The first step in the process of utilizing water resources, then, is to

detect a humidity gradient, locate the source, and colonize such a suitable site.

Humidity-detecting receptors and their associated sensillae have been termed

hygroreceptors. They have been found, characterized, and more extensively studied in

the cockroach (Altner et al. 1977, Loftus 1976, Nishino et al. 2003, Roth and Willis

1952, Tichy and Kallina 2010, Tominaga and Yokohari 1982, Winston and Green 1967,

Yokohari 1978, 1981, Yokohari and Tateda 1976). Potential or putative hygroreceptors

have also been found in termites (Yanagawa et al. 2009, 2010). Given how closely

related cockroaches and termites are evolutionarily (Inward et al. 2007) and the

apparent conservation in insect hygroreceptors, it is likely that termites use similar

structures to find the moisture they require. In species that are bound by their galleries,

these structures would enable them to detect humidity gradients within their food

source. Substrate moisture is also known to affect termite distribution, foraging

behavior, and survival (Collins 1991, Cornelius and Osbrink 2010b, Gautam and

Henderson 2011, Kulis et al. 2008, McManamy et al. 2008, Steward 1982, Su 2003,

Woodrow et al. 2000). This supports the idea that termites have a way of determining

between more and less favorable conditions in terms of humidity after a colony has

become established, as well as the ability to modify behavior accordingly.

After detecting and finding a water resource, the termite must appropriately and

efficiently use it. Depending on their environment, a given species of termite may have

to deal with an excess or lack of moisture. Despite the implication that all (or most)

24

termites have to deal with a constant lack of water, it is important to note that there is a

large number of termite species that inhabit rainforests and other habitats with high

natural moisture levels. However, it has been suggested that species acclimated to drier

habitats (such as Cr. brevis) will die from overexposure to water, or “water poisoning”

(Collins 1969, Minnick et al. 1973, Steward 1982, 1983, Rudolph et al. 1990, Woodrow

et al. 2000). The effects of flooding and heavy rainfall are also a consideration, and not

only for termite species that prefer drier habitats (Forschler and Henderson 1995,

Cornelius and Osbrink 2010a, Owens et al. 2012). Figure 1-2 is a general diagram of

water management in the termite individual and termite aggregations, illustrating that

water is taken in, used, and lost in various processes, and representing the balance

between internal body water, external water sources, and desiccation conditions that

affect both. As individuals, termites must find water and ingest it, then use it for

physiological processes, excreting a portion as a component of waste. Additionally,

evaporation (through respiration and other movement through the cuticle) removes

some portion of the water taken in. This becomes a sort of metabolic algorithm. Food

consumption results in nutrients, energy, environmental water consumption, metabolic

water production, and waste, while consumed water (free or bound) results in energy

and waste, but can also be stored in glands. Ultimately, they must balance the use and

storage of the water they gain against the constant loss of that water from their bodies

through various processes.

An individual termite, then, uses (putative) hygroreceptors on its antennae to

detect humidity gradients in the environment, allowing them to avoid, modify, or recruit

to a given microhabitat. Their antennae (and associated receptors) detect changes or

25

differences in humidity (Yanagawa et al. 2009, 2010), but it is also possible termites

locate free water, bound water, and/or condensed moisture as a result of tunneling

behavior. The primary reason for foraging when water sources are lacking is the search

for water replacement and not food, with termites relying on stored nutritional reserves

until water is found. When a suitable water source is found, or when water is more

readily available, then food sources could become the primary goal of foraging.

After water is found, the termite consumes it for immediate use, for storage in

labial glands (Gautam and Henderson 2014), or use in metabolism. Some of this water

is then lost to evaporation through the cuticle from respiration and other processes. A

high percentage of this water is extracted (absorbed and resorbed) from the gut and

used in cellular homeostatic processes. As a result of habitat, termite species must

differentially retain water that is mixed with undigested material. The gut and rectal pads

accomplish this to varying degrees (depending on necessity) by retaining water from

consumed material and feces within the alimentary canal. In desiccation-tolerant

species, such as the Cryptotermes spp., most of the water is removed, leaving

characteristic pellets as waste. In other species, liquid feces is excreted and used in

construction and other behaviors (Grube and Rudolph 1999b). In this way, termites find

and use water resources to achieve an internal homeostasis based on the initial state of

their surroundings and their internal milieu.

When other individuals are included, additional behavioral considerations must

be taken into account. Interactions between individuals modulate behavior. Trophallaxis

(stomodeal and proctodeal), aggregation, and cannibalism under desiccative stress are

all additional factors to be considered. It should be noted that even though water

26

management is vital to the individual, a single individual has been observed being

sacrificed for the successful management of a group of termites (Collins 1991).

Environmental modulation is also important. Moisture is retained and managed in the

nest to keep humidity levels at or near an optimum through cuticular evaporation as well

as through transport of water resources within the nest (Noirot 1970, Gautam and

Henderson 2014).

As a colony, organization of the nest and adjoining foraging tunnels as well as

the age and size of the colony are components to be considered in the management of

water as a resource. The colony must address water management in order to achieve

(and keep) a general homeostasis. As the colony moves (e.g. forages, changes food

sources, or swarms) it must weather changes in environmental humidity and water

availability in order to keep an optimum environment in which to function most

efficiently. In some species, liquid feces are used to line the walls of the nest and

foraging tunnels, and to build shelter tubes or carton material in order to take advantage

of evaporation of water through the cuticle. Gradual drying and hardening of the outer

walls of shelter tubes, for example, creates a more easily regulated environment,

buffers against unfavorable external conditions, and provides a microhabitat with a

volume that lends itself to the maintenance of relative humidity, or RH (the ratio of the

partial pressure of water vapor to the equilibrium vapor pressure of water at the same

temperature), at a high level and near the optimum. Thus, termites use water resources

to achieve an optimal environment both internally as individuals and externally as a

colony based on the state of their ambient surroundings.

27

Objectives

Despite the research already conducted on desiccation tolerances (Collins 1963,

Collins 1966, Collins 1969, Collins 1991, Strickland 1950), effects of flooding (Forschler

and Henderson 1995, Cornelius and Osbrink 2010a, Owens et al. 2012), cuticular

studies (Collins 1963, Collins 1966), and other ecological and physiological work on

termites (Woodrow et al. 2000, Collins 1997, Noirot 1970, Su 2003), it is clear that

differences within and between genera regarding habitat preference and water

management has not been undertaken to an appreciable extent. This study provided

new insight into how termites manage their water resources both behaviorally and

physiologically, as well as bolster the information presently available. This information

may also provide insight into how and why over 360 species of termites emerged and

spread as pests of man-made structures, agriculture, forests, and pastures (Krishna et

al 2013). The overall hypothesis is that termites use a suite of behavioral and

physiological means to address paramount water management concerns based on the

changes in their environment both as individuals and as a group (colony) or risk the

death of enough individuals to incur colony collapse. Their ability to establish nest

homeostasis, as well as to tolerate changing conditions, especially in terms of RH and

water availability, is key to establishing themselves as pests or nuisance species and

expanding their ranges through natural and human-mediated means.

Chapter 2 reports on a histological approach to observing differences in cuticle

thickness, rectal pad width, and spiracle morphology. These difference may aid in

explaining how and why these four species are able to exploit their various habitats. In

Chapter 3 a study is presented that examined the survival of four termite species at

different relative humidities to elucidate differences in tolerance to desiccation. Cr.

28

brevis, whose endemic habitat is unusually stable and humid despite a lack of rainfall

(Scheffrahn et al. 2009), was observed acclimatizing and feeding efficiently at moderate

and high levels of humidity (Steward 1982). This was an interesting discovery and

indicates that Cr. brevis can tolerate a wider range of habitat and microhabitat

conditions in terms of RH and water availability than is implied in their common name

(West Indian drywood termite) and in the literature. In Chapter 4, observations are

presented on changes to relative humidity levels, resulting in RH equilibria, when 25

termites were placed in an empty chamber. Humidity preference using a multiple-choice

apparatus is examined in Chapter 5. The species-specific preference for a more

favorable environment was shown to be of importance therein. In Chapter 6, the ability

of termites to utilize different water sources was examined. Lastly, Chapter 7 examined

the ability of a group of termites to respond critically and efficiently to a drastic change in

humidity and water availability. The wetwood (nests in wood and soil high in moisture)

species couplet, C. formosanus and N. jouteli, and the drywood (nests in wood

generally low in moisture) species couplet, Cr. cavifrons and Cr. brevis, were compared

separately. The use of refugia and construction behavior were key components of their

responses.

This project provided information about the requirements for adequate and

efficient water management in termites not only at the individual level, but also at the

group level. The ecology and behavior of pest termite species are of interest for study

because of the intense physical damage and economic cost termites can affect. While

understanding how termites find and use different types of water resources is unlikely to

lead to methods for preventing them from achieving their required homeostasis (with

29

regard to humidity) within a colony and body water content within an individual, the level

of humidity-dependence and related behavioral plasticity for each pest species is key to

invasion prevention and may aid in bolstering current methods of control by highlighting

any sensitivities to RH levels and water availability. If the behaviors responsible for

achieving homeostasis and preventing desiccation could be disrupted, control methods

would be improved. It may also shed light on how natural selection and speciation

events can lead to symbioses. In the case of Cr. brevis, it should be noted that, with

isolated, rain protected outdoor populations found only in Hawaii, Honduras, Key West

(FL), and common outdoor populations found in coastal Peru and Chile (Scheffrahn et

al. 2000, 2009), this species has entered into a pseudo-commensal relationship with

humans (termites benefit, humans indirectly affected negatively), with expansion in

distribution of this species mirroring human globalization, but also, interestingly, not

including southeast Asia.

30

Figure 1-1. Termite species used for experimentation, with pseudergate (left), soldier (center) and winged imago (right) shown. A) Coptotermes formosanus Shiraki. B) Neotermes jouteli (Banks). C) Cryptotermes cavifrons Banks. D) Cryptotermes brevis (Walker). Photographs courtesy of author.

31

Figure 1-2. Diagram of a general termite water budget. Ellipses indicate ambient environment for each level: individual, group, and colony. A, B, and C are behaviors commonly associated with each level.

32

CHAPTER 2 HISTOLOGICAL APPROACH TO EXAMINING PHYSIOLOGICAL RESISTANCE TO

DESICCATION

Introduction

As important as water is to all life on Earth, different organisms employ a

diversity of behavioral and physiological methods, and have various morphologies for

obtaining and utilizing water resources. After finding water and getting it into its body, an

organism must retain it for use in biological processes. Water management for eusocial

insects begins with the individual. These individuals must use efficient means to prevent

and/or avoid desiccation. Small insects such as termites, have a more difficult time

preventing desiccation due to their small size. Body water is lost via evaporation

through the cuticle, buccal and anal openings, and spiracles. The cuticle of insects is

often heavily sclerotized, making it an initial line of physical defense to predators and

pathogens. This cuticle also includes microstructure for aid in the prevention of water

loss, namely, the cement and wax layers. The ability of termites to retain their body

water is of concern here. The soft cuticle of termites not only costs them in terms of

individual defense from predators and pathogens, but may also in terms of the balance

of the internal milieu (i.e. total body water content).

Another point of egress for water resources is through the spiracles. Respiration

through spiracles on the abdomen and thorax naturally includes water loss as vapor

along with gas exchange. During regular respiration when spiracles are open, water

vapor from the termite body is allowed to escape with gaseous CO2. The amount of

water lost in this way may be far less when compared to losses through the mouth,

anus, and cuticle, but deserves mention. It is possible that the drywood, dampwood,

and subterranean termites have spiracles with a similar structure, in addition to being

33

similar in function. It is also possible that they are completely different, and lend

themselves to the prevention of body water loss through some structural component.

Fecal deposition is yet another way water is lost from the body. This depends on

the ability of insects to retain or reabsorb water from their feces as it passes through the

gut. Defecation events are one of the termite’s biological processes that they are better

able to control water loss through. In drywood termites especially, the hard pelleted

frass is a reflection of rectal pads that remove much of the water from consumed wood.

These termites primarily acquire water resources metabolically from the dry wood they

live in and feed on because direct water resources are so infrequent in these habitats.

kalotermitid dampwoods, too, excrete pelleted frass that softens and clumps due to the

moist conditions these termites live in. Subterranean termites often use their more liquid

feces in construction of carton and repair of the nests and galleries. The differences in

frass is mainly due to differences in rectal pad physiology and, in fact, differences in the

rectum of termites have been recorded (Noirot and Noirot-Timothee 1977).

Taking into account the different habitats these termites are found in, are there

differences in the thickness of the cuticle, rectal pad morphology, and spiracle

morphology that could be responsible for the various ways termite species deal not only

with desiccative stresses, but also their normal environmental conditions? To address

this question, I prepared termite specimens of four species from varying natural

conditions for sectioning and staining, as well as general dissection. Staining was used

to evaluate thickness of the cuticle as well as the width of the rectal pads. Dissections

were undertaken to examine the spiracles. It was hypothesized that the drywood

termites would have thicker cuticles as well as larger rectal pads than the “wetwoods.”

34

The spiracles of the termites were observed through dissections without staining. It was

also hypothesized that the spiracles of each of the four species would be similar in

structure.

Materials and Methods

Individuals from colonies of C. formosanus, N. jouteli, Cr. cavifrons, and Cr.

brevis were collected in Broward County, Florida. These species were selected because

they inhabit different habitats and, presumably, different microhabitats in terms of

available water and humidity level. Different colonies of C. formosanus were collected

from bucket traps as described in Su and Scheffrahn (1986). Colonies of N. jouteli, Cr.

cavifrons, and Cr. brevis were assumed to be different when collected from different

pieces of wood even if collected from the same location since multiple colonies could be

inhabiting a single piece of wood. The wood was split to expose individuals that were

then aspirated and placed into containers. Groups of C. formosanus collected from

bucket traps were transferred to polystyrene containers that contained a wood food

source and were regularly misted with water to keep the relative humidity in the

containers high (>95%). These containers were then placed in an incubator with an

average temperature of 26.4°C and an average RH of 41.5%. Cryptotermes termites

were kept in polystyrene boxes and stored in the incubator as well. Stacked wood

shelters were provided as food and refuge for these species. Cr. cavifrons was also

provided with small water dishes made from shell vial caps cut to provide access to a

small reservoir of liquid water. Polystyrene boxes housed N. jouteli and wood supplied

to provide food and shelter, and were regularly misted with water to keep the humidity at

>95% RH. These boxes were also kept in said incubator. Populations of each termite

35

species survived well in these conditions and were used for experiments as needed.

Termites were kept in these boxes in the incubator for no more than 6 months before

use.

Individual worker or pseudergate (for the kalotermitid species) termites were

randomly sampled from these populations for dissection and sectioning. The

characteristics of the cuticle, rectal pads, and spiracles of the four termite species were

examined in relation to their micro-habitat preferences and desiccation tolerances.

Termite specimens were prepared for histological sectioning by first having the heads

and legs removed, and were then placed in Bouin’s fixative solution (75% aqueous

picric acid, 20% formaldehyde, 5% acetic acid) for at least seven days (Martoja and

Martoja-Pierson 1967). They were then dehydrated for 30 min three times in successive

solutions of 75% and 95% ethanol, a single time in pure n-butanol for 30 min, and for 24

hrs three additional times in containers of pure n-butanol.

Following dehydration, specimens were passed through a single mixture of 50%

paraffin wax and 50% n-butanol for 6 hrs at 60°C, and three times in containers of pure

paraffin wax for 24 hrs at 60°C in order to replace the butanol with paraffin. Specimens

were then embedded in blocks of pure paraffin wax, sectioned with a microtome (~7µm

sections), attached to slides, and stained with a modified Azan Heidenhaim protocol

(Mayer et al. 1979, Chouvenc et al. 2009). The stained slides (with cover slips and

Permount mounting medium) were observed at varying magnification with a compound

microscope (Leica DM500 B, Leica Microsystems GmbH, Wetzlar, Germany) coupled

with a DFC450 Leica camera, as well as a dissecting microscope (Olympus SZX9,

36

Olympus Corporation, Shinjuku, Tokyo, Japan) in order to measure the thickness of the

cuticle and rectal pads, and observe the morphology of the spiracles.

Pictures of the stained abdominal section slides were taken with Leica

Application Suite (LAS) software and stored for measurement analysis with GIMP (The

GIMP team, GIMP 2.8.14, www.gimp.org, © 1997-2014). Measurement of the width of

the cuticle was taken where staining had allowed for determination of the cuticle

thickness (i.e. cuticle was colored pink or purple). Attempts were made to take

measurements as close to perpendicular to the plane of the cuticle as possible. Analysis

of the rectal pads was made using the width of the pads observed through transverse

sections of the abdomen. Observation of the spiracles was made using the compound

and light microscopes and analysis was solely observational and based on morphology.

All statistical analysis was carried out using JMP statistical software (JMP® Pro,

2013. version 11.0. SAS Institute Inc., Cary, NC). An ANOVA was used to examine

differences in overall cuticle thickness and rectal pad width. Tukey’s HSD post hoc tests

were used to separate differences between all pairs. Species was the factor and cuticle

thickness or rectal pad width (microns) were the response variables. Morphology of

spiracles was purely observational and did not include statistical analysis.

Results

The ANOVA test indicated there was a significant difference in cuticle thickness

(df=3, F=55, P-value<.0001) and rectal pad width (df=3, F=48.4, P-value<.0001) among

species. When separating differences using Tukey’s HSD, all four species were found

to be significantly different, with N. jouteli having the thickest cuticle and C. formosanus

having the thinnest (Table 2-1). Rectal pad width was not significantly different between

N. jouteli and Cr. brevis (Table 2-2), but they were significantly thicker than Cr. cavifrons

http://www.gimp.org/

37

and C. formosanus, with C. formosanus having the thinnest rectal pads (Table 2-2).

Examples of the stained cuticle and rectal pads can be seen in Figure 2-1 and 2-2,

respectively. Figure 2-3 is an example of the abdominal spiracles after dissection of a

specimen of N. jouteli. N. jouteli was used as an example because it was easy to locate

the spiracles of this species and more difficult to locate and photograph the spiracles of

the smaller remaining three species. All four species had spiracles on each of the 8

tergites of the abdomen, situated near the pleural membrane. In Cr. cavifrons, Cr.

brevis, and N. jouteli, the spiracles take on a “j” shape with an extension protruding from

the main channel of the spiracle. In C. formosanus the spiracle lacked this

protuberance, but still had a “j” shape. Examples of the spiracles of the four termite

species observed through dissections and slide squashes can be seen in Figure 2-4.

Stained and sectioned specimens, in conjunction with unstained dissection slides

revealed the probable morphology of the spiracular structures. In C. formosanus the

trachea attached to the underside of the main structural curve (beginning of the tail of

the “j”), with no additional arm present (Figure 2-5). In the kalotermitids, however, the

trachea appeared to connect beneath the structure, in between the additional atrial arm

and the main structural curve (beginning of the tail of the “j”) (Figure 2-6). The structure

of C. formosanus appeared to be longer and thinner than the structures found in the

kalotermitids.

Discussion

The cuticle (and sclerotization) of insects is an important component of defense

from external factors, as well as stability in regards to the internal and external milieu.

The permeability of the cuticle to water in termites is key to this stability and resistance

to desiccation. A general lack of sclerotization in termites might seem to compound their

38

difficulty with losing water. The thickness of the cuticle was of major concern in this

study, with the assumption being that a thicker cuticle contributes to a lower percentage

of body water lost to evaporation over a given amount of time (lower CP). This

assumption is not supported when looking at previous work on cuticular permeability in

beetles and cockroaches (Appel et al. 1983, 1986, Hadley 1977, 1978, Monzer and

Srour 2009, Weissling and Giblin-Davis 1993). Even more heavily sclerotized insects

can lose water easily as reflected in high CP values. Locke (1965), Beament (1945) and

Wigglesworth (1945) also found that treatment of the insect cuticle with wax layer

disruptants (i.e. - peanut oil, abrasive dusts) caused the rapid loss of body water.

Similar results were seen in studies on the cuticle of termites by Collins (1969) and

Sponsler and Appel (1990). It seems then, that the thickness of the cuticle is not as

important as its composition (layering and hydrocarbons), with a lipid layer being the

main barrier to water loss (Beament 1961) and disruption of these cuticular layers

leading to an increase in rates of water loss, as was seen when drywood termites were

treated with sorptive dusts which lead to rapid desiccation (Ebeling and Wagner 1959).

Representative photographs of the cuticles of the four termite species are seen in

Figure 2-1. The fact that C. formosanus had the thinnest cuticle makes sense when we

recall that individuals and small groups of this species did not last long in dry conditions.

The cuticle did not allow for efficient trapping of body water within the termite. C.

formosanus termites were unable to prevent water from evaporating from the surface of

the cuticle to an appreciable degree, and, thus, they dried out and died. This is a

possible reason why these termites rely so heavily on behavioral mechanisms and

modification of their microhabitat with body water to prevent desiccation. Cr. brevis and

39

N. jouteli had similarly greater cuticle thicknesses than C. formosanus and Cr. cavifrons.

Since Cr. brevis lives in such dry habitats, this makes sense. It is easier for water to be

held within the body in a desiccating environment with a thick cuticle that includes a wax

layer barrier to water loss (Beament 1961). N. jouteli, on the other hand, lives in an

environment of moist wood, so a thicker cuticle does not seem necessary. These

termites are single-piece feeders, however, living within their food source. If this food

and shelter source is disturbed, moisture may be lost and, possibly, lost quickly. Having

a cuticle that aids in tolerating such less-than-favorable conditions helps give termites

time to find shelter. Their large size and an ability to tolerate losing much of their body

water are additional factors aiding in this desiccation prevention. Cr. cavifrons is

intermediate to the previous species, having a cuticle thicker than C. formosanus, but

thinner than Cr. brevis and N. jouteli. When we think of their natural habitat and the

results of the previous experiments in this study, we have a possible reason why this

would be observed. Cr. cavifrons can be found living in wood with both low and high

moisture content, and in a natural habitat unlike the ones the other species inhabit.

Inherent in this habitat are fluctuations in humidity and moisture availability, as well as

possible disturbances to their single-piece food source. They would, therefore, not only

be equipped to tolerate the changes in the conditions of their environment, but also be

equipped to adequately prevent water from escaping the body in normal and stressed

situations.

In addition to the thickness of the cuticle, the rectal pads were also examined and

compared. Representative photographs of the rectal pads of the four termite species

are seen in Figure 2-2. The results indicated that N. jouteli and Cr. brevis had

40

statistically larger rectal pad widths than did C. formosanus and Cr. cavifrons. The rectal

pads of these species were generally shorter and thicker, often taking a swollen, more

bulbous shape. This probably relates to the fact that these two species produce pelleted

frass. In the habitat of N. jouteli, however, these pellets can be found loose or clumped

depending on the wood moisture content (Scheffrahn and Su 2007a). The rectal pads of

these species act as clamp-like sponges, absorbing water from waste to be returned to

other parts of the body and producing six-sided fecal pellets. In C. formosanus, the

rectal pads also function in retaining water as needed, resulting in more liquid fecal

matter that can be used in nest modification and construction of carton material with

antibiotic properties (Chouvenc et al. 2013). This proved to be a more difficult

measurement process to undertake, and it is prudent to note that the difficulty in taking

these measurements may have had an undesired effect on the overall trend as

measurement error.

Lastly, the morphology and possible functional mechanisms of the spiracles were

also examined. These structures are involved in respiration and link the outside

environment with the internal milieu through sclerotized holes in the cuticle. They are

another orifice (besides the mouth and anal cavities) through which water (vapor) can

escape the body. The function of the protuberance (atrial arm) that was observed in N.

jouteli and the Cryptotermes species is unclear, though it may function as a point for

muscle attachment, be an extension of the atrium, a separate air sac-like structure

aiding in gas exchange, or a moisture/humidity trap aiding in the prevention of

evaporative water loss (Snodgrass 1993). Muscle attachments were observed with the

slides, but exactly how these muscles were connected and their role in closing the

41

spiracles was unclear. The closing apparatuses were clearly of the internal variety in all

four species and the atria of the spiracles of insects have been noted to be long and

tubular in some cases (Snodgrass 1993). The spiracle structure observed was clearly

connected to and separate from the trachea as evidenced by the lack of taenidia within

this structure and the clear taenidial rings seen in the tracheal trunk. Further study on

the ultrastructure of the spiracle of the termite is warranted, but it is clear a structural

difference exists.

42

Table 2-1. Cuticle thicknesses (mean±SEM) of four termite species

Species Cuticle Thickness

(microns)a Observations

N. jouteli 2.51±0.05a 339

Cr. brevis 2.28±0.03b 180

Cr. cavifrons 2.10±0.03c 222

C. formosanus 1.81±0.03d 196 a Values followed by the same letters are not significantly different at the ɑ=0.05 (Tukey's HSD).

43

Table 2-2. Rectal pad widths (mean±SEM) of four termite species

Species Rectal Pad Widths

(microns)a Observations

N. jouteli 171.6±4.8a 55

Cr. brevis 160.5±4.2a 66

Cr. cavifrons 142.6±3.7b 59

C. formosanus 101.8±4.8c 62 a Values followed by the same letters are not significantly different at the ɑ=0.05 (Tukey's HSD).

44

Figure 2-1. Transverse sections of stained abdominal cuticle at 200x magnification. A) C. formosanus, B) N. jouteli, C) Cr. cavifrons and D) Cr. brevis. Arrows indicate cuticle. Photographs courtesy of author.

45

Figure 2-2. Transverse sections of stained rectal pads at 200x magnification. A) C. formosanus, B) N. jouteli, C) Cr. cavifrons and D) Cr. brevis. Arrows indicate individual rectal pads. Photographs courtesy of author.

46

Figure 2-3. Abdominal spiracles on the tergites and near the pleural membrane of N. jouteli. Abdominal spiracles of C. formosanus, Cr. cavifrons, and Cr. brevis were found in similar locations on the body. Photograph courtesy of author.

47

Figure 2-4. Dissected spiracles at various magnifications. A) C. formosanus, B) N. jouteli, C) Cr. cavifrons and D) Cr. brevis. Arrows indicate spiracular protuberances (atrial arm). Photographs courtesy of author.

48

Figure 2-5. Comparison of artist interpretation and photograph of the attachment of the

spiracle structure and the trachea of C. formosanus. A) Drawing of spiracle-tracheal attachment. B) Photograph illustrating spiracle-tracheal attachment. a: spiracle cap (peritreme) b: atrium c: atrial arm of spiracle structure d: tracheal trunk e: cuticle. Drawing and photograph courtesy of author.

49

Figure 2-6. Comparison of artist interpretation and photograph of the attachment of the spiracle structure and the trachea in the kalotermitid species used (Cr. brevis, Cr. cavifrons, and N. jouteli). A) Drawing of spiracle-tracheal attachment. B, C, and D) Photographs illustrating spiracle-tracheal attachment. a: spiracle cap (peritreme) b: atrium c: atrial arm of spiracle structure d: additional atrial arm of spiracle structure e: tracheal trunk f: cuticle. Drawing and photographs courtesy of author.

50

CHAPTER 3 WATER LOSS TOLERANCE OF FOUR TERMITE SPECIES EXPOSED TO VARIOUS

RELATIVE HUMIDITIES

Introduction

Water is of paramount importance to the overall health and proper function of all

organisms. There are few animals that can survive extended periods of time without

water in some form or another being available to them. Examples of organisms that

undergo anhydrobiosis include the tardigrades, nematodes, and the sleeping

chironomids (Polypedilum vanderplanki) (Hinton 1960). Termites, however, are not able

to employ this state and must locate and utilize water resources to prevent and/or

tolerate desiccation. They live in various habitats and microhabitats with differing levels

of ambient humidity and availability of water. Just as with other organisms, within their

natural environments, termites also have optimal conditions that they employ behavioral

and physiological means to achieve or find. This begs the question as to what termites

do when their optimal conditions fail to be maintained or cannot be found. Specifically,

how well do termites survive when placed in conditions where relative humidity varies

from very low to very high? In order to address this question, four species of termite that

are naturally found in various (micro) habitats were used. Humidity chambers were

constructed to observe termite survival in conditions presumably closer to and further

from their natural conditions in terms of relative humidity. It was hypothesized that the

Cr. brevis, a termite found in wood generally devoid of moisture (characterized as a

drywood termite), would survive better in lower RHs and worse in higher RHs. The

converse was hypothesized for Cr. cavifrons (a “drywood” termite that can also be found

in moist wood), as well as the dampwood termite, N. jouteli, and the subterranean

termite, C. formosanus (characterized as wetwoods termites).

51


Termites used for this study were collected and stored as described in Chapter 2.

Desiccative humidity chambers consisted of clear polystyrene containers (Tri-State

Plastic, Inc., Kentucky, USA) (17.15x12.22x6.03 cm) and fitted lid with internal relative

humidity conditions stabilized using three saturated salt solutions, silica gel, and water

(Figure 3-1). These three solutions, silica gel, and water were chosen to create the

humidity conditions in each chamber after referencing Rockland (1960) and Winston

and Bates (1960). The container lip was coated with Eboline petroleum jelly (Eboneen

Products Company, Inc., Hartford, CT, USA) to create a seal between the lid and

bottom. Temperature and humidity levels were measured daily using an Amprobe

THW3 probe (Amprobe®, Everett, WA, USA) fitted with a plastic collar to allow for

upright free-standing while taking readings. Access to the chamber for the probe was

provided by an approximately 1.59 cm hole in the lid. These holes were plugged with

rubber stoppers while not in use. Probe readings were used to determine the average

RH of each chamber. The average RHs (±SEM) of the chambers and their associated

stabilizing material were as follows: 92.0±0.07% (H2O), 72.9±0.08% (NaCl), 55.7±0.09%

(Mg(NO3)2), 34.3±0.04% (MgCl2), and 18.2±0.14% (silica). Groups of ten termite

individuals per species were introduced into each chamber. Termites were placed in

small Fisherbrand® disposable Petri dish bottoms (35x10 mm). Holes were cut into the

lids of these Petri dishes to allow air movement and prevent termite escape. Dishes

were also provisioned with a small piece of wood (Pinus sp.) as a food source (15x15x9

mm). After the termites were introduced to their respective dishes, the container lid was

replaced. The three salt solutions were kept in large Pyrex glass Petri dish bottoms

(93x22 mm) in the center of their respective containers. The silica gel covered the

52

bottom of the container in an approximately 2.54 cm layer with the dishes holding the

termites partially sunk into the silica particles. Silica gel was not placed in a dish

because preliminary tests indicated that silica gel in a Petri dish did not provide a

surface area as effective at stabilizing the RH of the chamber as a layer along the

bottom of the container did. The water chamber held a glass Petri dish bottom

containing deionized water and Whatman #1 qualitative circle filter papers (55 mm

diameter) that had been cut in half (7 semicircles), wetted, and attached to the side of

the dish (partially submerged) to facilitate greater evaporation of water through

increased surface area. Temperature and humidity readings were recorded every 24

hours post-introduction for twelve days (preliminary experiments suggested Cr. brevis

would die in the high humidity chamber at around 10 days). Daily counts of living

termites were recorded for a 12 day period. A follow-up experiment was also conducted

to examine recovery of N. jouteli individuals that exhibited visible reduction in mass from

loss of body water. At the conclusion of the 12 day study, the surviving N. jouteli

individuals from the various RHs were weighed and then placed in a Petri dish that

provided food and water for 1 week. Following this week, surviving termites were

counted and weighed again to examine recovery with adequate food and water

resources.

Statistical analysis was carried out using JMP statistical software. An analysis of

variance (ANOVA) for a 4x5 factorial experimental design was conducted to test for a

difference. Species and relative humidity (RH) were the factors and percent survival the

response variable. Percent survival values were arcsine-square root transformed before

53

analysis. Tukey’s HSD post hoc test was used to evaluate all pairwise differences at

ɑ=0.05. Data from six replications were analyzed.

Results

After 12 days, statistical analysis provided evidence for significant differences in

species and RH, as well as the interaction between the two (Table 3-1). In this case,

only C. formosanus exhibited a significant difference in survival among RH levels. They

survived significantly better in the 92% RH water chamber and died within a few days

when housed in RH conditions ≤72.9% (Table 3-2). There was no significant difference

in survival at any RH level (18.2 to 92%) for Cr. cavifrons, Cr. brevis, or N. jouteli (Table

3-2). However, the appearance of N. jouteli individuals in all of the humidity chambers

except the water chamber were visibly smaller (desiccated) at the conclusion of the

experiment than at the beginning. Whereas C. formosanus individuals died in all

chambers save the H2O chamber, N. jouteli individuals were visibly reduced in body

mass at RHs lower than 92%. The Cryptotermes species did not exhibit this change. At

all RHs, other than two of six instances in the 92% water chamber, the N. jouteli groups

decreased in mass (ranging from 4.4 to 151.2 mg loss per group of 10). Of the 180 N.

jouteli individuals that survived in the various RH levels and were then placed in the

food and water dish, 172 recovered. These 172 individuals exhibited a total weight gain

of approximately 327 mg. The remaining eight individuals were not found, indicating

cannibalism was a component of recovering from desiccation. The N. jouteli groups of

ten individuals exhibited a decrease in mass of approximately 42% of their weight as a

loss of body water to evaporation due to desiccation.

54

Discussion

The results of this study show that C. formosanus is far less capable of tolerating

desiccation than N. jouteli, Cr. cavifrons, and Cr. brevis. This is in agreement with

previous studies by Collins (1958, 1963, 1966, 1969) and Khan (1980) on kalotermitid

and rhinotermitid species, and reflected in the quick desiccation and death of the

subterranean termites. The overall mean survival of termites in the RHs and overall

mean survival of the species were affected by the survival of C. formosanus. Due to the

100% mortality of C. formosanus in RH ≤72.9%, results were skewed toward

significantly higher overall termite survival at 92% humidity and significantly lower

overall survival for this species. Termite survival means for the remaining three species

and in the remaining four RH levels did not express such an overall difference. The fact

that N. jouteli did not express a difference in survival in the various chambers and

individuals were observed to lose body mass and remain alive, is likely due, in part, to

their body size and thus, the amount of body water available to be lost. However,

Strickland (1950), noted that desiccation tolerance was not correlated to size, indicating

that differences in the ratio of surface area to volume are not the only factors in species

differences in survival under such experimental conditions. The ability to create and

utilize metabolic water from food sources is also a likely factor (i.e. consumption of drier

food sources versus more moist sources). Nakayama et al. (2004c), in a study on

Reticulitermes speratus (Kolbe) and C. formosanus, found evidence suggesting that in

termites, there are thresholds for recovery from desiccation treatments that change

depending on humidity level and period of desiccation. There is also evidence for this

with the survival and condition of the Neotermes species in this study. Termites can rely

on fat body reserves, breaking them down to utilize bound water in order to delay death

55

from desiccation. When more favorable conditions return, they are generally able to

recover, as was seen with the recovery of the desiccated Neotermes individuals. This

recovery has also been noted in another kalotermitid, I. minor by Pence (1956). In the

case of C. formosanus, they did not have enough body reserves to rely on for the time

period of the study, exhausted their body reserves rapidly, or desiccated so quickly that

reserves were generally not a factor in tolerating desiccation.

The Cryptotermes species also did not express a difference in survival in any of

the chambers. As was noted by Collins (1958) and Pence (1956), the Cryptotermes

species in this study were also far less active than the subterranean and dampwood

species, and tended to huddle with little movement unless disturbed. This was likely a

behavioral mechanism to prevent desiccation from water loss through evaporation from

the body by decreasing overall surface area. Collins (1991) discussed the demand for

higher environmental moisture in some kalotermitids and termopsids (family

Archotermopsidae), something seen with Neotermes. She also noted that the inability to

survive in such high moisture environments is not always seen in Cr. brevis, with some

populations able to adjust to such conditions. The time period of this study likely failed

to show species differences in desiccation tolerances among three kalotermitids that

would be evident with extended periods of time. Despite this, both Cryptotermes

species are probably capable of tolerating a wide range of humidity conditions (at least

for a time). Since C. formosanus lives and forages in soil, they are less likely to

encounter severely dry conditions. N. joueli and Cr. cavifrons, however, generally live

and feed in pieces of wood that are more likely to be subject to RH changes. Given that

Florida can have months-long rainy seasons and extended dry seasons, it would

56

certainly be beneficial for them to be able to tolerate a range of RH and water

availabilities. Cr. brevis, however, is not found in wood exposed to free water, and

probably has a tolerance range that lies exclusively within lower RHs and a general lack

of water availability. Future studies should extend the length of time similar experiments

are run. Additionally, this was a no-choice test, and each species will employ whatever

means necessary to stay alive as individuals and as a group. Cannibalism, another

behavioral mechanism, was seen by Collins (1991), but was not explicitly observed in

the present study. It was evident, however, with the recovery of Neotermes individuals

from desiccation in which several bodies were missing. In this study, the rapid deaths of

C. formosanus individuals indicates that this species was unable to utilize certain

behaviors at all or efficiently enough to reduce the loss of body water and prevent

desiccation within the time frame of the experiment and probably only delayed their

inevitable death from desiccation. Their physiology was also clearly unable to aid in

coping with the conditions of the chambers with RHs below 72.9%, as they were

observed to lose much of the water from their bodies and die of desiccation. As there

were not clear changes in behavior in N. jouteli, and the Cryptotermes species, their

physiologies must have been able to help them survive and flourish in the various RH

conditions during the time period of the study. What, exactly, these physiological

mechanisms were, remain to be elucidated, but they are surely related to cuticular

permeability and percentage of total body water, which will be examined in the next

chapter.

57

Table 3-1. Analysis of variance for a factorial experiment to evaluate the effects of humidity on worker survival of four termite species over 12 days

Source df Termite survivala

MS b F P-value

Species (S) 3 10 406.1 <.0001

Chamber (C) 4 0.82 33.38 <.0001

SxC 12 0.74 29.91 <.0001

Error 100 0.02 --- ---

a Each combination of treatments was replicated 6 times.

b Mean square.

58

Table 3-2. Effects of five RH levels on worker survival (%) of four termite species over 12 days (mean±SEM)

RH levelabc

(92.0±0.07%) (72.9±0.08%) (55.7±0.09%) (34.3±0.04%) (18.2±0.17%)

[23.5±0.06°C] [23.6±0.06°C] [23.5±0.06°C] [23.7±0.06°C] [23.6±0.06°C]

Species

Overall mean survival for speciesd

C. formosanus 100±0.0Aa 0.0±0.0Ab 0.0±0.0Ab 0.0±0.0Ab 0.0±0.0Ab 20.0±7.4A

N. jouteli 98.3±1.7Aa 100±0.0Ba 96.7±2.1Ba 95.0±3.4Ba 98.3±1.7Ba 97.7±0.9B

Cr. cavifrons 98.3±1.7Aa 100±0.0Ba 96.7±2.1Ba 98.3±1.7Ba 90.0±6.3Ba 96.7±1.5B

Cr. brevis 95.0±2.2Aa 93.3±4.9Ba 96.7±5.4Ba 96.7±2.1Ba 91.7±4.0Ba 93.7±1.7B Overall mean survival at RH leveld

97.9±0.8a 73.3±8.9b 71.3±8.7b 72.5±8.8b 70.0±8.6b

a Means for the combination of treatment (species x RH level) followed by the same lowercase letters within a row or means followed by the same capital letter within a column are not significantly different at the ɑ=0.05 level (Tukey's HSD).

b Value for the combination (species x RH level) of treatments are means of 6 observations.

c Mean RH±SEM produced by water, three salts, and silica gel found in parentheses (mean temperature±SEM in brackets).

d Values for each main treatment effect (species, RH level) are means of 30 and 24 observations, respectively. Means followed by the same letter are not significantly different at the ɑ=0.05 level (Tukey's HSD).

59

Figure 3-1. Setup for examining termite survival when exposed to various relative humidities. A) Humidity chambers with lids removed. a: H2O dish with filter paper ring (92.0±0.07%) b: NaCl dish (72.9±0.08%) c: Mg(NO3)2 dish (55.7±0.09%) d: MgCl2 dish (34.3±0.04%) e: silica gel layer (18.2±0.14%) f: wood food source g: holding dish with modified lid. B) Humidity chambers with lids in place h: rubber stopper i: temperature/humidity probe j: chamber lid. Cardboard box cover not shown. Photographs courtesy of author.

60

CHAPTER 4 CUTICULAR PERMEABILITY, BODY WATER LOSS, AND RELATIVE HUMIDITY

EQUILIBRIA OF FOUR TERMITE SPECIES

Introduction

In order to keep a colony healthy, individuals (workers) must find water and use it

to stabilize the nest and foraging tunnels they use as their home at or near homeostasis

(the maintenance of relatively stable internal conditions) in terms of water availability

and humidity. The ability of termites to meet this equilibrium (stable internal conditions

despite external condition changes) involves behavioral and physiological processes.

One such way involves the direct movement and use of body water from individual

termites to modify the microhabitat of the nest (e.g. - with feces), a tunnel, or even a

food source (Gallagher and Jones 2010, Nakayama 2004a, 2004b, Grube and Rudolph

1999a, 1999b, Grube et al. 1997). Another way is the indirect (passive) use of body

water, or cuticular permeability (CP) and the associated evaporation of water through

the cuticle, mouthparts, anus, and spiracles. If this body water (as vapor) is contained

within the nest and galleries, the RH of these areas will reach an equilibrium at which

the termites will not (or very slowly) lose body water. Cuticular permeability is the

amount of water lost (μg) per unit surface area (cm2) per unit time (hr) per unit

saturation deficit (mmHg), which is often used to describe and compare evaporative

water loss from the cuticle of insects and other arthropods (Wigglesworth 1945, Edney

1977). Previous work on water loss rates, percent total body water content (%TBW) and

cuticular permeability (CP) in termites was undertaken by Cook and Scott (1932),

Collins (1958, 1963, 1966, 1969), Edney (1977), Rust et al (1979), Sponsler and Appel

(1990), Shelton and Appel (2000, 2001), Shelton and Grace (2003), and in cockroaches

by Appel et al. (1983, 1986). These studies indicated that the use of Meeh’s formula,

http://www.sciencedirect.com/science/article/pii/S1095643302002702#BIB32



61

2/3 power of body weight multiplied by 12 (Edney 1957, Edney and McFarlane 1974,

Mead-Briggs 1956), for determining surface area for use in calculating CP can be

problematic and, thus, results can be somewhat inconsistent. However, in general,

these previous studies indicated that when placed in desiccative conditions,

rhinotermitids and those kalotermitids associated with more moist habitats had higher

%TBW, higher rate losses, higher CP values, and shorter survival times than did the

kalotermitids associated with drier habitats. Rate loss was shown to decrease over time,

and upon death, those termites associated with arid habitats had lost a lower

percentage of body mass. Given this information, it is clear that behavioral plasticity and

modification of the environment are important components of resistance to desiccation.

Water lost through the cuticle, mouthparts, anus, and spiracles should be reused, if

possible, before being lost entirely. Termites live in various and constantly changing

habitats, and in generally stable microhabitats, with various levels of ambient humidity

and available water. The question is whether, in a confined environment, the water from

a termite’s body will modify a less favorable environment towards the RH equilibrium of

a more favorable one. To this end, four termite species from different

habitats/microhabitats were used to observe whether a low relative humidity

environment would be significantly changed in terms of RH through the loss of water

resources from the bodies of both living and dead termites. CP values for the four

termite species were measured to compare with the confined space RH equilibria

determined through this study. It was hypothesized that living groups of all four termite

species would significantly change the humidity of the environment in the enclosed

experimental chamber and that there would be significant differences among the four

62

species. Additionally, dead groups of all four termite species were hypothesized to raise

the humidity of the environment in the enclosed experimental chamber to 100% RH.


Cuticular Permeability and Body Water Loss

Termites used for this study were collected and stored as described in Chapter 2.

Groups of dead termites, as well as groups of live termites, were used to evaluate if

there was a difference between physical water loss and physiologically mediated water

loss from their bodies. Dead termites were obtained by killing termites with ethyl acetate

(fumes) and immediately using then for experimentation. Rate of weight loss as water

vapor was calculated for 10 groups of 10 live and 10 groups of 10 dead individuals for

each species over 12 hours. The initial biomass of these groups of living and dead

termites was recorded. The groups were then placed in a desiccator of indicating

Drierite in a polystyrene box that was kept in an incubator (≈41.5% RH and 26.4°C).

Temperature and humidity levels were measured using an Amprobe THW3 probe fitted

with a plastic collar to allow for upright free-standing during readings. Readings were

taken to determine the range of RHs and temperatures over the 12 hour study. Access

to the chamber for the temperature and humidity probe was allowed through an

approximately 1.59 cm hole in the clear box lid. The drying weight of the termite groups

was taken every 2 hours for 12 hours (six readings). The percentage total body water

content (%TBW) was calculated by taking the difference between the initial (wet) mass

and the final (dry) mass, dividing by the wet mass and multiplying by 100. This loss in

mass was assumed to be due to loss of water from the body. In addition, the rate of

%TBW loss was found by subtracting the wet mass from each hourly mass and dividing

by the wet to dry mass difference and multiplying by 100. Cuticular permeability was

63

calculated similarly to Sponsler and Appel (1990) by using mass loss after 12 hours.

Surface area was estimated using Meeh’s formula and the saturation deficit calculated

as the difference in the vapor pressure of water at a certain RH and temperature and

the vapor pressure of saturated air at the same temperature (Edney 1977).

RH Equilbria

Equilibria chambers were constructed using wide-mouthed clear plastic jars

(Uline, Wisconsin, USA). The jars were 53 mm in height and 52 mm in diameter (without

lid). The bottoms of the jars were scratched with sandpaper to facilitate termite

movement by providing friction for their tarsal claws. Temperature and humidity levels

were measured using the probe mentioned in the rate loss study. Access to the

chamber for the temperature and humidity probe was allowed through an approximately

1.59 cm hole in the clear plastic lid. Groups of 25 live termites and groups of 25 dead

termites were introduced into their respective chambers. After termites were placed into

the jars, the chamber was sealed using Eboline petroleum jelly to limit air movement

into and out of the chamber. A small amount of petroleum jelly was also used to create

a seal between the probe collar and the lid. Cardboard or plastic cups with the bottoms

cut off were placed (inverted) over the chambers. Two layers of black satin cloth were

then draped around the humidity probe and the cup to provide a dark environment for

the termites (Figure 4-1). Temperature and humidity readings were recorded every 4

hours for approximately two days to determine the relative humidity level that was

achieved by the termite groups (as plateaus) presumably through cuticular and

respirative evaporation of water.

Statistical analysis was conducted using the JMP statistical software and

SigmaplotTM (v12.5, Systat Software, Inc., San Jose, CA). A 2x4 factorial analysis was

64

conducted with species and status (alive or dead) as the factors and initial mass,

%TBW, or CP value as the response variable. Only the percent values for TBW were

arcsine-square root transformed before the analysis of variance (ANOVA) was used to

test whether there was a significant difference among initial mass, %TBW, or CP value

for the four species. Post hoc Student’s t tests were used to separate the differences for

each pair at ɑ=0.05. Data from ten replications were analyzed.

Regression analyses were also used to evaluate rates of %TBW loss. A Z-test

where 𝑍 = |𝐵1 − 𝐵2|/√𝑣𝑎𝑟(𝐵1) + 𝑣𝑎𝑟(𝐵2) was conducted on the slopes of the

regression equations Y = A + B*t, to examine differences in the rates of %TBW loss

between all species pairs at ɑ=0.05. Because a curve fitting equation that was

applicable to the data of all four termite species could not be determined, approximation

of equilibrium using the average %RH levels was used (Figure 4-2 and 4-5). The 16th

hour after introduction was used as the approximate onset of equilibria. A 2x4 factorial

analysis was conducted with species and status (alive or dead) as the factors and RH-

equilibrium at 16 hours as the response variable. The RH percent values were arcsine-

square root transformed before the analysis of variance (ANOVA) was used to test

whether there was a significant difference among RHs for the four species. Post hoc

Student’s t tests were used to separate the differences for each pair at ɑ=0.05. Data

from four replications were analyzed.

Results

The following are the results for the studies conducted on living termites. The

ANOVA tests on initial mass (df=3, F=1654.5, P-value<.0001), %TBW (df=3, F=81.8, P-

value<.0001), and CP values (df=3, F=113.3, P-value<.0001) of live termites revealed

65

that there were differences between species. Student’s t post hoc tests separated the

differences for live termites (Table 4-1). Initial mass was found to be significantly

different for all pairs tested except Cr. cavifrons and C. formosanus, with N. jouteli being

the heaviest and C. formosanus and C. cavifrons the lightest (Table 4-1). Additionally,

%TBW was found to be significantly different for all four species (Table 4-1). Initial

mass, %TBW, and CP values were determined for an average individual of each

species. The highest CP value was found for an average N. jouteli individual, and was

significantly different from the other species. The lowest CP values belonged to the

Cryptotermes species, which, while not different from each other, were significantly

different from the other two species. Lastly, the CP value for C. formosanus was also

different when comparing among species and was intermediate in value between N.

jouteli and the Cryptotermes species (Table 4-1). The RH in the desiccation chamber

fell in the range of 0.3-3.3%RH (1.6±0.16 average), with a temperature range of 25.2-

26.6°C (26.1±0.08 average). There was a linear relationship between cumulative %TBW

loss and time (Table 4-2 and Figure 4-3). The slopes of each line were positive

indicating an increase in losses over time. However, the rate of %TBW loss was found

to be not significantly different between all pairs (all Z<1.96 at ɑ=0.05). We can see that

for the four termite species, in general, as the CP value increases, so does the RH

equilibrium level (Figure 4-4). This relationship indicates RH equilibria can generally be

used in place of CP values and that termites that are less able to prevent body water

from evaporating through the cuticle reach higher RH equilibrium levels because of this

water loss.

66

The ANOVA on the %RHs attained in the chambers indicated that there was a

significant difference in the RH equilibria attained by each termite species and a control

at 16 hours after introduction (df=4, F=27.4, P-value<.0001). Post hoc Student’s t tests

revealed that N. jouteli and C. formosanus, with the higher two RH equilibria, were not

significantly different from each other, but were significantly different from the lower

equilibria of Cr. cavifrons and Cr. brevis (as well as the control) (Table 4-3 and Figure 4-

2). The Cryptotermes species and control were also not significantly different from each

other (Table 4-3). Readings from temperature and humidity probes indicated that the

ambient conditions of the laboratory were 47.3 %RH and 23.9°C on average.

The following are the results for the studies using dead termites. The ANOVA

tests on initial mass (df=3, F=4944.4, P-value<.0001), %TBW (df=3, F=114.5, P-

value<.0001), and CP values (df=3, F=583.6, P-value<.0001) of dead termites, revealed

that there were differences between species. Student’s t post hoc tests separated the

differences for dead termites (Table 4-1). Initial mass was found to be significantly

different for all pairs tested except Cr. cavifrons and C. formosanus, with N. jouteli being

the heaviest and C. formosanus and C. cavifrons the lightest (Table 4-1). Additionally,

%TBW was found to be significantly different for all species except N. jouteli, which was

only significantly different from C. formosanus (Table 4-1). Initial mass, %TBW, and CP

values were determined for an average individual of each species. The CP values for

dead termites were significantly different from each other when comparing species. The

highest value was found for N. jouteli, while the lowest values was found for Cr. brevis.

The CP values for C. formosanus and Cr. cavifrons were intermediate in value between

those of N. jouteli and Cr. brevis (Table 4-1). The RH in the desiccation chamber fell in

67

the range of 4.5-17.8 %RH (11.5±0.72 average), with a temperature range of 25.4-

26.8°C (26.2±0.10 average). These values for %RH for dead termites in the desiccation

chamber were higher than those found for live termites presumably because

physiological mechanisms for preventing water loss were not available for dead

individuals. There was a linear relationship between cumulative %TBW loss and time

(Table 4-2 and Figure 4-6). The slopes of each line were positive indicating an increase

in losses over time. However, the rate of %TBW loss was found to be not significantly

different between all pairs (all Z<1.96 at ɑ=0.05). We can see that for the four termite

species, in general, as the CP value increases, so does the RH equilibrium level (Figure

4-7). This relationship indicates RH equilibria can generally be used in place of CP

values and that termites (i.e. dead individuals) that are not able to prevent body water

from evaporating through the cuticle reach higher RH levels because of this water loss.

The ANOVA on the %RHs attained in the chambers indicated that there was a

significant difference in the RH equilibria attained by each termite species and a control

at 16 hours after introduction (df=4, F=50.5, P-value<.0001). Post hoc Student’s t tests

revealed that all four species and the control were significantly different from each other,

except the two Cryptotermes species, which were not different from each other (Table

4-3 and Figure 4-5). N. jouteli had the highest RH equilibria and Cr. cavifrons the lowest

of the four termite species (Table 4-3). The overall comparison of individual living and

dead termite initial masses, %TBW, and CP value revealed significant differences.

These significant differences were found between living and dead individuals of C.

formosanus, Cr. brevis, and N. jouteli for initial mass, C. formosanus, Cr. cavifrons, and

N. jouteli for %TBW, and Cr. cavifrons and N. jouteli for CP value. The overall

68

comparison of RH-equilibria of living and dead termites indicated that dead termites of

each of the four species resulted in significantly higher RH levels (Table 4-3). As

previously indicated, readings from temperature and humidity probes indicated that the

ambient conditions of the laboratory were 47.3 %RH and 23.9°C on average.

Discussion

The average RH curves (Figure 4-2 and 4-5) indicate that the body water of living

and dead individuals of each species increased the %RH in their chambers before

achieving a plateau around hour 16. The levels reached with dead termites were higher

than those seen with live termites (Table 4-3). However, the only species to reach levels

near 100% RH was N. jouteli. This is interesting because it indicates that while the

physiological mechanisms present for living termites are eliminated with dead termites,

physical mechanisms remain that prevent body water from rapidly leaving a corpse. If

the termites are thought of as membrane bound water reservoirs, living and dead

termites lose water across this membrane, but dead termites are unable to regulate this

loss through physiology as the live ones are. The membrane, however, is physically

able to continue preventing water loss after death, partially trapping water inside the

corpse, with the permeability of the cuticle to water and the breakdown of the corpse

ultimately leading to a change in the RH levels seen. The use of ethyl acetate to kill the

termites may have also affected the RH levels reached. Closure of the spiracles to

avoid ethyl acetate vapor entering the respiratory system may have also lead to the

prevention of water vapor diffusing and evaporating from the spiracles. However, the

amount of water lost through respiration and whether all or a portion of the spiracles

were closed is unknown. Additionally, the multi-compartment water loss model (Machin

69

1981 and Sponsler and Appel 1990) that assumes that cuticular and extracellular water

is lost more rapidly than tissue or intracellular water would be supported here. Live

termites that are under desiccative conditions would have to rely on tissue and

intracellular water as they are exposed to these conditions for longer and their cuticular

and extracellular water resources are being exhausted (see recovery of N. jouteli in

Chapter 3). Dead termites, however, would be expected to decompose, releasing water

until it is completely evaporated (in this case until the RH level reaches 100%). Here,

the RH levels were significantly higher for dead termites than living termites, but the

cuticles of the dead individuals still physically prevented the rapid loss of water (as

vapor) reflected in the fact that the RH levels did not reach 100% (except with N. jouteli)

in the chambers even 44 hours after death. In terms of CP value, the value determined

for C. formosanus workers in Sponsler and Appel (1990) (37.49±2.32) was

approximately three times higher than the value found for living termites in this study

(13.0±0.75). This discrepancy is likely due to their use of mass loss after 2 hours, a

period of time that represented the maximum water gradient between the termites and

the chamber, resulting in maximal water loss (Sponsler and Appel 1990). The current

CP value for living N. jouteli was the only value comparable to the CP values of several

of 10 species of adult male cockroaches (Appel et al. 1983). This value was also

comparable to those found for C. formosanus workers by Shelton and Grace (2003).

The CP value found for live workers of C. formosanus was not comparable to those of

the same species and caste, as well as workers of other subterranean species in

previous studies (Shelton and Grace 2003, Hu et al. 2012). The CP values for the

Cryptotermes species, however, were comparable to values found for Incisitermes spp.

70

of drywood termites by Rust et al (1979). Comparing the RH equilibria and the CP

values for each species indicates that over time, C. formosanus and N. jouteli lose more

body water through the cuticle to achieve higher RH equilibria, whereas the

Cryptotermes species lose very little from their bodies resulting in much lower RH

equilibria. However, determining RH equilibria for a given volume does not involve the

use of surface area estimates (such as Meeh’s formula), but can still relate loss of body

water through the cuticle to RH and temperature. The cuticle thicknesses measured in

Chapter 2 did not strongly support the idea that this thickness correlates to lower CP

values. In this case, the thinnest cuticle corresponded to the second highest CP value

(C. formosanus), but the thickest cuticle belonged to N. jouteli which also had the

highest CP value for both living and dead individuals. The cuticle thicknesses for the

Cryptotermes species were between these extremes. This indicated that the thickness

of the cuticle is not the only factor in determining how readily body water evaporates

from the cuticle and highlights the problems with surface area estimates for determining

CP values.

Additionally in this study, RH plateaus were evaluated as an alternative to CP

calculation due to the problems with using formulas (e.g. Meeh’s formula) to estimate

surface area of small organisms such as termites. With the determination of a RH

equilibrium and the time taken to reach this RH level, the rate of water loss from the

body (%TBW loss) can be calculated and dependence on surface area estimates is

avoided. This is a simpler method for approximating the level of water loss from the

body of a termite over time and the humidity level required to resist desiccation. The

results of this study showed that there were species-specific RH equilibria plateaus

71

reached by the four species. The water loss from the bodies of groups of termites due to

the CP of each species resulted in a homeostasis with the RH conditions of their

surroundings (chamber). That is to say, for the internal conditions of the termite to

remain relatively constant, body water lost through the cuticle as vapor reached a point

of balance with the RH humidity conditions of the closed system of the chamber. If we

think of this in terms of diffusion of vapor or gas, the vapor pressure from the RH in the

atmosphere of the confined space of the chamber and the vapor pressure from

evaporation from the termite body will reach an equilibrium. Because the initial

conditions of the chamber are at lab ambient (low RH), water moves from the termite

body into the atmosphere until the equilibrium is reached and water is no longer lost

across the cuticle. Ultimately, the loss of water from the termite body is affected by the

makeup of the cuticle (i.e. - lipid and/or cement layers), and activity of the mouth, anus,

and spiracles. Shelton and Appel (2000) found that water loss is coupled to the cyclic

release of CO2 from the respiratory system (spiracles), but that this did not contribute to

reductions in water loss in R. flavipes (Kollar) alates. They also found water loss

coincided with CO2 release in I. minor, and that cuticular water loss and respiratory

water loss made up the majority of daily water loss. Defecation events also added to

water loss in this species and surely do in others as well (Shelton and Appel 2000).

If water resources are readily available, a RH equilibrium achieved through loss

of body water should be stabilized. If these conditions are not met, the termite

eventually loses most of its %TBW (desiccates) and has to either locate an adequate

water source, more favorable surroundings, or face mortality. Given the natural habitats

of the four termite species, these equilibrium plateau levels make sense. N. jouteli and

72

C. formosanus require a high %RH and, thus, would need a higher RH equilibrium in

the chamber and CP values that allow for this. The Cryptotermes species, however, do

not require a high %RH microhabitat, and so, did not need as high of a RH equilibrium

in their chambers. In fact, their equilibria and CP values are so low that they lost very

little body water and the RH equilibria were not significantly different from the control.

This indicates that these species employ physiological control of water loss to keep

such valuable water resources retained within the body.

The high initial body mass and %TBW of N. jouteli lent themselves to the highest

plateau of the four species (Table 4-1). C. formosanus, however, had the lowest living

initial body mass and second lowest dead initial body mass, lowest living and dead

%TBW, and only reached a RH plateau around 75%. This is interesting because C.

formosanus was observed in the experiment in Chapter 3 to require a humidity level of

at least 90% (with a food source) to survive an extended period of time. This indicates

that more individuals in the chamber would have allowed for a higher %RH while not

using as much of the TBW of each individual, but that this equilibrium was adequate for

the number of termites used, the volume of the chamber, and the time period of the

study. While C. formosanus and Cr. cavifrons did not differ from each other in living and

dead initial mass, live Cryptotermes spp. exhibited the middle two initial masses and

%TBWs. Both these species also equilibrated at the lower two RH levels. Overall

differences seen with the comparison of living and dead individuals for initial mass,

%TBW, and CP value are most likely attributed to colony differences, as living and dead

termites were examined separately and at different times in this study because there

were not enough termites available to run them at the same time. We would expect

73

these values not to be different if samples came from the same laboratory population or

if the same specimens could be used while living and after death. Despite this, however,

the values associated with living and dead individuals were relatively close. Additionally,

the %TBW of all four species (living or dead) are within the range (~60-80%) of content

levels for insects found by Edney (1977).

The humidity chambers used herein were relatively large in volume (≈112.5 cubic

centimeters) in comparison to their normal living conditions. It should be easier for

individuals and groups of termites to attain these equilibria in the foraging tunnels and

galleries in the wood and soil of their microhabitats, especially when the substrate often

holds water as well. It is also important to note that modification of an environment

through behavior is an active process whereas equilibria were attained passively. The

termites could not actively prevent body water from evaporating across the cuticle other

than by decreasing activity, the possible regulation of respiration, and reducing

defecation events. The combination of both active and passive desiccation tolerance

factors results in the overall ability of different termite species to tolerate various

environmental conditions.

Termites must often create (behaviorally) or locate habitats (during colony

foundation) that lend themselves to reducing or tempering the effects of cuticular

permeability. If they do not find a habitat/microhabitat that is suitable in terms of RH and

water availability during colony foundation, the colony will not survive to become

established. The preceding study was influenced by the work of Appel et al. (1983,

1986) on the water relations in cockroaches and the work of Sponsler and Appel (1990)

on the aspects of water relations in termite workers, soldiers, and alates. The

74

information provided from these previous studies included information regarding

cuticular permeability and rate losses, and highlight the need for species ill-equipped to

prevent body water from escaping, to counteract this problem through behavioral

mechanisms. This is most clearly illustrated in the case of C. formosanus. Individuals

and groups of this species do not survive long in an environment with a low %RH and

lack of water resources, but presented with enough water, a substrate, and the

opportunity to modify their immediate environment through construction behaviors, they

often survive very well. This was seen in Chapter 3 in an environment with a high %RH.

Given that evaporation of termite body water creates RH equilibria that help prevent

desiccation, termites should select environments (when given a choice) where

evaporation and loss of this water to the atmosphere is at a minimum. The next chapter

deals with how termites distribute themselves when given a choice of RH levels.

75

Table 4-1. Initial mass (mg), percentage of total body water (%TBW), and cuticular permeability (CP) value of an average live individual worker of four termite species exposed to 0.3-3.3% RH and ≈26.1°C and an average dead individual worker of the same four species exposed to 4.5-17.8% RH and ≈26.2°C

Variable Species Live

individual Dead

individual

Initial massa

C. formosanus 2.97±0.10Aa 3.85±0.06Ab

Cr. brevis 7.50±0.21Ba 5.44±0.06Bb

Cr. cavifrons 3.29±0.06Aa 3.71±0.05Aa

N. jouteli 23.7±0.41Ca 19.3±0.19Cb

%TBWa

C. formosanus 70.8±0.63Aa 63.8±0.83Ab

Cr. brevis 76.7±0.45Ba 78.1±0.56Ba

Cr. cavifrons 73.7±0.39Ca 75.6±0.50Cb

N. jouteli 80.2±0.23Da 76.6±0.50BCb

CP valueab

C. formosanus 13.0±0.75Aa 14.6±0.25Aa

Cr. brevis 3.80±0.59Ba 2.55±0.18Ba

Cr. cavifrons 2.66±0.29Ba 4.83±0.20Cb

N. jouteli 22.7±1.40Ca 29.9±0.96Db amean±SEM not followed by the same uppercase letter within a column and same lowercase letter within a row are significantly different (Student’s t at ɑ=0.05) bmean cuticular permeability (µgH2O*cm-2*h-1*mmHg-1) values for mass loss of an individual after 12 hours

76

Table 4-2. Regression equations of cumulative percentage of total body water content (%TBW) lost over time (hours) for live termites of four species exposed to 0-3.3% RH and 26°C and dead termites of the same species exposed to 4.5-17.8% RH and ≈26.2°C

Status Species Regression equation ± SEMa r2 P-value

Live

C. formosanus Y = (4.71±1.5)±(3.68±0.20)X 0.83 <0.001

N. jouteli Y = (0.48±0.83)±(2.59±0.12)X 0.82 <0.001

Cr. cavifrons Y = (0.07±0.25)±(0.54±0.03)X 0.78 <0.001

Cr. brevis Y = (0.22±0.11)±(0.12±0.02)X 0.47 <0.001

Dead

C. formosanus Y = (4.71±1.5)±(3.68±0.20)X 0.83 <0.001

N. jouteli Y = (0.66±0.21)±(1.14±0.03)X 0.96 <0.001

Cr. cavifrons Y = (0.85±0.26)±(0.81±0.04)X 0.89 <0.001

Cr. brevis Y = (0.19±0.15)±(0.38±0.02)X 0.84 <0.001 aY is percentage total body water loss and X is time in hours. Same experimental design as in Table 4-1.

77

Table 4-3. Species-specific RH equilibriums produced from evaporation of body water from groups of 25 live workers and groups of 25 dead workers of four species of termite after 16 hours

Species Live Dead

C. formosanus 75.7±1.9Aa

81.2±1.8Ab

Cr. brevis 53.5±0.95Ba

65.1±3.2Bb

Cr. cavifrons 54.4±1.4Ba

64.9±1.2Bb

N. jouteli 80.6±4.6Aa

93.7±3.3Cb

control 49.0±0.79Ba

49.4±0.99Da aMean±SEM not followed by the same uppercase letter within a column and same lowercase letter within a row are significantly different (Student’s t at ɑ=0.05)

78

Figure 4-1. Setup for determining relative humidity equilibria in a confined space for four termite species. A) Uncovered evaporation chamber. a: temperature/humidity probe b: probe collar c: petroleum jelly seal. B) Covered evaporation chamber. d: cloth covering e: cup covering. C) Close-up of jar and lid components of evaporation chamber. f: chamber lid with probe access hole g: jar chamber with scuffed base. Photographs courtesy of author.

79

Figure 4-2. Species-specific average %RH equilibria curves with SEM bars for live termites. Red circle indicates time used for estimate of RH equilibria for all four species with 25 live termites and ambient conditions of 47.3% RH and 23.9°C.

80

Figure 4-3. Relationship of cumulative percentage of total body water content (%TBW) lost over time for live individuals of four termite species exposed to 0-3.3% RH and 26°C. Same experimental design as in Tables 4-1 and 4-3.

81

Figure 4-4. Relationship between cuticular permeability (CP) values of an average live individual from 10 replicates of 10 live individuals and mean RH equilibria (RH-EQ) of 4 replicates of 25 live individuals for four species of termites.

82

Figure 4-5. Species-specific average %RH equilibria curves with SEM bars for dead

termites. Red circle indicates time used for estimate of RH equilibria for all four species with 25 dead termites and ambient conditions of 47.3% RH and 23.9°C.

83

Figure 4-6. Relationship of cumulative percentage of total body water content (%TBW)

lost over time for dead individuals of four termite species exposed to 4.5-17.8% RH and ≈26.2°C. Same experimental design as in Tables 4-2 and 4-4.

84

Figure 4-7. Relationship between cuticular permeability (CP) values of an average dead

individual from 10 replicates of 10 dead individuals and mean RH equilibria (RH-EQ) of 4 replicates of 25 dead individuals for four species of termites.

85

CHAPTER 5 RELATIVE HUMIDITY PREFERENCE OF FOUR TERMITE SPECIES IN A MULTIPLE-

CHOICE ARENA

Introduction

Previous studies reported termite survivorship when placed in an environment

with varying levels of relative humidity, as well as the evaporation of body water to affect

a humidity equilibrium in a confined space. However, the methods for obtaining this

information were predicated on the fact that the termites were not given a choice of

environment. The question then arose that if the four species of termite were given a

choice of RH conditions, what preference would they exhibit? An environment in which

RH is higher makes reaching a higher RH equilibrium and, thus, survival, easier

because the termites will not lose as much body water through the cuticle, preventing

desiccation. Alternatively, a higher RH environment may not be of preference to some

termites even if they are able to tolerate such conditions, and may even make survival

more difficult. Those species that naturally inhabit wetter environments are expected to

exhibit a preference for artificial environments with higher humidity levels. The converse

in terms of preference is also expected. However, preference is not necessarily defined

by a narrow range in terms of environmental conditions, and various factors may affect

differences in preference. As the experiment in Chapter 3 suggested, the Cryptotermes

species and N. jouteli are probably able to tolerate a wider range of humidities than C.

formosanus is, as evidenced by the latter’s quick desiccation. This does not mean that

the former species prefer all conditions within said range, however. For example, N.

jouteli is known to live in damp wood sources and can tolerate a wide range of humidity

levels for a time with a threshold of recovery (Chapter 3), but would likely show a

preference for wood with a certain range of moisture contents.

86

Although Collins (1958, 1969, 1991), Khan (1980), and Strickland (1950) found

that different termite species can tolerate various (sometimes extreme) conditions in

terms of humidity and temperature for varying amounts of time, there are undoubtedly

conditions that termites prefer over others that they can tolerate. The assumption was

that the four termite species would exhibit preferences for RH levels that more closely

approximate the RH conditions of their natural habitats. The goal of the study, then, was

to observe whether the termites would aggregate within one or more chambers with

certain RH levels and exhibit a humidity preference. It was hypothesized that the

drywood species, Cr. brevis, would aggregate in the lowest humidity chamber, the

“wetwood” species, C. formosanus and N. jouteli, would do so in the highest humidity

chamber, and that Cr. cavifrons would aggregate in more than one chamber based on

the fact this species can be found in dry to relatively wet wood.


Termites used for this study were collected and stored as described in the

previous chapters. Multiple-choice arenas were constructed by connecting 3-ounce

clear plastic jars, associated plastic lids, and lid linings (Uline, Wisconsin, USA) with

tubing. Four chambers with varying RH levels were connected to a central chamber

containing indicating Drierite (Figure 5-1). The average RHs (±SEM) of the chambers

and their associated stabilizing material were as follows: 90.7±0.24% (H2O),

71.7±0.20% (NaCl), 52.0±0.12% (Mg(NO3)2), 34.1±0.16% (MgCl2), and 8.4±0.48%

(Drierite). Access to the chambers for the temperature and humidity probes were

allowed through approximately 1.59 cm holes in the lids. These holes were plugged with

rubber stoppers while not in use. Temperature and humidity levels were measured

before termites were introduced to the arena using an Amprobe THW3 probe fitted with

87

a plastic collar to allow for upright free-standing. Probe readings were used to determine

the average RH of each chamber from readings taken before termites were introduced.

Relative humidity stabilizing solutions/materials were placed in the bottom of the jar

chambers. Chambers housed modified plastic vials (25 mm inner diameter, 15 mm

height) cut to hold the termites while preventing their escape (holding dish). A hole (6

mm diameter) was drilled into the wall of each of the four radial chambers (2.5 cm from

base) to allow the tubing (5.5 mm inner diameter, 50 mm long) to connect their elevated

holding dishes to the elevated holding dish in the central introduction chamber. This

tubing was scratched using a Dremel® multi-purpose cutting bit attachment to provide

friction for their tarsal claws and facilitate termite movement between and among the

chambers. Groups of 25 termites were placed into the introduction chamber for N.

jouteli, while 50 termites were used for the other three species. The large size of N.

jouteli meant 50 individuals would not fit in one holding dish, which, because of area

available to hold individuals, would influence where preference was exhibited. Termites

were allowed to move and acclimate to any chamber for at least 12 hours (maximum 16

hours). Black satin cloth covered the arenas to provide darkness. Counts of termites

within each chamber were then recorded. Termites found in the connecting tubes were

counted as occurring in the chamber the tube was connected to.

Statistical analysis was conducted using the JMP statistical software. A 4x5

factorial analysis was conducted with species and RH level as the factors and termite

preference (percent found in chamber) as the response variable. The percent values

were arcsine-square root transformed before the analysis of variance (ANOVA) was

used to test whether there was a significant difference among RHs for the four species.

88

Post hoc Student’s t tests were used to separate the differences for each pair at ɑ=0.05.

Data from six replications were analyzed.

Results

The ANOVA for the 4x5 factorial experimental design indicated differences for

RH and species-RH interaction (Table 5-1). Overall distribution was significantly higher

in the 90.7% RH chamber (50.5% of all termites) when compared with the other

chambers. Overall RH mean was affected by the preferences exhibited by each species

for a given RH level (chamber). Post hoc Student’s t tests indicated the following when

comparing preference between species for given RHs (Table 5-2). Percentages of total

individuals for a given species found in a given RH chamber after 12-16 hours are

shown in parentheses. In some cases, “preferences” are supported as significantly

higher statistically, but are low enough that they likely reflect a transient state in a given

RH. C. formosanus (89.3%) and N. jouteli (91.3%) preferred the 90.7% RH significantly

more than Cr. cavifrons (18.3%) and Cr. brevis (3.0%) did. Cr. cavifrons (42.3%)

preferred the 71.7% RH significantly more than C. formosanus (4.0%), N. jouteli (4.7%),

and Cr. brevis (7.0%). Cr. brevis (30.7%), C. formosanus (3.3%), and N. jouteli (3.3%)

all preferred the 52% RH significantly more than Cr. cavifrons (1.0%). C. formosanus

(2.0%), Cr. brevis (17.7%) and Cr. cavifrons (21.0%) preferred the 34.1% RH

significantly more than N. jouteli (0.7%). Lastly, C formosanus (1.3%), Cr. cavifrons

(17.3%), and Cr. brevis (42.3%) preferred the 8.4% RH significantly more than N.

jouteli. In fact, N. jouteli individuals were not found in the chamber with 8.4% RH at all

(Table 5-2). The species with the highest percentage of termites found in the RH

chambers were as follows: in the 90.7% chamber it was N. jouteli with 91.3±2.8%, in the

71.7% chamber it was Cr. cavifrons with 42.3±19.0, in the 52.0% chamber it was Cr.

89

brevis with 30.7±19.5, in the 34.1% chamber it was Cr. cavifrons with 21.0±13.2%, and

in the 8.4% chamber it was Cr. brevis with 42.3±15.9%. Comparing RH preferences

within species revealed that the wetwood (nests in wood or soil high in moisture)

species of C. formosanus and N. jouteli that preferred the 90.7% RH significantly more

than the other four RHs (89.3 and 91.3 %, respectively). Cr. cavifrons only exhibited a

significant difference in preference between the 52% and 71.7% RHs (1.0 versus 42.3

%, respectively). Cr. brevis only showed a significant difference in preference between

the 90.7% and 71.7% RH and the 8.4% RH (3.0 and 7.0 versus 42.3%, respectively). It

should be noted that approximately 58% of Cr. cavifrons and 43% of Cr. brevis

individuals were located within connecting tubes.

Discussion

Preferences for RH levels were observed for each termite species. Preference of

termites in humidity gradient arenas has also been reported in a study by Gautam and

Henderson (2011), who found that C. formosanus workers (and soldiers) aggregated in

the highest RH area of the arena, as would be expected. Cr. brevis was characterized

as feeding efficiently at either medium (≈ 60%) or high (≈ 90%) RH in Steward (1982).

Woodrow and Grace (1999) found that Cr. brevis infesting structural lumber in Hawaii

were associated with a wide range of wood core temperatures, with a minimum of

13.9°C and a maximum of 43.3°C (24.3°C mean), and an even wider range of ambient

humidities, with a minimum of 27.2% RH and a maximum of 98.2% RH (75.1% RH

mean). This supports the idea that the drywood species are able to tolerate a wide

range of RHs. In addition, a study on Microcerotermes beesoni Snyder (family

Termitidae) by Sen-Sarma and Chatterjee (1966) found that this termite is highly

hygropositive, orienting and relocating to the 90-95%RH section of their apparatus. This

90

makes sense, given that this species inhabits environments in India similar to those

inhabited by C. formosanus.

The natural habitats of the four species used in this study reflect what we see in

the results. C. formosanus is a subterranean species that requires not only high RHs,

but also other sources of water for normal activity and survival (Gautam and Henderson

2011). The dampwood termite, N. jouteli, which lives within its moist food source does

as well. Although the habitats of the Cryptotermes species are similar to each other in

that both inhabit and feed within wood, Cr. brevis, is almost exclusively limited to dry

timber found in structures and furniture, whereas Cr. cavifrons can be found in dry to

moist wood in natural habitats such as forested areas. Cr. brevis was found in a natural

woodland area in Hawaii (Scheffrahn et al. 2000) as well as its endemic origin in

Northern Chile and Peru. The endemic climate of this drywood is unusually stable and

humid despite a lack of rainfall (Scheffrahn et al. 2009). The discovery of the endemic

origin, additional natural and urban populations of Cr. brevis in Hawaii, and the climatic

characteristics associated with these areas, lend themselves to the idea that this

species is capable of tolerating a wider range of temperature and humidity conditions

than previously thought (Scheffrahn et al. 2000, 2009, Steward 1981, 1983). Despite

being almost exclusively found associated with human structures and goods, Cr. brevis

appears to be able to tolerate and acclimate to different conditions in order to prevent

death from desiccation, water poisoning, or overheating.

These results generally agree with the results of the aforementioned studies, with

the “wetwood” species aggregating in the high RH chamber, and the drywood species

not exhibiting a major preference for any chamber. However, the proportion of drywood

91

individuals that were located within tubes and not actual chambers reflects the idea that

huddling in a small space is an important behavioral mechanism for resistance to

desiccation. This huddling has been noted by Grasse and Chauvin (1944) as the “group

effect.” This aggregation and the increase in survival it provides is also well supported in

the literature for several species including Cr. brevis (Abushama 1974, Ahmad et al.

1982, Malik and Sheikh 1990, Minnick et al. 1973, Pence 1956, Sen-Sarma and

Chatterjee 1966). Results from a study by Cabrera and Rust (1996) on I. minor,

suggested that temperature and humidity levels have a greater influence on water loss

for termite individuals than on groups. They noted, however, that temperature seemed

to have the greatest effect on aggregation behavior even though this species was

known to aggregate in the absence of strong external stimuli. This indicates that

aggregation may serve additional purposes in addition to helping tolerate desiccation.

All four species exhibited this behavior though, with the drywood species aggregated

within the tubes at the percentages reported. In addition, the drywood species were

observed to have begun sealing the tubes with feces, another possible behavioral

mechanism for resisting desiccation. Construction behaviors will be examined in

Chapter 7.

We can see that relative humidity is an important factor in the survival of termite

individuals and groups. Given a choice of RH level, they will exhibit preferences

accordingly and employ behavioral mechanisms to prolong the life of the group. An

interesting question is whether behavior before or after making a microhabitat choice is

more important. It is possible that, once microhabitat and grouping are established, the

search for a more favorable environment is not undertaken as long as the conditions

92

have exceeded a tolerance/favorable threshold. The group does not relocate to a more

favorable environment if they have already acclimated to one that is not unfavorable.

The size of the group and/or the size of the microhabitat may also have a bearing on

such aggregation, searching, or foraging behavior. Future studies dealing more directly

with the interaction of huddling behavior, group size, and RH humidity is warranted.

93

Table 5-1. Analysis of variance for a factorial experiment to evaluate preference of workers of four species of termite exposed to an arena with five different RH levels for 12-16 hours


MS b F P-value

Species (S) 3 0.004 0.03 0.9925

Chamber (C) 4 1.287 11.2 <.0001

SxC 12 0.854 7.43 <.0001

Error 100 0.115 --- ---


b Mean square.

94

Table 5-2. Effects of species and relative humidity on preference of termites in a multiple-choice arena (mean±SEM)

RH levelabc

(90.7±0.24%) (71.7±0.20%) (52.0±0.12%) (34.1±0.16%) (8.4±0.48%)

[23.4±0.06°C] [23.5±0.06°C] [23.4±0.06°C] [23.4±0.06°C] [23.4±0.05°C]

Species

Overall mean preference for

speciesd

C. formosanus 89.3±2.5Aa 4.00±1.5Ab 3.30±1.4Ab 2.00±0.9Ab 1.30±0.8Ab 20.0±6.5A

N. jouteli 91.3±2.8Aa 4.70±1.2Ab 3.30±2.6Ab 0.70±0.7Bb 0.00±0.0Bb 20.0±6.7A

Cr. cavifrons 18.3±16.4Bab 42.3±19.0Ba 1.00±1.0Bb 21.0±13.2Aab 17.3±8.8Aab 20.0±6.0A

Cr. brevis 3.00±2.3Ba 7.00±4.3Aa 30.7±Aab 17.7±10.9Aab 42.3±15.9Ab 20.0±5.8A

Overall mean preference at RH leveld

50.5±9.3a 14.5±5.7b 9.6±5.3b 10.3±4.4b 15.3±5.5b

a Means for the combination of treatment (species x RH level) followed by the same lowercase letters within a row or means followed by the same capital letter within a column are not significantly different at the ɑ=0.05 level (Student's t).

b Value for the combination (species x RH level) of treatments are means of 6 observations of 50 termites per arena (25 termites for N. jouteli).

c Mean RHs produced by water, three salts, and silica gel found in parentheses (mean temperature±SEM in brackets).

d Values for each main treatment effect (species, RH level) are means of 30 and 24 observations, respectively. Means followed by the same letter are not significantly different at the ɑ=0.05 level (Student's t).

95

Figure 5-1. Setup for determining RH level preferences of four termite species. A) Arena with chamber lids in place. a: chamber lid b: rubber stopper c: jar chamber. B) Arena with chamber lids removed. d: filter paper semicircle e: Drierite introduction chamber f: H2O chamber g: MgCl2 camber h: Mg(NO3)2 chamber i: NaCl chamber. C) Close-up of arena components housing termites. j: connecting tube k: holding dish l:filter paper food source. Cloth covering for arena not shown. Photographs courtesy of author.

96

CHAPTER 6 UTILIZATION OF WATER SOURCES BY FOUR SPECIES OF TERMITE

Introduction

The availability of water as humidity is an important component of the

environmental conditions that affect the health and survival of termite individuals and

their colony. Humidity is not the only water source available, however. There are other

types of water resources that termites might use at any given time. These include water

accumulated due to dew point depression, metabolic water obtained through the

breakdown of food, free liquid water (if encountered), as well as water bound to various

substrates (e.g. soil, food, cadavers). We have seen how well termites survive under

varying levels of RH, their preferences when given a choice of RH levels, and how body

water evaporation (CP) leads to a RH equilibrium. A further step in examining how

termites manage their water resources is to examine how well termites survive when

exposed to various sources of water. Will groups of termites from four species be able

to utilize these different water sources to prevent desiccation?

Depending on the amount and how often it rains, free water is generally not

readily available for long in the environments termites are found. However, soil

moisture, water vapor, water bound in living or dead nestmates, and/or the moisture of a

food source are often available. Different termite types (i.e. – drywoods, dampwoods,

subterranean) utilize or avoid certain environmental conditions in which these moisture

sources are present. Whereas drywoods depend heavily on metabolically derived water,

dampwood and subterranean termites that live in high RH environments likely

supplement environmental obtained water with that derived metabolically (Brammer and

Scheffrahn 2007, Lee and Wood 1971, Scheffrahn and Su 2007b). As it would be

97

advantageous to be able to utilize all types of water resources, if termites cannot utilize

all types, there must be a cost-benefit relationship intricately linked to a given species’

normal, preferred habitat. How, then, do different termite species normally associated

with a given habitat type react to a different one? Will they adequately acclimate to this

new environment and be able to utilize the water sources available? To answer the

question of whether the termites could utilize different water sources, an experiment

was set up that placed groups of four different termite species in chambers housing

different types of water resources. The goal was to test whether there was a difference

in survival of the termite species that would indicate their utilization of the various water

resources and whether the termites would acclimate to the environmental conditions

over time. It was hypothesized that, over four weeks, the wetwood species normally

found nesting in wet wood and soil, C. formosanus and N. jouteli, would survive better in

the chambers with water sources directly available, whereas the drywood species would

survive better in chambers without water sources directly available.



previous chapters. Experimental chambers were constructed using 3-ounce clear plastic

jars with plastic lids and lid liners (Uline, Wisconsin, USA). The bottoms of these jars

were scratched with sandpaper to provide friction for termite tarsal claws and facilitate

termite movement. Temperature and humidity levels inside the jars were measured

using an Amprobe THW3 probe fitted with a plastic collar to allow for upright free-

standing during the experiment. Access to the chamber for the probe was allowed

through an approximately 1.59 cm hole in the lid and liner. These holes were plugged

with rubber stoppers while not in use. These readings were only taken as a means to

98

monitor the RH of the chambers used in the experiment. Five chambers that provided

various sources of water, or the lack thereof, were used for the experiment (Figure 6-1).

These included: chambers with a piece of dry wood (Pinus sp.) (18x18x8 mm), a piece

of dry wood with a humidity source, a piece of wet wood, a piece of dry wood with wet

sandy soil, and a piece of dry wood with a source of free water. Humidity sources were

composed of filter paper strips (2x6 cm) that had been wetted with deionized (DI) water.

Two of these wetted strips were attached to the sides of the jar chambers near the top

lip to prevent termite access to the moist paper. The wet wood was soaked in deionized

water for at least one day before being placed into a chamber. Free water sources were

provided by modifying the cap of a Fisherbrand Type 1, 1 dram glass shell vial (15x45

mm) into a small water reservoir (water dish). Wet, sandy soil chambers were made by

adding approximately 1.5 mL DI water to 5 grams of sandy soil that had been sterilized,

dried in a 60°C oven for 2 days, and passed through a #25 sieve. After 10 termites were

placed into the jars, the lids were screwed on and stopped. Once a week (as needed),

3-5 drops of water were added directly to the wood in the wet wood chambers, to the

soil in the soil chambers, to the water dishes of the wood with liquid water chambers,

and to the filter paper humidity sources. Each week for four weeks, three sets of each of

the chambers were broken down and the termite survival recorded (12 sets) for each

species.

The data were analyzed using the JMP statistical software and a 4x5 factorial

experimental design with termite species and water source as the factors and

percentage survival as the variable. Percentage survival data were arcsine-square root

transformed before analysis. Each week was analyzed separately. ANOVAs were

99

conducted to test for significant differences and LSM Student’s t tests were used to

separate the differences for each pair at ɑ=0.05. Data from three replications were

analyzed.

Results

The RH means (±SEM) for the chambers were as follows: dry wood

(49.6±0.61%), wet wood (91.0±0.22%), dry wood and free water (91.1±0.22%), dry

wood and wet soil (92.6±0.22%), and dry wood with humidity source (94.2±0.18%). The

temperature means (±SEM) for the chambers were as follows: dry wood (24.1±0.08%),

wet wood (24.2±0.08%), dry wood and free water (24.2±0.07%), dry wood and wet soil

(24.2±0.07%), and dry wood with humidity source (23.0±0.03%). The ANOVAs for each

week revealed significant differences in termite species and species-water source

interactions for every week, but differences in water source only for week 1 (Table 6-1).

After 1 week of exposure (Table 6-2), mean separating post hoc LSM Student’s t

tests revealed that species mean survival was significantly lower for C. formosanus and

Cr. brevis compared with N. jouteli. Mean water source survival was found to be lower

for the dry wood chamber when compared with the other four chambers. When

comparing survival for species with each water source, significantly lower survival was

found for C. formosanus exposed to dry wood. Percentages of survival in the five water

source chambers for both N. jouteli and Cr. cavifrons were not significantly different.

Lower survival was found for Cr. brevis exposed to wet wood and soil when compared

with dry wood coupled with a RH source. When comparing within water sources, Cr.

brevis exhibited lower survival when exposed to wet wood when compared with C.

formosanus and N. jouteli. For dry wood, lower survival was found for C. formosanus

compared with the remaining species. Lower survival was found for Cr. brevis exposed

100

to soil when compared with C. formosanus. No difference was found for any species

exposed to wood with free water. Finally, lower survival was found for Cr. cavifrons

exposed to dry wood with a RH source, when compared with N. jouteli and Cr. brevis

(Table 6-2).

After 2 weeks of exposure (Table 6-3) post hoc separation tests revealed that

termite species mean survival was higher for N. jouteli than for C. formosanus and Cr.

brevis. Survival for Cr. cavifrons was also higher than for Cr. brevis. Mean survival

among the water sources was lower for dry wood when compared with wood with free

water and wood with a RH source. Comparing survival within a species showed lower

survival of C. formosanus exposed to dry wood only. N. jouteli and Cr. cavifrons again

showed no difference in survival when exposed to any of the water sources. Lower

survival of Cr. brevis was exhibited for wet wood when compared with dry wood, wood

with free water, and wood with a RH source. When comparing within water sources for

each species, lower survival was seen for Cr. brevis exposed to wet wood and soil, as

well as for C. formosanus exposed to dry wood. C. formosanus was found to have lower

survival as compared with Cr. cavifrons exposed to wood with a free water source, and

no difference in survival was found for any species exposed to wood with a RH source

(Table 6-3).

After 3 weeks of exposure (Table 6-4), post hoc separation tests revealed that

termite species mean survival was higher for N. jouteli than for C. formosanus and Cr.

brevis. Survival for C. formosanus and Cr. cavifrons was also higher than for Cr. brevis.

No difference in mean survival was found among the water sources. Comparing survival

within a species revealed lower survival for C. formosanus exposed to dry wood only.

101

No difference in survival was found for N. jouteli and Cr. cavifrons exposed to the five

water sources. Lower survival was exhibited for Cr. brevis exposed to wet wood when

compared with dry wood. Comparing within water sources for each species indicated

lower survival for Cr. brevis exposed to wet wood and wood with a RH source, and for

C. formosanus exposed to dry wood. Cr. brevis also exhibited lower survival when

exposed to soil when compared with C. formosanus and N. jouteli. No difference was

found for survival of any species exposed to wood with free water (Table 6-4).

Lastly, after four weeks of exposure (Table 6-5) post hoc separation tests

revealed that termite species mean survival was significantly higher for N. jouteli and Cr.

cavifrons than for C. formosanus. Survival of all three was also significantly higher than

Cr. brevis. Mean survival among the water sources was found to be lower for termites

exposed to wood with free water when compared with those exposed to wet soil.

Comparing within species revealed lower survival for C. formosanus exposed to dry

wood and wood with free water. No significant differences in survival were seen for N.

joueli and Cr. cavifrons exposed to any of the water sources. Cr. brevis exhibited lower

survival when exposed to wet wood, wood with free water, and wood with a RH source,

when compared with survival when exposed to dry wood. Comparing within water

sources for each species showed lower survival for Cr. brevis exposed to wet wood and

wood with a RH source, as well as for C. formosanus exposed to dry wood. Cr. brevis

also exhibited lower survival than C. formosanus when exposed to wet soil. C.

formosanus exposed to wood with free water exhibited lower survival when compared

with N. jouteli and Cr. cavifrons, with N. jouteli also exhibiting higher survival than Cr.

brevis (Table 6-5). There was no obvious trend of increase, decrease, or stability in

102

survival with time of exposure from week to week. However, looking at the percentages

of survival after 4 weeks, there were three instances where no termites died (C.

formosanus with wet wood and Cr. cavifrons with dry wood and dry wood with a RH

source) and one instance of all the termites dying (C. formosanus with dry wood), all

others experienced decreases in survival of various degrees. This indicates that, in

general, with an increase in time of exposure to the conditions of the chambers, termite

survival would be expected to decrease.

Discussion

The effect of time of exposure on survival was that, in general, the longer

termites were exposed to a given water source (or lack thereof) and the conditions

created in the chambers, the lower their survival. This indicates that in conditions that

were considered favorable for a given species, this mortality was likely due to normal

mortality (lifespan) or handling stress. Despite fluctuations from week to week

(increases, decreases, and stability in survival percentages), when looking at the overall

species and water source means, there was always a decrease in survival from the

beginning of the experiment to the conclusion of a given week. However, separations

were influenced by the 100% morality within 1 week of all C. formosanus individuals

exposed to dry wood, something that was not seen with any of the other species. They

were also influenced by the incidence of replicates that did not lose any individuals (0%

mortality). Without a water source, C. formosanus desiccated and died quickly.

Additionally, since C. formosanus depends so heavily on external water sources, and

without such a source to use for food manipulation, they could not utilize the dry piece

of wood. In several replicates, all Cr. brevis individuals died when exposed to water

sources of wet wood, wood with water, wet soil, and wood with a RH source. While this

103

mortality was due primarily to direct contact with water, bacterial and fungal pathogens

may have also been factors. This also affected the separation of differences, but

probably not to the degree of the subterranean die-off.

This decrease in survival over time is a main function of the conditions of the

chamber and the ability of the termites to utilize the water resources in these conditions.

It is also partly due to normal mortality rates, handling stress, and group size. In the

case of Cr. brevis, it is also due to the opportunity and ability of the termites to avoid

direct contact with the water sources. Previous studies have noted greater survival in

groups of termites versus individual termites (see aggregation literature cited in Chapter

5). In this study, with such a small group size (10), losing one to a few termites reduced

survival percentages drastically and diminished the group, likely affecting the efficacy of

group behaviors. Similar studies with larger groups should be conducted in future work,

especially given the fact that termite colonies can be very large.

The results from this study again illustrate the necessity for environments with

water sources readily available for C. formosanus and N. jouteli. Without adequate

water in the air or substrate, C. formosanus desiccates quickly and dies. Despite a

dependence on moist environments, N. jouteli is better able to tolerate a lack of water

resources than C. formosanus. This tolerance is exemplified in the relatively high

proportions of survival of N. jouteli in this experiment after exposure to each water

source, albeit with flattened abdomens when direct water sources were lacking. This

supports what was seen in the RH level study. C. formosanus, N. jouteli, and Cr.

cavifrons survived relatively well in all the chambers in which water was directly

available as free water, moist food, moist substrate, or high RH. Other termite species

104

with similar habitats and lifestyles would be expected to as well. Studies examining the

tunneling activity of C. formosanus and R. flavipes, as well as C. gestroi and

Heterotermes tenuis (Hagen) (Su and Puche 2003, Arab and Costa-Leonardo 2005)

found that these species tunneled more in substrates higher in moisture content.

Additionally, the effect of substrate and food moisture levels on survival, consumption,

and distribution of Microcerotermes crassus Snyder, Macrotermes carbonarius (Hagen),

M. gilvus (Hagen), C. gestroi (Wasmann), C. formosanus, R. speratus, R. flavipes, R.

tibialis (Banks), and R. virginicus (Banks) also support the findings presented here

(Delaplane and La Fage 1989, Green et al. 2005, Hu et al. 2012, McManamy et al.

2008, Nakayama et al. 2005, Wong and Lee 2010). Termites are more likely to be

found, survive longer, and consume food resources at higher rates in environments in

which water resources are more likely to persist.

Cr. cavifrons and N. jouteli exhibited an ability to tolerate conditions in which

water sources were lacking, as well as utilize water sources when they were present (as

indicated by relatively high survival throughout). In the case of Cr. cavifrons, this is a

reflection of their natural habitat of forest wood where humidity and moisture levels are

more variable than the structural timber habitat of Cr. brevis. Cr. cavifrons also exhibited

sequestration behaviors (esp. wood with RH source), sealing themselves within the

wood block provided, avoiding the conditions of the chambers almost completely. Cr.

brevis, on the other hand, exhibited relatively high survival in the dry wood chamber, but

poorer survival when they were exposed to conditions in which direct contact with water

was possible. Bloating and morbidity from water poisoning were observed when Cr.

brevis was exposed to such conditions (i.e.-soil and wet wood chambers). This is in

105

accordance with results from previous work by Collins (1969), Minnick et al. (1973),

Steward (1982, 1983), Rudolph et al. (1990), and Woodrow et al. (2000), in which they

found evidence for water toxicity in drywood termite species. As long as individuals of

Cr. brevis can find a way to avoid this, such as huddling on the wood block provided,

they were better able to tolerate these less-than-favorable environments.

The natural habitat of Cr. brevis was reported from its endemic origin to be humid

and stable (Scheffrahn et al. 2009), indicating that this species can survive well in humid

conditions as long as direct contact with water resources is avoided. In this study, the

high percent survival of Cr. brevis in the drier chambers reflects Collins’ (1991) note of

this species’ ability to withstand drying effectively enough to die of starvation before

dying from desiccation. The natural habitat of C. formosanus is generally quite moist

(subterranean), but they employ behavioral and physiological means to modify their

environments when foraging (e.g.-construction of shelter tubes) or when the nest and/or

galleries are damaged. N. jouteli, may be able to do something similar, but they live

within their damp food source and do not forage as extensively. It is reasonable to

believe that the Cryptotermes species also employ behavioral and physiological means

to tolerate less-than-favorable conditions. The aim of the subsequent chapter is the

examination of behavioral and physiological means to resisting desiccation as a group

in Chapter 7.

106

Table 6-1. Analysis of variance for a factorial experiment to evaluate the survival of termites exposed to various types of water sources from 1 to 4 weeks

Run time


MS b F P-value

1 week

Species (S) 3 0.21 3.2 0.0322

Water source (W) 4 0.32 5 0.0023

SxW 12 0.42 6.6 <0.0001

Error 40 0.06 --- ---

2 weeks

Species (S) 3 0.81 6.1 0.0016

Water source (W) 4 0.26 2 0.1191

SxW 12 0.41 3 0.004

Error 40 0.13 --- ---

3 weeks

Species (S) 3 1.06 8.57 0.0002

Water source (W) 4 0.17 1.4 0.2516

SxW 12 0.44 3.5 0.0012

Error 40 0.12 --- ---

4 weeks

Species (S) 3 1.35 10.6 <0.0001

Water source (W) 4 0.2 1.6 0.1938

SxW 12 0.58 4.6 0.0001

Error 40 0.13 --- ---


b Mean square.

107

Table 6-2. Effects of water source and species on worker survival (%) of four species of termite after 1 week (mean±SEM)

Water sourceabc

wet wood dry wood wet soil wood+water wood+RH

(91.1±0.43%) (50.5±1.81%) (92.9±0.38%) (90.8±0.29%) (94.5±0.34%)

[24.1±0.10°C] [24.0±0.11°C] [24.1±0.09°C] [24.1±0.10°C] [23.1±0.04°C]

Species


C. formosanus 93.3±6.7Aa 0.0±0.0Ab 100±0.0Aa 100±0.0Aa

93.3±3.3ABa

77.3±10.4A

N. jouteli 96.7±3.3Aa 90.0±10.0Ba

96.7±3.3ABa 93.3±3.3Aa 100±0.0Aa

95.3±2.2B

Cr. cavifrons

86.7±8.8ABa 90.0±10.0Ba

93.3±6.7ABa

86.7±13.3Aa 76.7±3.3Ba

86.7±3.7AB

Cr. brevis

63.3±21.9Ba 86.7±8.8Bab

73.3±13.3Ba

80.0±10.0Aab 100±0.0Ab

80.7±5.9A

Overall mean survival with water source exposured

85.0±6.6a 66.7±12.1b 90.8±4.5a 90.0±4.3a 92.5±3.0a

a Weekly mean RH±SEM associated with each water source found in parentheses (mean temperature±SEM in brackets).

b Means for each combination of treatment (species x water source) followed by the same lowercase letters within a row or means followed by the same capital letter within a column are not significantly different at the ɑ=0.05 level (Student's t).

c Values for each combination (species x water source) of treatments are means of 3 observations.

d Values for each main treatment effect (species, water source) are means of 15 and 12 observations, respectively. Means followed by the same letter are not significantly different at the ɑ=0.05 level (Student's t).

108

Table 6-3. Effects of water source and species on worker survival (%) of four species of termite after 2 weeks (mean±SEM)

Water sourceabc


(90.6±0.55%) (49.3±1.1%) (91.9±0.38%) (91.2±0.47%) (93.9±0.40%)

[24.1±0.10°C] [24.0±0.12°C] [24.0±0.09°C] [24.1±0.10°C] [23.2±0.02°C]

Species


C. formosanus 90.0±5.8Aa 0.0±0.0Ab 96.7±3.3Aa 63.3±31.8Aa 90.0±10.0Aa

68.0±11.2AC

N. jouteli 96.7±3.3Aa

90.0±10.0Ba 96.7±3.3Aa 96.7±3.3ABa 96.7±3.3Aa

95.3±2.2B

Cr. cavifrons 83.3±6.7Aa 90.0±5.8Ba 80.0±11.5ABa 100±0.0Ba 86.7±3.3Aa

88.0±3.1AB

Cr. brevis

26.7±21.9Ba

73.3±21.9Bb 56.7±29.6Bab

80.0±11.5ABb

80.0±10.0Ab

63.3±9.4C


74.2±9.8ab 63.3±12.4b 82.5±8.4ab 85.0±8.5a 88.3±3.7a

a Weekly mean RH±SEM associated with each water source found in parentheses (mean tempeature±SEM in brackets).




109


Water sourceabc


(90.9±0.44%) (48.8±0.91%) (92.9±0.49%) (90.9±0.58%) (94.2±0.44%)

[24.4±0.24°C] [24.2±0.24°C] [24.4±0.22°C] [24.4±0.24°C] [22.8±0.05°C]

Species


C. formosanus 90.0±5.8Aa 0.0±0.0Ab 96.7±3.3Aa

90.0±10.0Aa 90.0±5.8Aa

73.3±10.1A

N. jouteli

90.0±10.0Aa

83.3±12.0Ba 96.7±3.3Aa 86.7±8.8Aa 100±0.0Aa

91.3±3.5B

Cr. cavifrons 83.3±8.8Aa 96.7±3.3Ba

86.7±6.7ABa 93.3±3.3Aa 90.0±0.0Aa

90.0±2.4AB

Cr. brevis

26.7±26.7Ba

80.0±11.5Bb

50.0±26.5Bab

60.0±30.6Aab

36.7±12.0Bab

50.7±10.0C


72.5±10.3a 65.0±12.0a 82.5±8.3a 82.5±8.2a 79.2±8.0a





110


Water sourceabc


(91.5±0.34%) (49.8±1.0%) (92.9±0.50%) (91.8±0.42%) (94.1±0.30%)

[24.3±0.13°C] [24.2±0.14°C] [24.3±0.11°C] [24.3±0.12°C] [23.0±0.04°C]

Species


C. formosanus 100±0.0Aa 0.0±0.0Ab 96.7±3.3Aa 33.3±33.3Ab 96.7±3.3Aa 65.3±12.4A

N. jouteli 86.7±8.8Aa 90.0±0.0Ba

93.3±3.3ABa 96.7±3.3Ba 83.3±6.7Aa

90.0±2.4B

Cr. cavifrons 86.7±8.8Aa 100±0.0Ba

90.0±10.0ABa

73.3±14.5BCa 100±0.0Aa

90.0±4.3B

Cr. brevis

26.7±14.5Ba 83.3±8.8Bb 56.7±3.3Bab

36.7±31.8ACa

26.7±14.5Ba

46.0±8.8C


75.0±9.5ab 68.3±12.2ab 84.2±5.4a 60.0±13.0b 76.7±9.6ab





111

Figure 6-1. Setup for examining utilization of various water sources (or lack thereof) by four termite species. A) Dry wood chamber. B) Wet wood chamber. C) Dry wood and free-water source chamber. D) Wet, sandy soil and wood chamber. E) Dry wood and humidity source chamber. a: wood block b: water dish c:soil layer d: filter paper humidity source e: rubber stopper f: cloth covering. Photograph courtesy of author.

112

CHAPTER 7 SURVIVORSHIP, PREFERENCE, AND BEHAVIOR IN RESPONSE TO A RH AND

MOISTURE AVAILABILITY SHIFT

Introduction

The necessity and ability of termites to prevent desiccation and survive

desiccative stress is reflected in the allocation of labor to construction, repair, and water

management within their nests and galleries. This ability is found behaviorally at the

colony level. For some species this is generally achieved through aggregation, a

decrease in activity, and simple construction (sealing small holes). Other species may

employ extensive construction of carton lattices and movement of moisture within the

nest. In addition, to achieve and stabilize a RH equilibrium within the galleries of the

colony, individual termites must work efficiently and in unison. Nests located in moist

substrates or managed to contain water resources would be expected to have humidity

levels at or near 100% RH for all or part of the life of the termite colony. Individual

termites are genetically programmed to respond to changes in the conditions of the

colony as well as the changes to nest homeostasis, including those involving humidity

levels and water availability. The ability of these individuals to act in concert and in an

efficient manner is critical to dealing with changes in a micro-habitat, such as when the

surrounding habitat changes due to biotic and/or abiotic factors. This includes intrusion

from a predator, damage to part of the nest and/or foraging area, or climatic changes

and weather patterns. The importance of preventing desiccation (or buffering

desiccative conditions) is highlighted by the fact that termites will succumb to

desiccation far quicker than they will die from starvation (Collins 1991). Many species of

termites are buffered from drastic changes in environmental conditions because they

live within their food source (one-piece feeders). Other species use construction and

113

foraging behaviors, creating nests in trees, epigeal mounds, above-ground shelter

tubes, and underground nests and foraging galleries in order to utilize or create micro-

habitats that are more favorable in terms of humidity. While there are some termite

species that forage above-ground, they tend to be exposed only for a limited period of

time. The construction of shelter tubes that connect the nest and galleries to food

sources and foraging trails are thought to protect against predators. While predation

may be a factor in their construction, shelter tubes also provide a barrier to water loss

for the colony, just as the cuticle is a barrier to water loss in individuals.

Homeostatic conditions are, by definition, relatively stable (optimum) conditions

that are regulated by an organism with regard to the various changing conditions of their

habitats or body. As Emerson (1956) states via Cannon, homeostasis is, “self-regulation

of optimal conditions of existence and survival.” In his paper, Emerson explored social

homeostasis of the termite nest in terms of construction behavior, regeneration (repair),

temperature, humidity, and the gaseous atmosphere. He indicated that the internal

humidity of the nest is likely an important homeostatic factor, and that construction and

repair of the nest is undertaken to establish a balance between nest and habitat

(including protection from predation). How, then, do termite species that inhabit certain

microhabitats respond to drastic changes in conditions from favorable to unfavorable (or

vice versa) in a short period of time, and how well do they survive? To test whether

group behavior could prevent collapse due to desiccation from a quick and drastic

change in humidity and water availability, two separate shelter/refuge studies were

conducted, one method for C. formosanus and N. jouteli (“wetwoods”), and another

method for Cr. cavifrons and Cr. brevis (drywoods). It was hypothesized that all four

114

species would sequester themselves within refugia (chambers) and utilize shelters to

buffer against a less favorable environment.



previous chapters. The dual-chamber arenas were constructed by modifying a plastic

container (with associated lid) and a Plexiglas tube (Figure 7-1). Clear polystyrene

containers (17.15x12.22x6.03 cm) were used as the main (outer) chamber (Tri-State

Plastic, Inc., Kentucky, USA). The circular inner chamber (refugia) was formed using a

piece of Plexiglas pipe (63 mm outer diameter, 57 mm inner diameter, wall 3 mm) that

was embedded in resin that coated the container floor (Table Top Crystal Clear resin

mix, Fiberglass Coatings, Inc., St Petersburg, FL, USA). White Creatology™ foam

(Wells Fargo Bank, Boston, MA, USA) with adhesive backing was glued to the

underside of the container lid and was used to create a seal with the pipe. A small hole

(0.635 cm) was drilled into the pipes to allow termite and air movement between the

inner and outer chambers. Circular holes approximately 1.59 cm in diameter were

drilled in all of the lids or coverings used in the experiment to allow humidity probe

readings that confirmed that the inner Plexiglas tubes were >90% RH and outer boxes

were <50% RH. The holes were plugged with rubber stoppers when not in use. Relative

humidity and temperature readings were taken using an Amprobe THW3 probe fitted

with a plastic collar to allow for upright free-standing. These readings were only taken

as a means to monitor the RH of the arena and incubator used in the experiment. Two

dry wood blocks (shelters) were placed as food and shelter in the apparatus, one within

the inner chamber, and the other in the outer chamber. Each shelter consisted of two

pieces of stacked pine wood (2.5x2.5x1.5 cm), one piece with a groove cut in it (0.635

115

cm diameter Ryobi straight router bit) (Figure 7-1 D). The cavity of each shelter was

aligned in parallel with the hole in the inner plastic pipe (Figure 7-1).

Approximately 60 grams of sterilized play sand was used as the substrate to

facilitate termite movement around the dual chamber apparatus. The sand was wetted

with approximately 20 mL deionized water until over-saturation. Excess water was then

drained off. After 24 hours, groups of 50 worker termites of the four species were

introduced through the probe access holes to the inner and outer chambers (25 to each

chamber). Initially, the conditions of both the inner and outer chambers were high in RH

and water availability (little to no desiccation) (Figure 7-1A). After 24 hours of

acclimating to this environment, the container lid was removed and replaced with a

circular plastic lid over the inner chamber for C. formosanus and N. jouteli (Figure 7-1B)

and a modified container lid that covered only the outer chamber (foam with backing

was used to create a seal) for Cr. cavifrons and Cr. brevis (Figure 7-1C). A quick and

drastic change to drier, less favorable conditions in the outer chamber followed for C.

formosanus and N. jouteli. This change to drier conditions was only applied to the inner

chamber for the Cryptotermes species. This difference in procedure (Figure 7-2) was

undertaken to make the outer chamber less favorable in the second part of the

experiment, with the assumption that C. formosanus, and N. jouteli require high RH and

moisture availability, and that the Cryptotermes species do not and that such levels may

be deleterious to these drywood species. Behavioral observations were recorded every

24 hours thereafter for four days. Four days was found to be the approximate amount of

time it took a control container’s inner chamber to equilibrate with the external incubator

conditions after the container lid was removed. After four days, the number of surviving

116

termites and whether they were found in the inner or outer chamber was recorded. The

entire experiment lasted five days.

Behavioral observations were based on how the termites reacted to their

environment and the change to this environment. This included whether they used the

chambers and/or wood shelters provided, and whether construction behavior was

exhibited. The hole connecting the inner and outer chambers was examined to see if it

was sealed as well and considered partially sealed if it was at least 50% closed off

through construction behaviors of the termites. A shelter was considered “in use” if

termites were found inside the shelter cavity, or atop the shelter, since the drywood

termites especially seemed to avoid direct contact with the wet sand. A shelter was

considered “sealed” if at least one side of the shelter cavity was closed off through

construction behaviors of the termites, with the number of sealed off sides also

recorded.

Behavioral observations were recorded and compiled. The data were analyzed

using the JMP statistical software. Factorial experimental designs (2x2), with species

and refuge chamber (favorable or unfavorable conditions) as the factors and preference

(as percent found) as the response variable, was used for analysis of the C.

formosanus-N. jouteli couplet and the Cr. cavifrons-Cr. brevis couplet separately.

ANOVAs were conducted to test for differences in preference between the refuge

chambers (as “favorable” and “unfavorable” conditions), respectively), as well as overall

survival of each species. Tukey’s HSD post hoc tests were used to separate the

differences among all pairs. Percent survival data were arcsine-square root transformed

before analysis. Data from six replications were analyzed.

117

Results

Readings taken from the temperature and humidity probe resulted in the average

values for the chambers over five days. In the experiment involving C. formosanus and

N. jouteli, the average readings (±SEM) for the sealed inner chamber were 95.9±0.11%

RH and 26.1±0.03°C, for the sealed outer chamber they were 96.0±0.2% RH and

26.3±0.05°C, and for the incubator/exposed outer chamber they were 41.5±0.46% RH

and 26.4±0.03°C. For the experiment using Cr. cavifrons and Cr. brevis, the average

readings for the sealed inner chamber were 96.1±0.2% RH and 26.2±0.06°C, for the

sealed outer chamber they were 95.0±0.35% RH and 26.2±0.03°C, and for the

incubator/exposed inner chamber they were 41.4±0.51% RH and 26.4±0.04°C.

Observational results were compiled and are presented in Table 7-1. In two of the six

replicates, C. formosanus sealed the hole connecting the chambers. In all of the six

replicates the inner shelter was in use, as compared with the outer shelter which was

only in use in three replicates. Openings to the shelter cavities were sealed by C.

formosanus nine times for the inner shelters, and only four times for the outer shelters in

use. Of the 300 C. formosanus individuals from six trials, 79% survived the change in

environmental conditions (over 99% of those were found in the inner chamber). In all six

replicates, N. jouteli at least partially (≥50%) sealed the hole connecting the chambers

(three times completely). In no replicates did they seal any of the wood shelters

provided, but this was due to their large size and activity, as they knocked the top of the

inner shelter off every time. Of the 300 N. jouteli individuals from six trials, 100%

survived the change in environmental conditions (over 99% of those were found in the

inner chamber). Cr. cavifrons never sealed the hole connecting the chambers. In every

118

replicate the wood shelters provided in each chamber were in use, but the cavity was

never completely sealed (one side in one case), although deposition was found within

some of the cavities. Of the 300 Cr. cavifrons individuals from six trials, 97% survived

the change in environmental conditions (54% in the inner chamber, 43% in the outer

chamber). Cr. brevis also never sealed the hole connecting the chambers. In every

replicate the wood shelters provided in each chamber were in use. Openings to the

shelter cavities were sealed by Cr. brevis seven times for the inner shelters, and only

four times for the outer shelters in use. Of the 300 Cr. brevis individuals from six trials,

97% survived the change in environmental conditions (47% in the inner chamber, 50%

in the outer chamber). Examples of sealed shelters are presented in Figure 7-3. Overall

survival of individuals for the four species is presented in Table 7-2. In terms of survival,

there was a significant difference in survival only between N. jouteli and C. formosanus

when comparing between species in each of the couplets (df=1 F=10.5 P-

value=0.0089).

The ANOVA for refuge chamber preference of the remaining living individuals of

C. formosanus and N. jouteli revealed a significant difference for refuge chamber

conditions only, and no significant differences for the remaining living individuals of the

Cryptotermes species (Table 7-3). Post hoc tests (Table 7-6 and Table 7-7) revealed no

overall difference in species preference for either couplet. There was a significant

difference in overall preference for the conditions of the refuge chambers, with the

favorable inner chamber housing a significantly higher percentage of termites (98%)

than the unfavorable outer chamber (2%) for C. fomosanus and N. jouteli. No difference

was found for these species within refuge chamber conditions, but a significant

119

difference was found between refuge chamber conditions for species (higher preference

for the favorable inner chamber conditions). There was no difference in the overall

preference for the conditions of the refuge chambers for Cr. cavifrons and Cr. brevis (51

and 49%). No difference was found for these species within refuge chamber conditions,

nor between refuge chamber conditions for species.

Discussion

The results of this study support the hypothesis that C. formosanus and N. jouteli

would sequester themselves within refugia and utilize shelters to buffer against a less

favorable environment. This hypothesis was not entirely supported for the Cryptotermes

species, as they did not sequester themselves in the favorable conditions of the inner

chamber, but did utilize the shelters. It is also evident that termites employ behavioral

suites to tolerate drastic changes in environmental conditions, in this case, to prevent

either desiccation or water poisoning. C. formosanus was not able to react to the

change efficiently enough in all cases, losing over 75% of the individuals in one

replicate, and had an overall survival significantly lower than N. jouteli (which lost no

individuals). Sealing of open areas or breaches, while a possible artefact of behavioral

adaptation to prevent predation, also aids in keeping the nest/galleries at homeostasis

with regard to humidity levels. The fact that the termites did not always completely seal

the chambers and shelters seems to indicate predation wasn’t the major reason for

sealing behavior for these groups of termites. The length of the experiment, however,

may have prevented the completion of construction undertaken to avoid predation.

Whether predation or colony homeostasis with respect to humidity (water resources) is

the more important reason for such construction behavior remains to be elucidated.

120

While N. jouteli sealed the hole connecting the chambers at least partially every

time, C. formosanus did so only a third of the time, but tended to seal the shelters

provided instead (losing about 30% of its individuals to desiccation). The vigorous

movement and size of N. jouteli rendered the inner shelters askew, effectively

eliminating them from consideration as a refuge. It is likely, based on behavior observed

on colonies kept in the laboratory, that they would have utilized such cavities if they

were larger or the shelter made of a single piece of wood. A future study should include

shelters large enough and made of a single piece of wood.

In both “wetwood” species, the majority of living termites were found in the inner

chamber, again exhibiting a preference for environments with high RHs. While C.

formosanus used wood particles from the shelters to seal the cavities as well as the

hole connecting chambers, N. jouteli used feces (or a wood/feces mixture) to seal this

hole. This is probably due to C. formosanus sequestering water resources internally at

the onset of unfavorably dry conditions, and N. jouteli using body water in its feces to

buffer against those conditions.

Whereas, N. jouteli and C. formosanus would seal the inner chamber, and stayed

in or relocated to where RH was high, neither of the Cryptotermes species did. They

always used the wooden shelters provided, however, cramming themselves into the

cavities and sealing one side of the cavity on most occasions (both sides in a few

cases), or huddling on the topside of the shelters. In both Cryptotermes species the

proportion of living termites was at 97% and was almost evenly split between the inner

and outer chamber. This indicated that these two species are able to tolerate both dry

121

and wet environments for a time, as long as they do not experience prolonged contact

with liquid water or heavy moisture.

Both Cr. cavifrons and Cr. brevis used feces to seal the shelters. In 7 of the 11

cases of shelter-sealing behavior exhibited by Cr. brevis, the inner shelter was the

shelter that was sealed. This showed that despite not indicating a statistical difference in

preference for the conditions of the chambers, individuals of this drywood species

exhibited a greater incidence of construction behavior in the drier conditions of the two

chambers. Cr. cavifrons did not exhibit cavity sealing to the extent that Cr. brevis did (1

versus 11 times), but the sole occurrence was found in the inner chamber. This does

not support the assumption that the dry conditions of the inner chamber were favorable

for the Cryptotermes species (especially Cr. brevis). In the outer chamber where RH

was high, the Cryptotermes species would not only utilize the shelter cavity, but would

also use the top of the shelter, presumably to avoid contact with the wet sand and

possibly to avoid the wet wood of the shelter as it absorbed moisture from the sand.

This again illustrates an avoidance by these species of direct contact with moisture. The

fact that such a high proportion of individuals survived in both of the chambers means

that the Cryptotermes species were able to tolerate high and low moisture levels of the

air and substrate through behaviors different to those exhibited by C. formosanus and

N. jouteli.

The sealing, huddling, and avoidance behaviors seen in the four termite species

were important because they provided groups of termites with refuge from and

tolerance of negative environmental conditions (either xeric or hygric conditions),

keeping the group alive as long as possible until conditions became more favorable or

122

the group collapsed. This is what we would expect to happen in natural environments as

conditions change for a multitude of reasons and for various periods of time, forcing

termite colonies to address such changes as efficiently as possible. It may not be

possible for groups of termites to efficiently relocate horizontally across a foraging area,

or vertically within the substrate, so it is important that they adapt their behavior to

tolerate less favorable conditions. This study was short in duration and an increase in

the time period is prudent in order to examine whether termite groups will relocate if

unfavorable conditions are prolonged, if they tend to “wait-out” less favorable conditions

rather than expose themselves to additional biotic and abiotic difficulties, and if

behavioral suites change as the group becomes more acclimated to a given condition.

123

Table 7-1. Shelter study data compilation from 6 replicates with 300 termites total

Species Chamber

seal (partial seal)a

Inner shelter in

use (seal)b

Outer shelter in

use (seal)b

Alive inner chamber

(percentage)c

Alive outer chamber

(percentage)c

C. formosanus 2 (0) 6 (9) 3 (4) 234 (78) 3 (<1)

N. jouteli 3 (3) ---d 0 (0) 299 (>99) 1 (<1)

Cr. cavifrons 0 (0) 6 (1) 6 (0) 161 (54) 129 (43)

Cr. brevis 0 (0) 6 (7) 6 (4) 142 (47) 149 (50) a hole connecting inner and outer chambers closed off through construction behavior (hole at least 50% closed)

b termites atop shelter or in cavity (number of shelter cavity openings closed off through construction behavior)

c number of termites alive (percentage of total termites alive)

d termite activity rendered shelters unusable

124

Table 7-2. Termite survival (%) in a dual chamber arena (mean±SEM)

Speciesa Observations Survival (%)b

N. jouteli 6 100±0.0a

C. formosanus 6 79.0±11.4b

Cr. brevis 6 97.0±1.4A

Cr. cavifrons 6 96.7±1.6A a N. jouteli/C. formosanus and Cr. brevis/Cr. cavifrons couplets were analyzed separately.

b Values followed by the same uppercase or lowercase letter are not significantly different at ɑ=0.05 (Student's t).

125

Table 7-3. Analysis of variance for a factorial experiment to evaluate preference of surviving termites after 4 days in a dual chamber arena

Species couplet Source df Termite preferencea

MS b F P-value

C. formosanus / N. jouteli

Conditions (C)

1 12.61 771.7 <.0001

Species (S) 1 0 0 1

CxS 1 2.01 2 0.1739

Error 20 0.02 --- ---

Cr. cavifrons / Cr. brevis

Conditions (C)

1 0.01 0.4 0.5361

Species (S) 1 0 0 1

CxS 1 0.03 1 0.3249

Error 20 0.02 --- ---


b Mean square.

126

Table 7-4. Preference of live individuals (%) of four termite species for two refuge chambers (mean±SEM)

Refuge chamberabc

Favorable conditions/inner chamber

Unfavorable conditions/outer chamber

(95.9±0.11%, 26.1±0.03°C) (41.5±0.46%, 26.4±0.03°C) Overall mean

preference for speciese Species coupletd [41.4±0.51%, 26.4±0.04°C] [95.0±0.35%, 26.2±0.03°C]

C. formosanus 96.7±2.8Aa 3.3±2.8Ab 50.0±14.2A

N. jouteli 99.7±0.33Aa 0.30±0.33Ab 50.0±15.0A Overall mean preference for refuge chambere

98.2±1.4a 1.8±1.4b ---

Cr. cavifrons 54.0±7.7Aa 46.0±7.7Aa 50.0±5.3A

Cr. brevis 48.8±1.5Aa 51.2±1.5Aa 50.0±1.1A Overall mean preference for refuge chambere

51.4±3.8a 48.6±3.8a ---

a Mean RH±SEM and temperature±SEM associated with each refuge chamber for C. formosanus/N. jouteli found in parentheses and Cr. brevis/Cr. cavifrons in brackets.

b Means for each combination of treatment (species x refuge chamber) followed by the same lowercase letters within a row or means followed by the same capital letter within a column are not significantly different at the ɑ=0.05 level (Tukey's HSD).

c Value for the combination (species x refuge chamber) of treatments are means of 6 observations.

dC. formosanus/N. jouteli and Cr. cavifrons/Cr. brevis couplets were analyzed separately.

e Values for the main treatment effects (species and refuge chamber) are means of 12 observations. Means followed by the same letter are not significantly different at the ɑ=0.05 level (Tukey's HSD).

127

Figure 7-1. Setup for examining the use of refuge and shelter, as well as associated behaviors by four species of termite when exposed to environmental change. A) Initial high RH and water availability arena. B) Arena with outer chamber exposed to incubator conditions for C. formosanus and N. jouteli. C) Arena with inner chamber exposed to incubator conditions for Cryptotermes spp. D) Close-up of inner/outer wood shelters showing views of sides and top. a: temperature/humidity probe b: outer shelter c: inner shelter d: inner chamber e: rubber stopper f: inner chamber lid g: inner chamber access hole h: foam seal i: modified lid foam seal j: sand layer of outer chamber k: initial arena lid l: modified outer chamber lid. Photograph courtesy of author.

128

Figure 7-2. Diagram illustrating dual chamber shelter study protocols for C.

formosanus/N. jouteli couplet and Cr. brevis/Cr. cavifrons couplet. Gray shapes indicate bottom section of arenas (dashed gray circles where the inner chamber tube is covered). Black shapes indicate lids of the arenas. Green arrows represent lids being added, red arrows lids being removed. Black arrows with text show the time the arena is in a particular conformation. Black “shift” arrow indicates when the arena has gone from being fully covered with initial lid, to partially covered (uncovered portions exposed to the conditions of the incubator).

129

Figure 7-3. Examples of shelter and chamber sealing. A) Shelter seal by C. formosanus, B) by Cr. brevis, C) by Cr. cavifrons and D) Partial chamber seal by N. jouteli. Photographs courtesy of author.

130

CHAPTER 8 CONCLUSIONS

The preceding project was motivated by the author’s interest in tolerance to

desiccation in insects known to be prone to problems associated with loss of body water

and their management of the water sources they may be exposed to. If we return to the

water budget model presented in Chapter 1 (Figure 1-2), we can now piece the previous

studies into a cohesive picture of termite water management. The dependence of all life

on water is reflected in the various lifestyles and habitats of termites. No matter if they

are found in xeric, mesic, or hydric environments, individuals of a termite colony must

have enough water to prevent their own desiccation, as well as the collapse of the

colony from the loss of too many individuals. The amount of water held and its retention

by an individual termite of a given species and a given colony varies. This is mainly due

to the variable sizes of individuals within a population. Individual termites of drywood

species such as Cr. brevis are better capable of retaining this body water, preventing its

evaporation through behavioral and physiological means. Other species, such as C.

formosanus, do not have a strong capacity to prevent water from evaporating from the

cuticle and must use their own behavioral and physiological means to counter this. For

a small group of termites, individual behavioral and physiological means benefit the

group as well as behavioral mechanisms exhibited by the group itself. Lastly, when the

group expands from having a few individuals to include many individuals (i.e. a colony),

additional mechanisms become necessary to protect this large number of individuals

from desiccation stresses. While sacrifice of some individuals may be necessary in

extreme conditions, general preventative and remedial measures include regulation and

repair of the construction of the nest and foraging galleries, movement of water

131

resources, and the division of labor that contributes to the eusociality of the colony. The

overall hypothesis that termites must address paramount water management concerns

based on the changes in their environment both as individuals and as a group (colony)

or risk the death of enough individuals to incur colony collapse is supported, and lends

itself to the idea that water management in termites can be characterized as having

three levels: individual, small group, and large group (colony).

At the individual level, water resources are located and taken in by termites that

make up a colony, but a portion of these resources is lost to evaporation and wastes

(Figure 1-2, water resource, water loss, and waste arrows). The water resource that

termites consume may be bound to a substrate, such as soil or a food, but may also be

free water that can be imbibed and stored in reservoirs (labial glands). Smaller amounts

of water may be obtained during trophallaxis and through cannibalism, interactions that

occur mainly between two individual termites (Figure 1-2, A). This water must then be

used efficiently to keep the individuals alive so they can properly develop and perform

their tasks as a specific caste in the colony. These tasks include foraging for food and

water, as well as the management of both. At the small group and large group levels

(Figure 1-2, C), division of labor leads to a social homeostasis in terms of water

management that must be regulated to prevent desiccation and colony collapse.

Regulation occurs through behavior and physiological mechanisms. Behavioral

mechanisms include cannibalism when water resources are low or absent, huddling to

decrease surface area available for evaporation, and water transport for the

modification of their immediate environment and the environment of the nest, foraging

tunnels, and galleries (Figure 1-2, B and C). Water lost from the body through the

132

cuticle and in wastes is recycled as vapor for RH equilibria in a confined space (Figure

1-2, reuse arrows). This is reflected in the use of moisture movement, fecal deposition,

and the construction of carton by some species, which aid in the stabilization of the RH

of their microhabitat. Other species hold on to their body water, using physiology to

prevent its escape through the cuticle, spiracles, and/or waste products. Thus, each

individual of a group contributes to the overall management of the colony’s water

resources (actively or passively), as do the groups themselves, taking on certain

behaviors as is required by the colony (i.e. behavioral castes).

We can see how important water resources are to the survival of termites, as

they will die from desiccation far earlier than from starvation (Collins 1991). Studies by

Collins (1969) indicated that different species of termites lived from minutes to days

when placed in extremely less-than-favorable conditions, indicating drastically different

abilities to tolerate desiccation and temperature stresses. In general, it was evident that

the “wet wood” species (i.e. C. formosanus and N. jouteli) require access to water and

high humidity environments at all times to function efficiently as is seen in the

desiccation chamber (Table 3-2), RH equilibria (Table 4-5 and Figure 4-2), RH

preference (Table 5-2), water availability (Table 6-2 to 6-5), and shelter studies (Table

7-2 and 7-4) conducted herein. The subterranean termite, C. formosanus, does not

tolerate low %RH conditions well, and must employ behavioral and physiological

measures to resist desiccation. They are not as adept at retaining body water as N.

jouteli, easily losing it through their thinner cuticle and in wastes from thinner longer

rectal pads (Figure 4-4, Table 2-1, and 2-2). In fact, body water from a small group of C.

formosanus (25 individuals) was seen to increase the RH in their immediate

133

environment towards a high %RH equilibrium in a confined space, indicating that body

water is a resource available for affecting the RH level of their microhabitat and a

component of resistance to desiccation (Table 4-5, Figure 4-2, and Figure 4-4). The

dampwood termite, N. jouteli, is able to tolerate desiccation better than C. formosanus,

in part because of their size, as well as the amount of total body water available to be

lost (Table 4-1). Body water of N. jouteli also increased the immediate environment

towards a high %RH equilibrium of a confined space. Shorter, thicker (larger) rectal

pads that can remove a high proportion of water from waste (pelleted frass) and a thick

cuticle that reduces the amount of water evaporated from the body are also factors in

this tolerance (Table 2-1 and 2-2). It is unclear whether differences in the morphology of

the spiracles of N. jouteli when compared to C. formosanus are also a factor in

desiccation tolerance (Figure 2-4 A and B). Both species utilize a microhabitat high in

%RH. However, whereas smaller colonies of N. jouteli find and live within its moist food

source, larger colonies of C. formosanus find and live in a microhabitat that they can

modify with water resources they find, as well as with available body water though

buccal and fecal deposition as needed. The rectal pads of C. formosanus do not retain

waste water to the degree that N. jouteli, Cr. cavifrons, and Cr. brevis can, aiding in the

regulation of more liquid feces as opposed to the pelleted frass of the latter species. N.

jouteli was also observed using fecal deposition to modify their environment in

laboratory conditions, but may or may not need to in their natural environments.

The drywood species both have small, slow-developing colonies found in low

%RH environments. Whereas Cr. cavifrons is found in natural wooded habitats, Cr.

brevis is found in the dry lumber of man-made structures. They do not require the level

134

of access to water and high humidity environments that the “wetwood” species do, as is

seen in the desiccation chamber (Table 3-2), RH equilibria (Figure 4-4 and 4-6), RH

preference (Table 5-2), water availability (Table 6-2 to 6-5), and shelter studies (Table

7-2 and 7-4). In fact, the evidence supports the idea that these species are not adept at

dealing with conditions where water is abundant, either as humidity or as free water,

and strive to avoid contact if possible (Table 6-2 to 6-5). However, using body water, Cr.

cavifrons and Cr. brevis did not increase the RH to a level significantly higher than the

control (Table 4-4 and Figure 4-2). Thus, this humidity level was far lower than the

humidity levels achieved by both C. formosanus and N. jouteli. As their common name

suggests, these drywood termites are also much more capable of tolerating low %RH

conditions than either of the “wetwoods.” Large rectal pads that can remove a high

proportion of water from waste (pelleted frass) and thick cuticles that reduce the amount

of water evaporated from the body are also factors in this tolerance (Table 2-1 and 2-2).

However, both Cr. brevis and Cr. cavifrons were observed using liquid feces to seal

cavities in order to create refuges from less-than-favorable conditions (Figure 2-3). It is

also unclear whether differences in the morphology of the spiracles of the drywoods

(and N. jouteli) when compared to C. formosanus are also a factor in desiccation

tolerance (Figure 2-4 A, C, and D).

Given the results presented in the previous chapters, we can see how important

water resources are to individual termites, as well as an entire colony. In general,

absence of water and/or food sources will lead to death, but death from starvation takes

far longer than death from desiccation. The lack or loss of a food source is problematic,

but easier for a colony to tolerate because starvation generally takes such an extended

135

period of time. Food sources, which cannot move, will not be lost to the conditions of

their environment, only depleted as it is consumed. As the source is depleted and

eventually exhausted, foragers continue to search for replacements (or supplements).

Loss (evaporation) of water resources, however, is more difficult to manage because

once the source is lost from the microhabitat it cannot be recouped, only replaced.

Problems stemming from desiccation stresses are relatively quick, requiring rapid,

efficient location of additional water resources. Water lost to evaporation from the body,

however, can be reused as vapor or liquid if condensation is possible. This can be

accomplished in some situations (e.g. small tunnels with low air movement), but the

dynamic nature of the environmental conditions of the general habitat can make this

difficult in the long run. Termites cannot obtain the necessary levels of water for the

colony from the ambient RH of their habitat, forcing foragers to search for additional or

supplemental water sources if unfavorable water availability conditions arise. A

prolonged lack of adequate water resources or a decrease in the ambient RH of the

microhabitat causes individuals to die of desiccation, which disrupts the social

homeostasis of the colony. If this is not compensated for, the colony risks collapse.

Without enough water to allow the body (individual or group) to function correctly,

finding food doesn’t matter. Some species are better equipped to tolerate a lack of

water for a longer period of time, but eventually must find it in one form or another.

Starvation takes its toll within a longer time frame when there is a lack of adequate food

resources for the colony. Depending on the termite species, the death of individuals can

be a matter of hours for desiccation and a matter of days or weeks for starvation.

136

The objectives set out for this project were generally achieved for each chapter,

providing information on the requirements for adequate and efficient management of

water by termites to avoid desiccation and provide a microhabitat equilibrium in terms of

moisture and humidity levels. The information gleaned from this study is unlikely to aid

directly in control of economically important termites. However, it does highlight the

importance of homeowner vigilance in terms of preventing conditions in and around the

home that provide termites with a favorable environment for establishment whether they

be economically important or nuisance pests. It is interesting to consider that average

humidity and dew point levels in conjunction with average temperature and stability of

habitat, may be a factor in the distribution of termites. For instance, these factors could

be a reason Cr. brevis is such a widespread pest species, and yet, has not been found

in Southeast Asia (Evans et al. 2013). Similarly, are these factors the reason the

biogeographical range of C. formosanus is absent west of Texas? These are likely not

the only reasons for such distribution patterns, however. Pathogens, predators, and

parasites are additional factors to consider. It is interesting to consider that Cr. brevis,

though originally found in naturally occurring dry wood, entered into a pseudo-

commensal relationship with humans (and possibly an allopatric speciation) with the

onset of a more globalized economy utilizing wooden ships and worldwide shipping of

palleted goods. The primary detection of major pest insect species (such as C.

formosanus and Cr. bevis) and our ability to track their expansion, is key to our

understanding of how these species are distributed around the globe and how they are

aided or deterred by human activities. Additional research into these aspects of pest

establishment and invasion is certainly warranted.

137

Summary: 1) Cuticle thickness, rectal pad proportion, and spiracle morphology are factors in

the ability of individuals of the four termite species to prevent and tolerate desiccation.

2) All four species exhibit RH equilibria, an alternative to cuticular permeability values in indicating differences in rates of body water loss through the cuticle. Higher RH levels were observed for groups of dead termites in comparison with groups of living termites.

3) N. jouteli and C. formosanus exhibit not only a preference, but the requirement for high RH environments and/or readily available water resources, though the latter can tolerate desiccation better than the former.

4) Cr. brevis exhibits a preference for lower RH environments reflecting the conditions of its human-mediated habitat, but can tolerate higher RH environments for some time as long as direct contact with water is avoided.

5) Cr. cavifrons exhibits a preference for a range of RH environments reflecting the conditions of its natural habitat in which RH levels are more variable and contact with water is more likely than in the habitat of Cr. brevis. They must also avoid direct contact with water when possible.

6) All four species employ construction behavior to tolerate less-than-favorable conditions and avoid desiccation. Aggregation (huddling) is another important factor in tolerating these desiccating conditions.

138

LIST OF REFERENCES

Abe, T., D.E. Bignell, and M. Higashi, (Eds.) 2000. Termites: evolution, sociality, symbioses, ecology. Dordrecht: Kluwer Academic Publishers, 488 pp.

Abe, T., N. Kirtibutr, and J.A. Holt, (Eds.) 1997. Global diversification of termites—its pattern and causal mechanism. Kyoto: Japanese Ministry of Education, Science, Sports, and Culture, iv + 377 pp.

Abe, T. 1987. Evolution of life types in termites. In: Kawano, S., Connell, J.H., and Hidaka, T. (Eds), Evolution and Coadaptation in Biotic Communities. University of Tokyo Press, Tokyo. pp 125-128.

Abushama, F.T. 1974. Water-relations of the termites Macrotermes bellicosus (Smeathman) and Trinervitermes geminatus (Wasmann). Z. Angew. Entomol. 75, 124-134.

Ahmad, M., M. Afzal, and Z. Salihah. 1982. The effects of different relative humidities on survival and moisture loss of workers and soldiers of Heterotermes indicola (Wasmann) (Isoptera: Rhinotermitidiae) under starvation conditions. Pakistan J. Zool. 14, 65-70.

Altner, H., H. Sass, and I. Altner. 1977. Relationship between structure and function of antennal chemo-, hygro-, and thermoreceptive sensilla in Periplaneta americana. Cell Tissue Res. 176, 389-405.

Appel, A.G., D.A. Reierson, and M.K. Rust. 1983. Comparative water relations and temperature sensitivity of cockroaches. Comp. Biochem. Physiol. 74A, 357-361.

Appel, A.G., D.A. Reierson, and M.K. Rust. 1986. Cuticular water loss in the Smokybrown cockroach, Periplaneta fuliginosa. J. Insect Physiol. 32, 623-628.

Arab, A., and A.M. Costa-Leonardo. 2005. Effect of biotic and abiotic factors on the tunneling behavior of Coptotermes gestroi and Heterotermes tenuis (Isoptera: Rhinotermitidae). Behav. Process 70, 32-40.

Arquette, T.J. 2013. Lipid, glycogen, proteins, uric acid, and body water in termites and other insects. Journal of the Mississippi Academy of Sciences. 58, 207-213.

Beament, J.W.L. 1945. The cuticular lipoids of insects. J. Exp. Biol. 21, 115-131.

Beament, J.W.L. 1961. The water relations of insect cuticle. Biol. Rev. 36, 281-320.

Bignell, D.E. 2000. Introduction to symbiosis. In Abe, T., Bignell, T.E., and Higashi, M., (Eds.), Termites: evolution, sociality, symbioses, ecology: 189–208. Dordrecht: Kluwer Academic Publishers, xxii + 466 pp.

139

Bignell, D.E., and P. Eggleton. 2000. Termites in ecosystems. In Abe, T., Bignell, T.E., and Higashi, M., (Eds.), Termites: evolution, sociality, symbioses, ecology, 363–387. Dordrecht: Kluwer Academic Publishers, xxii + 466 pp.

Brammer, A.S., and R.H. Scheffrahn. 2007. A drywood termite. http://entomology.ifas.ufl.edu/creatures/urban/termites/c_cavifrons.htm

Cabrera, B.J., and M.K. Rust. 1996, Effect of temperature and relative humidity on aggregation and water loss in the Western drywood termite, Incisitermes minor (Hagen). Proc. Nat. Conf. Urban Entomol. 39-41.

Chouvenc, T., N.-Y. Su, and A. Robert. 2009. Cellular encapsulation in the eastern subterranean termite, Reticulitermes flavipes (Isoptera), against infection by the entomopathogenic fungus Metarhizium anisopliae. J. Invertebr. Pathol. 101: 234–241.

Chouvenc, T., C.A. Efstathion, M.L. Elliott, and N.-Y. Su. 2013. Extended disease resistance emerging from the faecal nest of a subterranean termite. Proc. R. Soc. B. 280: 20131885.

Collins, M. 1958. Studies on water relations in Florida termites. I. Survival time and rate of water loss during drying. Quart. J. Fla. Acad. Sci. 21, 341-352.

Collins, M. 1963. Studies on water relations in North American termites I. Eastern species of the genus Reticulitermes (Isoptera, Rhinotermitidae). Ecology 44, 600-604.

Collins, M. 1966. Studies on water relations in North American termites II. Water loss and cuticular structures in eastern species of the Kalotermitidae (Isoptera). Ecology 47, 328-331.

Collins, M. 1969. Water relations in termites. In: Krishna, K., Weesner, F.M. (Eds), Biology of Termites, Vol. 1. Academic Press, NY. pp 433-458.

Collins, M. 1991. Physical Factors Affecting Termite Distribution. Sociobiology 19, 283-286.

Collins, M. 1997. The termites (Isoptera: Kalotermitidae, Rhinotermitidae, Termitidae) of the British Virgin Islands: Distribution, moisture relations, and cuticular hydrocarbons. Sociobiology 30, 63-76.

Cook, S.F., and K.G. Scott. 1932. The relation between absorption and elimination of water by Termopsis angusticollis. Biol. Bull. (MBL) 63, 505-512.

Cornelius, M.L., and W.L.A. Osbrink. 2010a. Effect of flooding on the survival of Formosan subterranean termites (Isoptera: Rhinotermitidae) in laboratory tests. Sociobiology 56(3): 699-711.

140

Cornelius, M.L., and W.L.A. Osbrink. 2010b. Effect of soil type and moisture availability on the foraging behavior of the Formosan subterranean termite (Isoptera: Rhinotermitidae). J. Econ. Entomol. 103, 799-807.

Delaplane, K.S., and J.P. La Fage. 1989. Preference for moist wood by the Formosan subterranean termite (Isoptera: Rhinotermitidae). J. Econ. Entomol. 82, 95-100.

Ebeling, W., and R.E. Wagner. 1959. Rapid desiccation of drywood termites with inert sorptive dusts and other substances. J. Econ. Entomol. 52, 190-207.

Edney, E.B. 1977. Water balance in land arthropods. Zoophysiology and ecology series. Vol. 9. Springer, Berlin. Pp. 282.

Edney, E.B., and J.M. McFarlane. 1974. The effect of temperature and transpiration in the desert cockroach, Arenivaga investigata and in Periplaneta americana. Physiol. Zool. 47, 1-12.

Edney, E.B., 1957. The water relations of terrestrial arthropods. Cambridge monographs in experimental biology. No. 5. Cambridge University Press, Cambridge, UK. Pp. 105.

Emerson, A.E. 1956. Regenerative behavior and social homeostasis of termites. Ecology. 37, 248-258.

Evans, T.A., B.T. Forschler, and J.K. Grace. 2013. Biology of invasive termites: A worldwide review. Annu. Rev. Entomol. 58, 455-474.

Forschler, B.T., and G. Henderson. 1995. Subterranean termite behavioral reaction to water and survival of inundation: Implications for field populations. Environ. Entomol. 24, 1592-1597.

Gallagher, N.T., and S.C. Jones. 2010. Moisture augmentation of food items by Reticulitermes flavipes (Isoptera: Rhinotermitidae). Sociobiology 55, 735-747.

Gautam, B.K., and G. Henderson. 2011. Relative humidity preferences and survival of starved Formosan subterranean termites (Isoptera: Rhinotermitidae) at various temperature and relative humidity conditions. Environ. Entomol. 40, 1232-1238.

Gautam, B.K., and G. Henderson. 2014. Water transport by Coptotermes formosanus (Isoptera: Rhinotermitidae). Environ. Entomol. 43, 1399-1405.

Grasse, P.P., and R. Chauvin. 1944. L’effect de group et la survie des neutress dans les societies d’Insectes. Rev. Sci. 82, 461-464.

Green, J.M., M.E. Scharf, and G.W. Bennett. 2005. Impacts of soil moisture level on consumption and movement of three sympatric subterranean termites (Isoptera: Rhinotermitidae) in a laboratory assay. J. Econ. Entomol. 98, 933-937.

141

Grube, S., and D. Rudolph. 1999a. The labial gland reservoirs (water sacs) in Reticulitermes santonensis (Isoptera: Rhinotermitidae): Studies of the functional aspects during microclimatic moisture regulation and individual water balance. Sociobiology. 33, 307-323.

Grube, S., and D. Rudolph. 1999b. Water supply during building activities in the subterranean termite Reticulitermes santonensis De Feytaud (Isoptera, Rhinotermitidae). Insectes Soc. 46, 192-193.

Grube, S., D. Rudolph, and I. Zerbst-Boroffka. 1997. Morphology, fine structure, and functional aspects of the labial gland reservoirs of the subterranean termite Reticulitermes santonensis de Feytaud (Isoptera: Rhinotermitidae). Int. J. Insect Morphol. & Embryol. 26, 49-53.

Hadley, N.F. 1977. Epicuticular lipids of the desert tenebrionid beetle, Eleodes armata: seasonal and acclamatory effects on chemical composition. Insect Biochem. 7, 277-283.

Hadley, N.F. 1978. Cuticular permeability of desert tenebrionid beetles: correlations with epicuticular hydrocarbon composition. Insect Biochem. 8, 17-22.

Hinton, H.E. 1960. A fly larva that tolerates dehydration and temperatures of -270°C to +102°C. Nature 188, 336–337.

Hu, J., K.-B. Neoh, A.G. Appel, and C.-Y. Lee. 2012. Subterranean termite open-air foraging and tolerance to desiccation: Comparative water relation of two sympatric Macrotermes spp. (Blattodea: Termitidae). Comp. Biochem. Physiol. A 161, 201-207.

Inward, D., G. Beccaloni, and P. Eggleton. 2007. Death of an order: a comprehensive molecular phylogenetic study confirms that termites are eusocial cockroaches. Biology Letters 3, 331–335.

Khan, M.A. 1980. Effect of relative humidity on survival of termites under starvation conditions. Z. Angew. Zool. 67, 133-178.

Klass, K.-D., C. Nalepa, and N. Lo. 2008. Wood-feeding cockroaches as models for termite evolution (Insecta: Dictyoptera): Cryptocercus vs Parasphaeria boleiriana. Molecular Phylogenetics and Evolution 46, 809–817.

Krishna, K., D.A. Grimaldi, V. Krishna, and M.S. Engel. 2013. Treatise on the Isoptera of the world. Bull. Nat. Hist. Mus. 377, 2704 pp.

Kulis, J., A.S. Sajap, and C.Y. Loong. 2008. Effect of moisture and relative humidity on survival and feeding activity of the Asian subterranean termite, Coptotermes gestroi (Isoptera: Rhinotermitidae). Sociobiology 52: 579-587.

Lee, K.E., and Wood, T.G. 1971. Termites and Soils. Academic Press. NY.

http://www.nature.com/nature/journal/v188/n4747/abs/188336a0.html

http://www.nature.com/nature/journal/v188/n4747/abs/188336a0.html

http://en.wikipedia.org/wiki/Nature_(journal)

142

Locke, M. 1965. Permeability of the insect cuticle to water and lipids. Science. 147, 295-298.

Loftus, R. 1976. Temperature-dependent dry receptor on antenna of Periplaneta. Tonic Response. J. Comp. Physiol. 111, 153-170.

Machin, J. 1981. Water compartmentalization in insects. J. Exp. Zool. 215, 327-333.

Malik, S.A., and K.H. Sheikh. 1990. Effect of different relative humidities on survival and moisture loss of termite workers and soldiers. Pakistan J. Zool. 22, 159-165.

Martoja, R., and M. Martoja-Pierson. 1967. Initiation aux technique de 1’ histologie animale. Masson, Paris.

Mayer, R. T., S.M. Meola, D.L. Coppage and J.R. Deloach. 1979. The pupal instar of Stomoxys calcitrans cuticle deposition and chitin synthesis. J. Insect Physiol. 25: 677-683

McManamy, K., P.G. Koehler, D.D. Branscome, and R.M. Pereira. 2008. Wood moisture content affects the survival of Eastern subterranean termites (Isoptera: Rhinotermitidae), under saturated relative humidity conditions. Sociobiology 52, 145-156.

Mead-Briggs, A.R. 1956. The effect of temperature upon the permeability to water of arthropod cuticles. J. Exp. Biol. 33, 737-749.

Messenger, M.T., N.Y. Su, and R.H. Scheffrahn. 2002. Current distribution of the Formosan subterranean termite and other termite species (Isoptera: Rhinotermitidae, Kalotermitidae) in Louisiana. Fla. Entomol. 85, 580-587.

Minnick, D.R., S.H. Kerr, and R.C. Wilkinson. 1973. Humidity behavior of the drywood termite Cryptotermes brevis. Environ. Entomol. 2, 597-601.

Monzer, M.A., and H.A. Srour. 2009. Desiccation intolerance of the red palm weevil, Rhynchophorus ferrugineus (Oliv) adults in relation to their cuticular hydrocarbons. Egypt. Acad. J. Biolog. Sci. 2, 47-53.

Nakayama, T., T. Yoshimura, and Y. Imamura. 2004a. Effects of the desiccation process on survival and food consumption of Japanese subterranean termites (Isoptera; Rhinotermitidae), Reticulitermes speratus and Coptotermes formosanus. Sociobiology 44, 127-138.

Nakayama, T., T. Yoshimura, and Y. Imamura. 2004b. Effects of desiccation treatment on Japanese subterranean termites, Reticulitermes speratus and Coptotermes formosanus (Isoptera: Rhinotermitidae). Sociobiology 43, 477-487.

Nakayama, T., T. Yoshimura, Y. Imamura. 2004c. Recovery of Japanese subterranean termites, Reticulitermes speratus and Coptotermes formosanus

143

(Isoptera: Rhinotermitidae), from desiccation treatment. Sociobiology 43, 489-499.

Nakayama, T., T. Yoshimura, and Y. Imamura. 2005. Feeding activities of Coptotermes formosanus Shiraki and Reticulitermes speratus (Kolbe) as affected by moisture content of wood. J. Wood Sci. 51, 60-65.

Nation, J. 2002. Insect Physiology and Biochemistry. CRC Press. Washington, DC.

Nishino, H., S. Yamashita, Y. Yamazaki, M. Nishikawa, F. Yokohari, and M. Mizunami. 2003. Projection neurons originating from thermo- and hygrosensory glomeruli in the antennal lobe of the cockroach. J. Comp. Neurol. 455, 40-55.

Noirot, C. 1970. The nests of termites. In: Krishna, K., and Weesner, F.M. (Eds.), Biology of Termites, vol. 2. Academic Press, New York, pp. 73-125.

Noirot, C., and C. Noirot-Timothée. 1977. Fine structure of the rectum in termites

(Isoptera): A comparative study. Tissue & Cell 9, 693-710.

Owens, C.B., N.-Y. Su, C. Husseneder, C. Riegel, and K.S. Brown. 2012. Molecular genetic evidence of Formosan subterranean termite (Isoptera: Rhinotermitidae) colony survivorship after prolonged inundation. J. Econ. Entomol. 105, 518-522.

Pence, R.J. 1956. The tolerance of the drywood termite, Kalotermes minor Hagen, to desiccation. J. Econ. Entomol. 49, 553-554.

Rockland, L.B. 1960. Saturated salt solutions for static control of relative humidity between 5° and 40°C. Anal. Chem. 32, 1375-1376.

Roth, L.M., and E.R. Willis. 1952. Possible hygroreceptors in Aedes aegypti (L.) and Blattella germanica (L.). J. Morphol. 91, 1-14.

Rudolph, D., B. Glocke, and S. Rathenow. 1990. On the role of different humidity parameters for the survival, distribution and ecology of various termite species. Sociobiology 17, 129-140.

Rust, M. K., and N.-Y. Su. 2012. Managing social insects of urban importance. Annu. Rev. Entomol. 57: 355-375.

Rust, M.K., D.A. Reierson, and R.H. Scheffrahn. 1979. Comparative habits, utilization and xeric adaptations of the Southwestern drywood termites, Incisitermes fruticavus Rust and Incisitermes minor (Hagen) (Isoptera: Kalotermitidae). Sociobiology. 4, 239-255.

Scheffrahn, R.H., J. Křeček, R. Ripa, and P. Luppichini. 2009. Endemic origin and vast anthropogenic dispersal of the West Indian drywood termite. Biol. Invasions 11, 787-799.

144

Scheffrahn, R.H., J. Křeček, J.A. Chase, B. Maharajh, and J.R. Mangold. 2006. Taxonomy, Biogeography, and Notes on Termites (Isoptera: Kalotermitidae, Rhinotermitidae, Termitidae) of the Bahamas and Turks and Caicos Islands. Annals of the Entomological Society of America. 99, 463-486.

Scheffrahn, R.H., N.-Y. Su, J.A. Chase, J.R. Mangold, J.K. Grace, and J.R. Yates III. 2000. First record of Cryptotermes cynocephalus Light (Isoptera: Kalotermitidae) and natural woodland infestations of C. brevis (Walker) on Oahu, Hawaiian Islands. Proc. Hawaiian Entomol. Soc. 34, 121-125.

Scheffrahn, R.H., and N.-Y. Su. 2007a. Florida dampwood termites. http://entomology.ifas.ufl.edu/creatures/urban/termites/neotermes.html

Scheffrahn, R.H., and N.-Y. Su. 2007b. West Indian drywood termite. http://entomology.ifas.ufl.edu/creatures/urban/termites/west_indian_drywood_termite.htm

Sen-Sarma, P.K., and P.N. Chatterjee. 1966. Humidity behavior of termites. I. Effect of relative humidity on the longevity of workers of Microcerotermes beesoni Snyder (Insecta: Isoptera: Termitidae) under starvation condition. Forest Bulletin. Forest Research Institute, Dehra Dun, pp 1-6.

Shelton, T.G., and A.G. Appel. 2000. Cyclic CO2 release and water loss in the western drywood termite (Isoptera: Kalotermitidae). Ann. Entomol. Soc. Am. 93, 1300-1307.

Shelton, T.G., and A.G. Appel. 2001. Cyclic CO2 release and water loss in alates of the Eastern subterranean termite (Isoptera: Rhinotermitidae). Ann. Entomol. Soc. Am. 94, 420-426.

Shelton, T.G., and J.K. Grace. 2003. Cuticular permeability of two species of Coptotermes Wasmann (Isoptera: Rhinotermitidae). Comp. Biochem. Physiol. A 134, 205-211.

Slaytor, M. 2000. Energy metabolism in the termite and its gut microbiota. In: Abe, T., Bignell, D.E., Higashi, M., 2000. Termites: Evolution, Sociality, Symbiosis, Ecology. Kluwer Academic, Dordrecht/Norwell. pp. 307–332.

Snodgrass, R.E. 1993. Principles of Insect Morphology. Cornell University Press, Ithafca, New York, USA.

Sponsler, R.C. and A.G. Appel. 1990. Aspects of the water relations of the Formosan and Eastern subterranean termites (Isoptera: Rhinotermitidae). Environ. Entomol. 19, 15-20.

Steward, R.C. 1981. The temperature preferences of building-damaging dry-wood termites (Cyptotermes; Isoptera). J. Therm. Biol. 6: 153-160.

http://entomology.ifas.ufl.edu/creatures/urban/termites/west_indian_drywood_termite.htm

http://entomology.ifas.ufl.edu/creatures/urban/termites/west_indian_drywood_termite.htm

145

Steward R.C. 1982. Comparison of the behavioral and physiological responses to humidity of five species of dry-wood termites, Cryptotermes species. Physiological Entomology 7, 71-82.

Steward R.C. 1983. The effects of humidity, temperature and acclimation on the feeding, water balance and reproduction of dry-wood termites (Cryptotermes). Ent. Exp. & Appl. 33, 135-144.

Strickland, M. 1950. Differences in toleration of drying between species of termites (Reticulitermes). Ecology 31, 373-385.

Su, N.-Y. 2002. Novel technologies for subterranean termites control. Sociobiology 40, 95–101.

Su, N.-Y. 2003. Overview of the global distribution and control of the Formosan subterranean termite. Sociobiology. 41: 7-16.

Su, N.-Y., and H. Puche. 2003. Tunneling activity of subterranean termites (Isoptera: Rhinotermitidae) in sand with moisture gradients. J. Econ. Entomol. 96, 88-93.

Su, N.-Y., and R.H. Scheffrahn. 1986. A method to access, trap, and monitor field populations of the Formosan termite (Isoptera: Rhinotermitidae) in the urban environment. Sociobiology 12, 299–304.

Su, N.-Y., and R.H. Scheffrahn. 1990. Economically important termites in the United Sates and their control. Sociobiology 17, 77-94.

Su, N.-Y., and R.H. Scheffrahn. 1998. A review of subterranean termite control practices and prospects for integrated pest management programmes. Integrated Pest Management Rev. 3, 1–13.

Su, N.-Y., and R.H. Scheffrahn. 2013. Formosan subterranean termite. http://entomology.ifas.ufl.edu/creatures/urban/termites/formosan_termite.htm

Tichy, H., and W. Kallina. 2010. Insect hygroreceptor responses to continuous changes in humidity and air pressure. J. Neurophysiol. 103, 3274-3286.

Tominaga, Y., and F. Yokohari. 1982. External structure of the sensillum capitulum, a hygro- and thermoreceptive sensillum of the cockroach, Periplaneta americana. Cell Tissue Res. 226, 309-318.

Weissling, T.J., and R.M. Giblin-Davis. 1993. Water loss dynamics and humidity preference of Rhynchophorus cruentatus (Coleoptera: Curculionidae) adults. Environ. Entomol. 22, 93-98.

Wigglesworth, V.B. 1945. Transpiration through the cuticle of insects. J. Exp. Biol. 21, 97-114.

146

Winston, P.W., and D.H. Bates. 1960. Saturated solutions for the control of humidity in biological research. Ecology 41, 232-237.

Winston, P.W., and C.C. Green. 1967. Humidity response from antennae of the cockroach Leucophaea maderae. Naturwissenschaften 54, 499.

Wong, N., and C.-Y. Lee. 2010. Influence of different substrate moistures on wood consumption and movement patterns of Microcerotermes crassus and Coptotermes gestroi (Blattodea: Termitidae, Rhinotermitidae). J. Econ. Entomol. 103, 437-442.

Woodrow, R.J., J.K. Grace, L.J. Nelson, and M.I. Haverty. 2000. Modification of cuticular hydrocarbons of Cryptotermes brevis (Isoptera: Kalotermitidae) in response to temperature and relative humidity. Environ. Entomol. 29, 1100-1107.

Woodrow, R.J., and J.K. Grace. 1999. Microclimates associated with Cryptotermes brevis (Isoptera: Kalotermitidae) in the urban environment. Pan-Pacific Entoologist. 75, 68-72.

Yanagawa, A., S. Shimizu, K. Noma, M. Nishikawa, O. Kazumasa, and F. Yokohari. 2009. Classification and distribution of antennal sensilla of the termite Coptotermes formosanus (Isoptera: Rhinotermitidae). Sociobiology 54, 327-349.

Yanagawa, A., T. Yoshimura, T. Yanagawa, and F. Yokohari. 2010. Detection of humidity differences by antennae in of the termite Coptotermes formosanus (Isoptera: Rhinotermitidae). Sociobiology 56, 255-269.

Yokohari, F. 1978. Hygroreceptor mechanism in the antenna of the cockroach Periplaneta. J. Comp. Physiol. 124, 53-60.

Yokohari, F. 1981. The sensillum capitulum, an antennal hygro- and thermoreceptive sensillum of the cockroach, Periplaneta americana L. Cell Tissue Res. 216, 525-543.

Yokohari, F., and H. Tateda. 1976. Moist and dry hygroreceptors for relative humidity of the cockroach, Periplaneta americana L. J. Comp. Physiol. 106, 137-152.

147

BIOGRAPHICAL SKETCH

John Greig Zukowski was born in Garzón, a village nestled in the foothills of the

Andes in southern Colombia. After being adopted at fifteen months of age by a couple

from the United States, he was raised in St. Paul, Minnesota. His high school career at

the military academy, St. Thomas Academy, in Mendota Heights, Minnesota, led him to

apply and matriculate at the University of Wisconsin-Madison. There he earned a

Bachelor of Science in biology and discovered a keen interest in insects. A brief stint of

work in a food plant, as well as a forensic laboratory, and a memorable European

adventure eventually led him to the University of Illinois at Urbana-Champaign, where

he earned his Master of Science degree in entomology and met his future wife. He

continued fostering his love of and interest in insects at the University of Florida, taking

a number of required (and unrequired) courses on his way to passing his qualifying

exams and entering doctoral candidacy. His work on water management in termites at

the Fort Lauderdale Research & Education Center with Dr. Nan-Yao Su, resulted in the

preceding dissertation. He graduated and earned his Doctor of Philosophy in

Entomology and Nematology from the University of Florida in fall of 2015.

water management in termites: a ......4-1 initial mass (mg), percentage of total body water (%tbw),...

Documents