Citation
Soil Displacement during Tunnel Excavation by the Formosan Subterranean Termite, Coptotermes formosanus Shiraki (Isoptera: Rhinotermitidae)

Material Information

Title:
Soil Displacement during Tunnel Excavation by the Formosan Subterranean Termite, Coptotermes formosanus Shiraki (Isoptera: Rhinotermitidae)
Creator:
LI, HOU-FENG ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Biomass production ( jstor )
Boxes ( jstor )
Density ( jstor )
Excavations ( jstor )
Particle density ( jstor )
Population density ( jstor )
Sand ( jstor )
Subterranean termites ( jstor )
Termites ( jstor )
Tunnels ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Hou-Feng Li. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/31/2007
Resource Identifier:
496801534 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

SOIL DISPLACEMENT DURING TUNNEL EXCAVATION BY THE FORMOSAN SUBTERRANEAN TERMITE, Coptotermes formosanus Shiraki (ISOPTERA: RHINOTERMITIDAE) By HOU-FENG LI A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Hou-Feng Li

PAGE 3

To Dr. How-Jing Lee, who led me into the world of entomology.

PAGE 4

ACKNOWLEDGMENTS I acknowledge and thank Dr. Nan-Yao Su for guiding me on the scientific road with clear direction. He offered an environment that allowed me to concentrate on cultivating a scientific mind, and provided the funds for researching and living. I would like to thank Dr. William H. Kern, Jr. who taught five good courses that gave me the fundamental knowledge needed for research. He always encouraged me, and I learned how to be a good instructor by observing his instruction. I also would like to acknowledge my committee members, Dr. Rudolf Scheffrahn and Dr. Samira Daroub, for all of their help and for reviewing my thesis. I would like to thank my lab partners for helping me with my research. Dr. Lee Sang-Hee, Dr. Paul Bardunias and Thomas Chouvenc discussed my experiments with me. Ron Pepin created excellent equipment for my experiments. Paul Ban taught me the practical knowledge for termite research and collected termites for my experiments. Ericka Helmick and Sandy Koi taught me to use English in proper ways. Finally, I would like to thank my parents and brother for their support of my wholehearted concentration on entomological research. Rou-Ling supports me with statistical and entomological expertise and shares with me when I am in distress. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ............................................................................................................vii LIST OF FIGURES .........................................................................................................viii ABSTRACT .......................................................................................................................ix CHAPTER 1 STRUCTURE AND FUNCTION OF WORKER MOUTHPARTS............................1 Materials and Methods.................................................................................................1 Termite..................................................................................................................1 Dissecting Mouthparts...........................................................................................1 Experimental Arena...............................................................................................1 Measurement of Capacity of Buccal Cavity..........................................................2 Results and Discussion.................................................................................................2 Structure of Mouthparts.........................................................................................2 Labrum and clypeus.......................................................................................2 Mandibles.......................................................................................................3 Maxillae..........................................................................................................3 Labium...........................................................................................................3 Movement of Mouthparts......................................................................................4 Tunneling behavior........................................................................................4 Stuffing behavior and the capacity of buccal cavity......................................4 Releasing behavior.........................................................................................4 2 DO FORMOSAN SUBTERRANEAN TERMITES COMPACT SAND PARTICLES DURING EXCAVATION OF TUNNELS?..........................................9 The First Hypothesis: Pushing Soil..............................................................................9 Materials and Methods........................................................................................10 Experimental arena.......................................................................................10 Experimental process...................................................................................10 Analysis........................................................................................................11 Results and Discussion........................................................................................11 The Second Hypothesis: Removing and Then Compacting Soil................................11 v

PAGE 6

Materials and Methods........................................................................................12 Experimental arena.......................................................................................12 Experimental process...................................................................................12 Variable: sand density..................................................................................12 Variable: available space..............................................................................14 Analysis........................................................................................................14 Results.................................................................................................................14 Discussion............................................................................................................16 Soil density and digging behavior................................................................16 Population density and depositing sand behavior........................................17 Mechanism to increase space.......................................................................18 Body size and digging strategies of fossorial animals.................................18 3 AVAILABLE SPACE AND WOOD CONSUMPTION...........................................30 Introduction.................................................................................................................30 Materials and Methods...............................................................................................30 Experimental Process..........................................................................................30 Decreased Weight and Volume of Wood............................................................31 Biomass Production and Survival Rate...............................................................31 Volume and Dry Weight of Carton Material.......................................................32 Metabolic Gas......................................................................................................32 Analysis...............................................................................................................32 Results.........................................................................................................................32 Wood Weight and Volume Loss.........................................................................32 Biomass Production and Survival Rate...............................................................33 Volume and Dry Weight of Carton Material.......................................................33 Metabolic Gas......................................................................................................33 Increased Available Space...................................................................................34 Discussion...................................................................................................................34 Mechanisms of Consuming Wood to Increase Available Space.........................34 Space in Gallery System......................................................................................35 Other Available Space in the Ground..................................................................36 4 SUMMARY AND CONCLUSIONS.........................................................................38 LIST OF REFERENCES...................................................................................................40 BIOGRAPHICAL SKETCH.............................................................................................42 vi

PAGE 7

LIST OF TABLES Table page 3-1 Decreased weight and volume of wood after exposure to Co. formosanus.............36 3-2 Termites’ average dry weight, biomass production, and survival rate.....................37 3-3 Weight, volume, and density of carton material......................................................37 3-4 Percentage of production of consumed wood..........................................................37 3-5 Volume of increased available space for termites....................................................37 vii

PAGE 8

LIST OF FIGURES Figure page 1-1 Lateral view of the worker head of Co. formosanus..................................................5 1-2 Dorsal view of the worker head of Co. formosanus...................................................6 1-3 Dorsal view of the mandibles of a worker of Co. formosanus...................................6 1-4 Lateral view of the left maxilla of a worker of Co. formosanus................................7 1-5 Ventral view of the worker head of Co. formosanus.................................................7 1-6 Load of worker buccal cavity.....................................................................................8 2-1 Schematic pushing sand hypothesis.........................................................................20 2-2 Experimental arena for testing the pushing soil hypothesis.....................................21 2-3 Measurement of width of tunnel and width of moved sand.....................................22 2-4 Experimental arena for testing the removing and compacting soil hypothesis........23 2-5 Definition of given area and sand area.....................................................................24 2-6 Polygon to measure sand area and given area..........................................................25 2-7 Measurement of the available space.........................................................................25 2-8 Sand Density Index of three sites.............................................................................26 2-9 Available space of the arenas at 0, 24, and 48 hours...............................................26 2-10 Population density effect on depositing sand...........................................................27 2-11 Population density effect on decreasing available space.........................................27 2-12 The relationship between soil bulk density and termites’ galleries in soil or roots.28 2-13 Fecal material between sand particles......................................................................28 2-14 Hypothesis of body size and tunneling strategies....................................................29 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science SOIL DISPLACEMENT DURING TUNNEL EXCAVATION BY THE FORMOSAN SUBTERRANEAN TERMITE, Coptotermes formosanus Shiraki (ISOPTERA: RHINOTERMITIDAE) By Hou-Feng Li May 2006 Chair: Nan-Yao Su Major Department: Entomology and Nematology The two main excavation strategies of fossorial animals are compacting soil and removing soil to another place. Formosan subterranean termites do not remove soil to a great extent, and previous researchers speculated that these termites compress soil to make subterranean gallery systems. The structure and function of Formosan subterranean termites’ mouthparts are described in Chapter 1. Two hypotheses of soil compacting in making tunnels were tested. Both of hypotheses were rejected in Chapter 2. We speculated that termites removed soil to somewhere unseen to our eyes. We offered another hypothesis that termites consume wood and gain available space to deposit soil for extending tunnels. In Chapter 3, we demonstrated this hypothesis. Three mechanisms of creating available space are discussed. First, metabolic reaction converts solid wood into metabolic gas. Second, cellulose structure is changed ix

PAGE 10

during the digestion process and becomes denser feces. Third, termites stuff feces into the pore space between sand particles. x

PAGE 11

CHAPTER 1 STRUCTURE AND FUNCTION OF WORKER MOUTHPARTS Materials and Methods Termite Workers, the undifferentiated larvae ( 3 rd instar) and soldiers of the Formosan subterranean termite (FST) Coptotermes formosanus Shiraki were collected from field colonies at Broward County, Florida, by using underground bucket traps (Su and Scheffrahn, 1986) baited with a bundle of spruce pine fir (SPF) wood (Picea sp.). Before testing, termites were kept in one-liter cylindrical plastic jars with pieces of moist SPF wood in the incubator at 28C. Dissecting Mouthparts Termites were put in 80% aqueous ethanol. Workers were set on blue sand ( Activa Products, Inc., Marshall, TX ) and their mouthparts were observed by using a dissecting microscope (Olympus Optical Co., LTD, Japan) and pictures taken by the Canon EOS digital camera (Canon Inc., Japan). Experimental Arena Multiple color and uniform size sand was the substrate for termite tunnel excavation. Sand pre-stain with one of colors including red, green, blue, purple and gray (Activa Products, Inc., Marshall, TX) were sifted through two sieves (corresponding to the U.S.A. standard test sieve scale of 50 and 45, Fisher Scientific, Pittsburgh, PA ) to obtain uniform particle sizes from 0.300 to 0.355 mm. FST could move these size of 1

PAGE 12

2 particles easily (Su et al., 1991). Thirty-five grams (g) of sand was poured into a two-dimensional arena (12.0 x 13.0 x 0.15 cm) consisting of two pieces of plexiglass sheets with 0.15 cm sand in the middle. Ten milliliters (ml) deionized water was added to the sand using a ten ml syringe (Becton, Dickinson and Company, Franklin Lakes, NJ). Ninety workers, ten soldiers and several pieces of wood (total 0.1-0.15 g) were introduced into a 1.2-cm triangle space (4 x 4 x 0.15 cm x 1/2) (a similar space to Fig. 2-4) situated at the corner of the arena. 3 Mixed color and uniform size sand is an ideal substrate for observing the termites’ tunneling behavior. Mixed color sand allows for recognizing the contour of each sand particle under the microscope. The contrast between the colored sand and appendages of mouthparts facilitates observation of how each appendage manipulates a sand particle, the smallest substrate unit. Measurement of Capacity of Buccal Cavity During tunnel excavating, termites stuffed sand particles into their buccal cavities for transportation. The number of sand particles moved over 5 centimeters (cm) by twenty workers was counted separately. The volume of the buccal cavity was estimated by counting the number of uniform size (0.300-0.355 mm) sand particles. Head width of twenty workers was measured as the indication of the termites’ body sizes. Results and Discussion Structure of Mouthparts Labrum and clypeus The labrum is a transparent, convex, and semicircular structure covering the mandibles. The posterior edge of the labrum is adjacent to the anterior edge of the clypeus (Weesner, 1969). Two lightly sclerotized projections on the labrum are divided

PAGE 13

3 by a longitudinal furrow. This longitudinal furrow also divides the two projections on the clypeus (Fig. 1-1 and 1-2). Mandibles Inner and front margins of the mandibles are well sclerotized when viewed laterally. The mandibles have a conical shape (Fig. 1-1). The upper sclerotized margin (Fig. 1-1) is a row of teeth (Fig. 1-3) and the base of it is the primary articulation point (Pa). The lower sclerotized margin (Fig. 1-1) is front margin of the mandible (Fig. 1-2) and the base is the secondary articulation point (Sa). Two primary articulation points (Pa) are clearly visible as reddish spots (Fig. 1-2). Both left and right mandibles have one apical tooth. The left mandible has three marginal teeth. The right mandible has two marginal teeth and one subsidiary tooth (Fig. 1-3). Maxillae The stipes is an elongated structure that supports the lacinia and galea distally and maxillary palps laterally (Fig. 1-1). The galeae encompass the upper and lateral sides of lacinia. The lacinia has an apical double-toothed, heavily sclerotized portion and a basal expansion with a row of long bristles (Fig. 1-4). Labium The labium consists of the postmentum and the prementum. The postmentum is a large shield like plate. The prementum consists of a pair of labial palps and ligula. The ligula is a four-lobed structure with two inner lobes, the glossae, and two outer lobes, the paraglossae (Figs. 1-1 and 1-5).

PAGE 14

4 Movement of Mouthparts Tunneling behavior The maxillae are important structures for tunneling. The bases of the maxillae, (stipes) move forward and backward to make the distal parts (lacinia) move. Workers use lacinia to dig out sand particles. The lacinia is shaped like a reaping hook but with the function of a pickax. The maxillary palps touch the sand particles in front of the lacinia. Workers used the maxillae to rotate the sand particles and stuff them into the mouth cavity enclosed by the maxillae and the mandibles (lateral), the labrum (top), and the labium (bottom). Workers used mandibles to hold the sand particles until they transported the sand to the deposit site. When the mandibles hold sand particles that projected above the dorsal surface of the mandibles, the labrum is deflected upward. The labium spreads downward and props up the sand particles. Pairs of paraglossa and glossa spread laterally. Stuffing behavior and the capacity of buccal cavity Before termites remove sand particles, the particles are rotated in the cavity by maxillae and stuffed deeply into the buccal cavity. During this process, some of sand particles may slip out of the cavity. Sand particle sizes from 0.300 to 0.355 mm, which is in the range that FST can move (Su et al., 1991), were used. The average number of sand particles carried by one worker per trip was 3.50.15 (MeanSE) (Figure 1-6). The average head width was 1.1790.008 mm (MeanSE). Releasing behavior Workers carried three or four sand particles from the excavated site and moved them to deposit sites. They deposited sand in the 1.2-cm 3 triangular space or along the

PAGE 15

5 tunnel that they were digging. Not all particles in one trip were always deposited at the same site. When termites moved, their antennae contacted the boundary of the tunnels. This contact may trigger a response for the termites to place sand at the deposit site. Workers twisted their heads against the walls of deposited sites and opened and closed their mandibles several times. After the mandibles closed, the maxillae moved forward and backward in higher frequency than the mandibles. The tip of maxillae (lacinia) pushed the sand particles to the wall and adjusted the position of each sand particle. The heavily sclerotized portion of the lacinia is easily recognized under magnification. Figure 1-1. Lateral view of the worker head of Co. formosanus. (A) Antennae. (C) Clypeus. (Ga) Galea. (L) Labrum. (La) Lacinia. (Lp) Labial palp. (M) Mandible. (Mp) Maxillary palp. (Pa) Primary articulation point. (Pm) Prementum. (Po) Postmentum. (S) Stipes. (Sa) Secondary articulation point.

PAGE 16

6 Figure 1-2. Dorsal view of the worker head of Co. formosanus. (A) Antennae. (C) Clypeus. (L) Labrum. (Lp) Labial palp. (Mp) Maxillary palp. (Pa) Primary articulation point. Figure 1-3. Dorsal view of the mandibles of a worker of Co. formosanus. (At) Apical tooth. (Mo) Molar plate. (Pa) Primary articulation point. (Sa) Secondary articulation point. (Su) Subsidiary tooth. (1-3) Marginal teeth.

PAGE 17

7 Figure 1-4. Lateral view of the left maxilla of a worker of Co. formosanus. (B) Bristles. (Ga) Galea. (La) Lacinia. (Mp) Maxillary palp. (S) Stipes. Figure 1-5. Ventral view of the worker head of Co. formosanus. (G) Glossa. (Ge) Gena. (Lp) Labial palp. (Mp) Maxillary palp. (Pg) Paraglossa. (Pm) Prementum. (Po) Postmentum. (S) Stipes.

PAGE 18

8 0123456789102345Number of sand particlesNumber of workers Figure 1-6. Load of worker buccal cavity.

PAGE 19

CHAPTER 2 DO FORMOSAN SUBTERRANEAN TERMITES COMPACT SAND PARTICLES DURING EXCAVATION OF TUNNELS? Many authors speculated or assumed that subterranean termites including Reticulitermes hesperus Banks (Ebeling and Pence, 1957), Co. brunneus Gay (Greaves, 1962), Co. acinaciformis (Froggatt) (Greaves and Florence, 1966) and Co. formosanus (King and Spink, 1969) are assumed to compact soil to make tunnel systems. The gallery system of Co. formosanus which excavated by King and Spink (1969) may extend 100 meters. However, there are no direct observations or experimental data in the literature to demonstrate that subterranean termites compact soil to make tunnels. If termites compact soil, there are two possible mechanisms of compacting. First, termites pushed compact soil toward walls, and second, termites remove soil particles to make tunnels and then compact the soil particles at the site of deposition. The First Hypothesis: Pushing Soil Ebeling and Pence (1957) described R. hesperus pressing large sand particles to either side by pushing heads and exterior surfaces of the mandibles and smaller sand particles were removed. Preliminary observation demonstrated FST workers always remove sand particles when tunneled in sand. In this experiment, we wanted to determine if they simultaneously compress sand particles to either side of their tunnels. Our null hypothesis (H 0 ) is that termites only remove sand particles during the excavation process. The alternative 9

PAGE 20

10 hypothesis (H 1 ) is that termites remove sand and simultaneously push sand particles (Fig. 2-1). If termites push sand, the lateral layer of sand along the tunnel becomes denser. The porosity of these pushed sand particles should then decrease. The pore space between sand particles is converted into the tunnel space. Materials and Methods Experimental arena Multiple color and uniform size (0.25-0.30 mm) sand was obtained by the same method in Chapter 1. Thirty-two g sand was filled in a two-dimensional arena space (12.0 x 12.0 x 0.15 cm) leaving a 1.2-cm 3 square space (2.8 x 2.8 x 0.15 cm) at one corner for loading termites (90 workers and 10 soldiers) and one piece of moist SPF wood (1.0 x 1.0 x 0.15 cm) (Fig. 2-2). Eight ml deionized water was added to sand using a 10 ml syringe. The hole on the top plexiglass sheet was plugged after introducing termites. Experimental process In the arenas, termites initially excavated three tunnels A, B, and C (Fig. 2-2). We only used the data from tunnel B because we can observe two sides of the tunnel B instead of one side of tunnel A or C. Four reference points, X, were printed on a transparency film (IKON Office Solutions, Malvern, PA) by a laser printer ( Hewlett-Packard Company, Palo Alto, CA). We attached these transparency films with reference points on arenas. After we took pictures, these four reference points helped us to match the positions of each picture. Two pictures of the same sample were taken with under-light source before and after termites dug tunnels beneath the reference points. These pictures were used to measure the width of tunnel (Wt) (Fig. 2-3 A and C). Two pictures of one sample were taken under microscope with the top-light source before and after termites dug tunnels

PAGE 21

11 beneath the reference points (Fig. 2-3 B and D). We compared these two pictures and then determined which particle in the previous picture was moved or remained stationary during this time period. We drew a line at the interface of the moved and stationary particles on the subsequent photograph (Fig. 2-3 D). The distance between two lines is the width of the moved sand (Wm). The moved sand included removed sand and pushed sand. Three replications were conducted with different termites colonies. Five sub samples were measured in each replication. Analysis We compared Wm and Wt by paired t-test (Excel 2000, Microsoft Corporation, Redmond, WA). H 0 : Wm = Wt. Termites move sand away and do not push the sand laterally when they dig tunnels. H 1 : Wm > Wt. Termites push sand laterally. Results and Discussion Wm and Wt were 3.160.18 mm (Mean SEM) and 2.900.15 mm respectively. There was no significant difference between Wm and Wt (n=15, p=0.12, paired t-test, Excel). The first hypothesis, which termites pushed soil to make tunnels, was rejected. We concluded that termites only move the sand particles away to make tunnels without pushing the sand particles to either side. The Second Hypothesis: Removing and Then Compacting Soil This hypothesis is that termites carry some soil at one site (excavation site) and then move it to another site (deposit site). When termites deposit the soil, they rearrange the particles at a greater density than before it was removed.

PAGE 22

12 We call the space in termites’ tunnels “available space.” If the soil is compacted, the volume of soil decreases and available space increases. During the excavation process, as more soil is compacted, more available space is created. Materials and Methods Experimental arena The arenas were the same as those used to test the first hypothesis except that sand particles size range in size between 0.3-0.355 mm and the starting space is a 1.2 cm 3 triangle (Fig. 2-4). Experimental process The test was done by using three group sizes, 50, 100, or 150 termites (90% workers and 10% soldiers) and replicates three times by using three colonies. Density of unexcavated sand was measured before termites were introduced. Densities of sand deposited in the starting space (1.2 cm 3 triangle) were measured at 24 and 48 hours, at which time termites tunneling activity more or less ceased. The available space was measured from digital images taken at 0, 24, and 48 hours. Variable: sand density Due to the difficulty in obtaining a reliable weight for a small block of sand (2.0 x 2.0 x 1.5 mm) in the two-dimensional arena, we use a Sand Density Index (SDI) instead of sand density. To obtain a larger block of sand to prevent the error of weighing is also difficult because termites moved and deposited small amounts of sand over 24 hours. We took pictures under a dissecting microscope. A given area of 2.0 x 2.0 mm (Fig. 2-5 A) was chosen at random from the arena sand to measure the area (mm 2 ) occupied by sand particles in contact with the top plexiglass sheet (Fig. 2-5 B), which was referred to as the “sand area.”

PAGE 23

13 We used the software, Version 6.1 of Matlab (Mathworks Inc., Natick, MA, 2001), to measure the precise area (pixels) of the approximately 2 x 2 mm given area, and sand area as shown in Fig. 2-5. We circumscribed the sand particles and the given area boundaries manually. A polygon, representing the shape of sand area or given area, made up of line segments between N vertices (xi, yi), i = 0 to N-1 (Fig. 2-6). Sand area and given area was determined by the following formula. The last vertex (xN, yN) is assumed to be the same as the first, so the polygon is closed. The area is given by 101121NiiiiiyxyxA After we obtained these two areas, the formula below was used for calculating SDI. %100areagiven area sand(SDI)Index Density Sand Before introducing termites, the SDI of unexcavated sand was randomly measured. Microsoft Excel 2000 (Microsoft Corporation, Redmond, WA) was used to draw a 2.0 x 2.0 mm grid that was printed on transparency films (IKON Office Solutions, Malvern, PA) by a laser printer ( Hewlett-Packard Company, Palo Alto, CA). This grid was placed over the arena. Five squares (given area) were randomly chosen by the RAND function of Excel 2000. A dissecting microscope and a digital camera were used to take photographs for calculating the SDI. The SDI of two deposit sites was measured by the same method using only areas that were larger than 2.0 x 2.0 mm.

PAGE 24

14 Variable: available space Although these two variables (sand density and available space) have a positive causal relationship, we measured them together for confirmation. Because the layer of sand in the middle of the two plexiglass sheets was very thin (1.5 mm), we were able to measure tunnel area in photos to calculate the available space. After we introduced termites for 0, 24, and 48 hours, we took photos and measured the available space (Fig. 2-7). We used the software GIMP Version 2.2 (Free Software Foundation, Inc., Boston, MA) to calculate the areas. In each experiment, the starting space at 0 hour is 100%. We used a percentage to represent the available space at 24 and 48 hours. The starting space in each experiment was the same 1.0 cm 2 . Analysis We compared the SDI of three sites (pooled data of three populations): unexcavated sites, deposited sites at 24 hour, and deposited sites at 48 hour. We also compared the available space at 0, 24, and 48 hours (pooled data of three populations) (SAS Institute, LSD test, =0.05). The effects of group size (50, 100, and 150 termites) on SDI (pooled data of two deposited sites) and available space (pooled data of 24 and 48 hours) were analyzed (SAS Institute, LSD test, =0.05). Results SDI of the two deposited sites at 24 and 48 hours was significantly lower than that of unexcavated sand (n=135; df=2; F=229.3; p<0.0001). The SDI of deposited sites at 24 hours and 48 hours were 54.25.27% (Mean SEM) and 72.81.06% respectively (n=45). The SDI of unexcavated sand was 82.96.91% (n=45) (Fig. 2-8).

PAGE 25

15 The available space in the arenas at 24 and 48 hours after introducing termites was significantly less than at 0 hour (n=27; df=2; F=62.87; p<0.0001). The available space at 24 and 48 hours were 89.01.27% (Mean SEM) and 87.81.46% respectively (n=9) (Fig. 2-9). In 24 hours, termites excavated tunnels and deposited sand into the introduction chamber. This displacement reduced the sand density (Fig. 2-8), so the available space significantly decreased (Fig. 2-9). During 24-48 hours, the density of deposited sand was only 10% lower than that of unexcavated sand (Fig. 2-8), so the reduction rate of available space slowed down (Fig. 2-9). As the sand density decreased, the available space also decreased. The positive causal relationship of deposited sand density and available space was confirmed. Results of this experiment rejected the second hypothesis that termites remove sand and then compact sand particles to increase available space. Group size had a significant effect on the densities of deposited sand (n=90; df=2; F=76.38; p<0.0001)(LSD test, SAS Institute). Termites groups of 50, 100 and 150 deposited sand with SDI of 51.65.2%, 68.02.17%, and 70.9.92% respectively (n=30, pooled data of SDI of two deposited sites with same group size). Sand deposited by groups of 50 termites was significantly looser than by groups of 100 and 150 termites (Fig. 2-10). Group size significantly affected available space (n=18; df=2; F=18.99; p=0.0002). Sand moved by termites of groups at 50, 100, and 150 altered the available space to 83.88.93%(Mean SEM), 91.01.76%, and 90.34.25% respectively (n=6, pooled data of available space at 24 and 48 hours with same group size). Available space

PAGE 26

16 with groups of 50 termites was significantly lower than for groups of 100 or 150 (Fig. 2-11). The starting space (introduction chamber) for each of the three populations was 1.0 cm 3 , so the average space per termite of 50, 100, 150 populations were 0.02, 0.01, and 0.007 cm 3 / termite respectively. Our experimental data showed termites’ sand deposit density changed between 50 and 100 termites groups. 0.01~0.02 cm 3 per termite is critical space pressure to adjust the deposition behavior. Discussion Soil density and digging behavior The density of unexcavated sand was 1.57.01 g/cm 3 (Mean SEM) (n=9) in our experiments. Under this sand density, our experimental results showed that FST did not compact sand. We speculate that subterranean termites cannot compact soil that is over 1.57 g/cm 3 in the field. The densities of clay, loam, and sand are about 1.5, 1.6, and 1.7 g/cm 3 respectively in the field (Brady and Weil, 2002). The best performance of depositing sand was 78% SDI (population density 100 between 24-48 hours). The SDI of 83% of unexcavated sand is equal to 1.57 g/cm 3 . If the correlation between sand density and SDI is linear, termites can deposit sand to 1.48 g/cm 3 . If termites encounter soil with a density lower than 1.48 g/cm 3 , they may compact soil to increase available space. Greaves and Florence (1966) suggested that the property such as soil bulk density, may limit formation of termite galleries. They excavated tree root systems infested by Co. acinaciformis colonies at four sites and suggested that when termites encountered high bulk density soil, termites tunneled in dead or living roots instead of soil.

PAGE 27

17 Greaves and Florence’s field data (1966) were redescribed by using regression analysis (Excel) (Fig. 2-12). The correlation between soil bulk density with the tunnels in roots or in soil was analyzed (proc corr, SAS Institute). The number of tunnels in soil was negatively correlated with soil bulk density (r= -0.614, p= 0.026). The data showed termites excavating few tunnel in soil with a bulk density over 1.4 g/cm 3 . Thus soil bulk density is a limitation for termites tunneling in soil. On the other hand, the number of tunnels in roots was not correlated with soil bulk density (r=0.463, p= 0.111). The data showed that termites always tunneled in roots whether the bulk density was high or low. The numbers of tunnels in roots and in soil, at the same sites, were compared by paired t-test. Termites excavated significantly more tunnels in the roots (p=0.013, Excel). The data showed that tunneling in roots is a primary excavating strategy, and rejected the hypothesis that tunneling in roots is a substitute for tunneling in soil, as mentioned by Greaves and Florence (1966). Population density and depositing sand behavior Population density is defined as the number of termites per space. In this study, termites with high population density deposited sand at high density. We discuss this relationship from two viewpoints. First, the available space in the arenas of all replications decreased in 24 hours, so the population densities increased. After the first 24 hours, the termites deposited sand at a higher density than in the initial 24 hours. Additionally, the group size is associated with population density. Each of the arenas in our experiment was left with a 1 cm 3 starting (open) space (Fig. 2-4.), therefore,

PAGE 28

18 population density is positive correlated to group size. The sand deposited by 100 and 150 termites was higher in density than that deposited by 50 termites. Although the data showed termites deposited sand more densely when the population density increase, the mechanism by which termites make such adjustment is unknown. We speculated that when population densities increase, the frequency of termites encountering each other increases. Increased physical stimulus from body sensillae result in such adjustment. Mechanism to increase space To discard with excretion is an important issue for termites because they live in a gallery system and excretion occupies available space. In our experiments, we did not observe termites’ excretion in the empty spaces. We observed that the fecal material was stuffed between sand particles (Fig. 2-13). Although removed sand is lower in density than stationary sand and termites lost some space, excretion was stuffed between the removed particles to compensate for the lost space. Body size and digging strategies of fossorial animals Both large animals such as the American badger (Taxidea taxus (Schreber)) (Eldridge, 2004) and seabirds (Bancroft et al., 2005) and small animals such as termites (this study) and ants (Lobry de Bruyn and Conacher, 1990) tend to remove soil to make a tunnel. On the other hand, some medium size animal such as snakes (Pennisi, 2005), and amphisbaenid lizards (Navas et al., 2004) compress soil to create passage. Two factors, body size and compactability of soil, may force animals to evolve their respective strategies. Body size and required tunnel volume are positively correlated with each other (dotted line in Fig. 2-14). The relationship between body size (B) and strength (S) is B 5/3 = S (Price, 1997). Strength increases faster than body size. Animals

PAGE 29

19 with more strength can compact soil to higher density and increase more space, but there are two limitations due to soil compactability. First, an animal must be strong enough to push the soil, so available space does not increase by compacting the soil until the animal obtains a certain level of strength which is related to a body size (the beginning of solid line in Fig. 2-14). The second limitation is the maximal soil bulk density. Compacting causes pore space between soil particles to decrease, thus increasing the animals’ tunnel space, so the soil bulk density increases. If there is no pore space between soil particles, the soil bulk density is 2.65 g/cm 3 because average soil particle is 2.65 g/cm 3 (Brady and Weil, 2002). When soil density is close to 2.65 g/cm 3 , the increased space by compacting a fixed amount of soil is limited (the plateau of solid line in Fig. 2-14). Although the soil types shift the curve (solid line in Fig. 2-14), these two limitations generally exist in all types of soil. We offer this hypothesis to explain why only medium size animals like some snakes use compacting strategies to move in soil. Large animals have enough strength to compress soil but the increased volume is limited and it is not enough for their big body to move underground. Small animals do not have enough strength to compress soil to make a tunnel, although they need much less space than larger body size animals. Only medium size animals have enough strength and at the same time require only medium size of space for their body (the solid line is higher than the dotted line, Fig. 2-14).

PAGE 30

20 WmWt WtWm Remove sand onlyHo: Wt = WmRemove and push sand simultaneouslyH1: Wt < Wm ABC WmWt WtWm Remove sand onlyHo: Wt = WmRemove and push sand simultaneouslyH1: Wt < Wm ABC Figure 2-1. Schematic pushing sand hypothesis. A) Position of sand particles before a termite digs a tunnel. B) The termite only removes two sand particles without pushing sand to make a tunnel. C) The termite removes two sand particles and pushed four sand particles (black ones) to either side. The dotted lines in B and C represent the interface between moved and un-moved sand particles that are identified by comparing with the sand position in A. The width between two dotted lines was defined as width of moved sand (Wm). (Wt) width of tunnels.

PAGE 31

21 Figure 2-2. Experimental arena for testing the pushing soil hypothesis.

PAGE 32

22 A B C D Figure 2-3. Measurement of width of tunnel and width of moved sand. A) and B) Before termites dug the tunnel beneath four reference points. C) and D) After the termites’ tunnel passed the four reference points (R). The black lines in B and D were drawn on the interface between moved and stationary sand by comparing B and D. (R) Reference point. (Wt) Width of tunnel. (Wm) Width of moved sand.

PAGE 33

23 Figure 2-4. Experimental arena for testing the removing and then compacting soil hypothesis.

PAGE 34

24 A B Figure 2-5. Definition of given area and sand area. A) A given area of approximate 2.0 x 2.0 mm square was chosen at random form the arena. B) Sand area represents the area of sand particles touching the top plexiglass sheet. The blue area is where the particles did not contact the top plexiglass sheet. A and B are the same photograph.

PAGE 35

25 Figure 2-6. Polygon to measure sand area and given area. A B C D Figure 2-7. Measurement of the available space (blue areas). A) 0 hour. B) 24 hours. C) 48 hours. D) 72 hours after termites were introduced.

PAGE 36

26 0%10%20%30%40%50%60%70%80%90%100%Excavation site(unexcavated sand)Deposited site 0-24hoursDeposited site 24-48hoursSand Density Index B C A Figure 2-8. Sand Density Index of three sites. Mean ( SEM). 80%85%90%95%100%02448Time (hours) after termite introductionAvailable space ABB Figure 2-9. Available space of the arenas at 0, 24, and 48 hours. Mean ( SEM).

PAGE 37

27 0%20%40%60%80%100%50100150Population densitySand Density IndexBAA Figure 2-10. Population density effect on depositing sand. Mean ( SEM). 75%80%85%90%95%50100150Population densitiesAvailable spac e BAA Figure 2-11. Population density effect on decreasing available space. Mean ( SEM).

PAGE 38

28 R2 = 0.6889R2 = 0.447-2024681012141.21.31.41.51.61.71.8Soil bulk densityNumber of galleries galleries in roots galleries in soil Figure 2-12. The relationship between soil bulk density and termites’ galleries in soil or roots (Adapted from Greaves and Florence, 1966). Figure 2-13. Fecal material between sand particles. The arrows point to the milky fecal material.

PAGE 39

29 Body sizeSpaceRequired tunnel volumeIncreased space by compacting soil SML Body sizeSpaceRequired tunnel volumeIncreased space by compacting soil SML Figure 2-14. Hypothesis of body size and tunneling strategies. (S) small; (M) medium; and (L) large size.

PAGE 40

CHAPTER 3 AVAILABLE SPACE AND WOOD CONSUMPTION Introduction Mound building termites excavate tunnels and move soil to construct their above ground mounds. There is unlimited space for them to deposit soil. However, subterranean termites seldom remove soil to a great extent. In Chapter 2, we rejected two hypotheses of compacting sand to build tunnels. If subterranean termites neither compact soil nor remove soil to a great extent, where do they deposit soil? We offer another hypothesis: subterranean termites gain the available space for depositing soil as a result of consuming wood. Materials and Methods Experimental Process Pieces of number two SPF wood (3.0 x 3.0 x 0.3 cm) were oven dried at 85C for 48 hours and cooled to room temperature before being weighed, and then soaked in deionized water for 24 hours to make the wood moist. Four pieces of moist wood and groups of 200 termites (180 workers, 20 soldiers) were placed in a Pyrex glass Petri dish (60 mm by 15 mm). The control treatment contained no termites. For experiments, termites from three different colonies were used, with four replicates from each colony. Each Petri dish was covered by foil and placed in the incubator (28C) for 30 days. At Day 15, Petri dishes of all treatments were opened for few seconds and deionized water was sprayed to add moisture. After 30 days, we measured the decreased weight and 30

PAGE 41

31 volume of wood, biomass production, weight and volume of carton material, and estimated the weight of metabolic gas. Decreased Weight and Volume of Wood We measured the dry weight of the wood at the beginning and after the 30-day experimental period to estimate wood weight loss. The decreased volume of wood is equal to the decreased weight divided by wood density. The average dry wood density was obtained by measuring five 3.0 cm cubes. Cubes were oven-dried at 85C for 48 hours and cooled to room temperature before being weighed. These five blocks and all experimental wood were from one piece of wood. Biomass Production and Survival Rate Biomass production was estimated as the total dry weight of the surviving termites at Day 30 minus the total termite dry weight at Day 0. However, it is impossible to measure the dry weight and then keep the termites alive for running experiments. In order to resolve this dilemma, we measured the fresh weight first and used dry / fresh ratio of the sub-samples from the same colony to estimate the dry weight. The dry / fresh ratio of sub-samples of each colony was obtained from two groups of 100 workers and two groups of 20 soldiers. We measured fresh weight first and measured dry weight after oven-dried at 85C for 24 hours. Dry / fresh ratios of workers and soldiers were used respectively to estimate the workers’ and soldiers’ experimental dry weight. Surviving termites after the 30-day experimental period were counted for calculating survival rate. According to the total dry weight and number of surviving termites, we can calculate the average survival termite weight.

PAGE 42

32 Volume and Dry Weight of Carton Material After 30 days, termites produced some carton material that was comprised of feces, saliva, corpses, and masticated wood. Carton material was put in a graduated cylinder with water to measure its volume and then measured its dry weight after oven-dried at 85C for 24 hours. We did not separate the composition of carton material because the termites glued them together. In our experiments, all material in the Petri dishes, except wood and live termites, were considered carton material. Metabolic Gas The amount of metabolic gas was estimated as the dry weight of the whole system at Day 0 minus the dry weight at Day 30. Analysis We used the t-test (Excel) to analyze the wood weight loss of experiments and controls (Table 3-1), density of carton material and wood (Table 3-3), and volume of consumed wood and carton material (Table 3-4). The paired t-test was used to analyze the average termite weight at Day 0 and at Day 30 (Table 3-2). The correlation between biomass production with survival rate was analyzed (proc corr, SAS Institute). Results Wood Weight and Volume Loss Wood weight losses for those with termites were significantly larger then controls (Table 3-1), indicating that wood weight loss was due to termite feeding. We measured the wood weight and density so the decreased volume of wood could be obtained (Table 3-1). The wood density of five 3-cm cubes was 0.37.00 g (n=5).

PAGE 43

33 We want to know decreased volume of wood, because it equates to the increased available space for termite. According to our hypothesis, this increased available space could be used for depositing soil during tunnel excavation. Biomass Production and Survival Rate Termites at Day 30 survived weighted (dry weight) more than those at Day 0 (Table 3-2). The total biomass after 30 days decreased due to termite mortality. There was a strong relationship between biomass production and survival rate (p=0.0002, r=0.8793). When survival rate decreased, the biomass production decreased, too. The lost biomass converted into carton material or metabolic gas. Volume and Dry Weight of Carton Material The volume, dry weight, and density of carton material of three colonies are presented in Table 3-3. The density of carton material was variable. The possible reason is the variable composition of feces, saliva, masticated wood, and corpses, were different among colonies. The carton material density of colonies IACH and 121 GI were not significantly different from the density of wood. However, the carton material density of colony 437 GI was significantly higher than wood density. When termites excavated pieces of wood, they masticated and removed wood rather than ingesting the wood. We observed more masticated wood material in experiments with colonies IACH and 121 GI than 437 GI. The higher percentage of masticated wood could decrease the density of carton material. Metabolic Gas The metabolic gas weight of colonies 121 GI, 437 GI and IACH was 0.31.00 g, 0.31.01 g, and 0.38.01g respectively.

PAGE 44

34 The consumed wood converted into metabolic gas, carton material and termites’ biomass production. The metabolic gas weight was 46~52% of consumed wood. Carton material weight was 48~55% of consumed wood. Biomass production weight was -0.27~0.46% of consumed wood (Table 3-4). The consumed wood did not convert into much biomass. On the other hand, termites lost their biomass due to conversion into carton material and metabolic gas. In general, half of the consumed wood became metabolic gas, and about half became carton material. Increased Available Space Available space for termites increased as a result of wood consumption by termites, but carton material also occupied some space. Because the volume of consumed wood was significantly more than that of carton material, the total available space increased by 35~64% of consumed wood (Table 3-5). Discussion Mechanisms of Consuming Wood to Increase Available Space There are two possible mechanisms to explain how termites gain available space after consuming wood. First, metabolic reactions convert solid wood into metabolic gas. CO 2 release of workers and soldiers was 2.95 and 3.22 mg g -1 h -1 respectively under individual isolation and in respirometry chamber situations (Shelton and Appel, 2001). Based on the data, 200 termites may have released 1.35 g CO 2 (0.57g x 720 h x 2.95 mg g -1 h -1 (180 workers) +0.06g x 720 h x 3.22 mg g -1 h -1 (20 soldiers)) during our 30-day experimental period. The experimental metabolic gas production of about 0.31~0.38g over 30 days was probably not an overestimate. The possible reason to explain the difference is that the respirometry chamber environment and individual isolation causes

PAGE 45

35 increased stress on the termites therefore causing the termites to increase CO 2 release. The released metabolic gas in our experiments was about 50% weight of consumed wood. It is an important mechanism to increase space. In the field, subterranean termites consume wood underground and metabolic gas could diffuse from the soil into the atmosphere. Second, wood structure is changed during digestion process and becomes a higher density excretion. This process also created additional space. The density of carton material of our experimental data could offer a clue for this inference. The carton material from one of the three colonies was significant denser than the food (Table 3-3). We collected fecal pellets from several Cryptotermes brevis (Walker) colonies from a wood cabinet. The wood density was 0.30 g/cm 3 , but the fecal pellets were 0.64 g/cm 3 . This suggests that drywood termites use fecal compaction to increase available space. Drywood termites may use both mechanisms to make gallery systems in a piece of wood. Incipient colonies of Cr. queenslandis (Hill) created galleries in blocks of wood without expelling fecal pellets during the first three months that was observed by X-ray equipment (Creffield, 1979). There was no solid material expelled from the introduced hole. The possible mechanisms were metabolic gas release and higher fecal pellet density to increase gallery space. Space in Gallery System An entire FST gallery system was excavated by King and Spink (1969). They described the total length of the tunnel as 580 meters and most of galleries shaped like horizontal slits which were no more than 0.3 cm high with width ranging from 0.3~3.8 cm (King and Spink, 1969). Assuming the average width of the tunnels at 2.0 cm, the total volume of the gallery system was 0.035 m 3 (0.3 x 2 x 58000 x 10 -6 ). The total area

PAGE 46

36 of 5,666 m 2 covered by the galleries was enormously large in comparison with the total gallery volume. Based on this field account, we know FST do not need much space to build a widely extended gallery system. Our experimental data (Table 3-5) showed that increased available space was 35~64% (average is 50%) of consumed wood volume. Thus termites had to consume only 0.07 m 3 of wood to gain 0.035 m 3 for the gallery system that covered 5,666 m 2 (King and Spink, 1969). Stem weight and root weight are positively correlated. Although the ratio of stem weight to root weight differs among plant species, it is close to 1: 1 (Mauseth, 2003). There were at least ten infested trees or stumps at King and Spink’s (1969) investigated site. These wood sources can be used as the termites’ food and also the source of available space to deposit excavated soil when the galleries were constructed. It is concluded that subterranean termites did not have to remove soil to a great extent in order to construct an extensive gallery system. Other Available Space in the Ground When termites excavate, they need space for depositing soil particles. Some biopores formed by organisms such as earthworms, other insects and plant roots could be an available space source. The amount and distribution of biopores depends on the activities of these organisms (Brandy and Weil, 2002). Table 3-1. Decreased weight and volume of wood after exposure to Co. formosanus. We compared decrease wood weight of experiments from each colony with that of the control (=0.05, t-test, Excel). Colony Wood weight loss (g) p Space increased due to consumed wood (cm 3 ) Experiments Control Experiments 121 GI 0.68.01 (n=4) <0.0001 1.840.03 (n=4) 437 GI 0.59.03 (n=4) 0.004 1.600.09 (n=4) IACH 0.800.02 (n=4) 0.02.00 (n=4) <0.0001 2.160.06 (n=4)

PAGE 47

37 Table 3-2. Termites’ average dry weight, biomass production, and survival rate. We compared the average termite weight at the Day 0 with Day 30 (=0.05, paired t-test, Excel). Average termite weight (mg) Colony n Day 0 Day 30 p Biomass production (mg) Survival rate 121 GI 4 0.72.00 0.76.01 0.0134 3.14.54 97.25.75% 437 GI 4 0.84.01 0.91.01 0.0067 -3.03.15 90.25.09% IACH 4 1.200.00 1.340.03 0.0197 -21.965.19 81.380.59% Table 3-3. Weight, volume, and density of carton material. Density of carton material was compared with wood density (=0.05, t-test, Excel). Colony n Carton material dry weight (g) Carton material volume (cm 3 ) Carton material density (g/cm 3 ) Wood density (g/cm 3 ) p 121 GI 4 0.360.01 0.920.05 0.400.02 0.1645 437 GI 4 0.280.02 0.570.04 0.500.01 0.0020 IACH 4 0.440.03 1.400.18 0.320.02 0.37.00 (n=5) 0.1245 Table 3-4. Percentage of production of consumed wood. Colony n Metabolic gas Carton material Biomass production/ loss 121 GI 4 46.19.79% 53.35.82% 0.46.08% 437 GI 4 52.58.32% 48.10.36% -0.68.02% IACH 4 47.351.88% 55.382.37% -2.740.60% Table 3-5. Volume of increased available space for termites. We compared the increased space by consumed wood with decreased space by carton material (=0.05, t-test, Excel). Colony n Space increased due to consumed wood (cm 3 ) Space decreased due to carton material (cm 3 ) p Total increased space (cm 3 ) and percentage of consumed wood 121 GI 4 1.840.03 0.920.05 0.0003 0.92.02, 50.20.99% 437 GI 4 1.60.09 0.57.04 <0.0001 1.02.05, 64.25.14% IACH 4 2.160.06 1.400.18 0.0139 0.76.15, 35.25.92%

PAGE 48

CHAPTER 4 SUMMARY AND CONCLUSIONS During the excavating process, FST loaded the buccal cavity with sand particles at the excavation site and moved it to the deposit site. The amount of sand particles that FST moved per trip was 3.5. The sand particle size ranged from 0.3 to 0.355 mm diameter. The maxilla is the most important component of the mouthpart for carefully manipulating each sand particle. This study showed that FST did not compact 1.57 g/cm 3 sand. Space was required for extending the gallery systems at this sand density. More experiments are needed to study if FST compact clay, loam, and sand that are 1.5, 1.6, and 1.7 g/cm 3 , respectively, in the field. We speculate it is difficult for FST to compact soil when soil density exceeded 1.48 g/cm 3 . Termites may utilize pores or biopores that exist in the soil for depositing soil. The amount of pores and biopores also depends on the type of soil and activities of other organisms. With the exception of the space created by physical processes and other organisms, FST can actively create space by consuming wood. Wood is termite’s food. The increased space is a by-product of wood consumption. Physical pores, biopores and the space created by wood consumption become the available space for FST to deposit sand and to extend tunnels. Where FST deposit soil during excavating is still unknown. We speculate they deposit soil at the closest space from the excavation site, because that is the most efficient way to move sand particles. 38

PAGE 49

39 There are three possible mechanisms for explaining how FST consume wood to gain available space. First, the digested food was converted into metabolic gas during digestion. Second, the lower density wood is converted into denser excretions. Third, the excretion is stuffed into the porous space between sand particles. Population density had a significant effect on termite soil compaction. Higher population density caused workers to compact sand more densely. The critical space pressure is between 0.01~0.02 cm 3 per worker. This is the first description of this phenomenon, but the mechanism is unknown. Frequency of physical stimulus to sensillae on the termite body surface could be a possible mechanism to adjust their behavior.

PAGE 50

LIST OF REFERENCES Bancroft, W. J., M. J. Garkaklis, and J. D. Roberts. 2005. Burrow building in seabird colonies: a soil-forming process in island ecosystems. Pedobiologia 49: 149-165. Brady, N. C., and R. R. Weil. 2002. Soil architecture and physical properties, pp. 94-133. In N.C. Brady and R.R. Weil (eds.), Elements of the Nature and Properties of Soils. Pearson Education Ltd., Upper Saddle River, NJ. Creffield, J. W. 1979. The use of x-rays in studies of drywood termites (Isoptera: Kalotermitidae). J. Inst. Wood Sci. 8: 100-104. Ebeling, W., and R. J. Pence. 1957. Relation of particle size to penetration of subterranean termites through barriers of sand or cinders. J. Econ. Entomol. 50: 690-692. Eldridge, D. J. 2004. Mounds of the American badger (Taxidea taxus): significant features of north American shrub-steppe ecosystems. J. Mammal. 85: 1060-1067. Greaves, T. 1962. Studies of foraging galleries and the invasion of living trees by Coptotermes acinaciformis and C. Brunneus (Isoptera). Aust. J. Zool. 10: 630-651. Greaves, T., and R. G. Florence. 1966. Incidence of termites in blackbutt regrowth. Aust. For. 30: 153-161. King, E.G.J., and W. S. Spink. 1969. Foraging galleries of the Formosan subterranean termite, Coptotermes formosanus, in Louisiana. Ann. Entomol. Soc. Am. 62: 536-524. Lobry de Bruyn, L. A., and A. J. Conacher. 1990. The role of termites and ants in soil modification: a review. Aust. J. Soil. Res. 28: 55-93. Mauseth, J. D. 2003. Roots, pp. 185-208. In J.D. Mauseth (eds.), Botany: an introduction to plant biology. Jones and Bartlett Publisher, Inc., Sudbury, MA. Navas, C. A., M. M. Antoniazzi, J. E. Carvalho, J. G. Chaui-Berlink, R. S. James, C. Jared, T. Kohlsdorf, M. D. Pai-Silva, and R. S. Wilson. 2004. Morphological and physiological specialization for digging in amphisbaenians, an ancient lineage of fossorial vertebrates. J. Exp. Biol. 207: 2433-2441. 40

PAGE 51

41 Pennisi, E. 2005. More than one way to dig a tunnel. Science 307: 347. Price, P. W. 1997. The world of the insect: size and scaling in moderately small organisms, pp. 37-56. In P. W. Price (eds.), Insect Ecology. John Wiley and Sons, Inc., New York, NY. Shelton, T. G., and A. G. Appel. 2001. Carbon dioxide release in Coptotermes formosanus Shiraki and Reticulitermes flavipes (Kollar): effects of caste, mass, and movement. J. Insect Physiol. 47: 213-224. Su, N.-Y., and R. H. Scheffrahn. 1986. A method to access, trap, and monitor field populations of the Formosan subterranean termite (Isoptera: Rhinotermitidae) in the urban environment. Sociobiology 12: 299-304. Su, N.-Y., R. H. Scheffrahn, and P. M. Ban. 1991. Uniform size particle barrier: a physical exclusion device against subterranean termites (Isoptera: Rhinotermitidae). J. Econ. Entomol. 84: 912-916. Weesner F. M. 1969. External anatomy, pp. 19-47. In K. Krishna and F. M. Weesner (eds.), Biology of Termites, vol. 1. Academic Press, New York, NY.

PAGE 52

BIOGRAPHICAL SKETCH Hou-Feng Li was born on 4 March 1979 in Taoyuan, Taiwan. He was fascinated with insects during his childhood and wished to be an entomologist when he was just ten years old. In 1994, he attended Chien-Kuo senior high school in Taipei and participated in the biology club. In 1995, he established the insect lecture in the club and started to do experiments for the science exhibition at the Department of Plant Pathology and Entomology in the National Taiwan University. In 1997, he graduated from high school and became a freshman in the department. From 1997 to 2001, he explored several research areas including aquatic insect investigation, insect ethology, neurophysiology, and chronobiology. Upon graduation from the university, he won five student academic awards and graduated in the first place of his class. In 2001-2003, he did compulsory military service as a secondary lieutenant in the tank company at Taiwan. After this, he returned to his previous department as a research assistant until the summer of 2004. Then, his new life began in the U.S.A. as a graduate student of the University of Florida, Department of Entomology and Nematology, where he did his research and studied at the Fort Lauderdale Research and Education Center. 42