Content uploaded by Nan-Yao Su
Author content
All content in this area was uploaded by Nan-Yao Su on Apr 08, 2015
Content may be subject to copyright.
HOUSEHOLD AND STRUCTURAL INSECTS
Response of the Formosan Subterranean Termites
(Isoptera: Rhinotermitidae) to Baits or Nonrepellent Termiticides
in Extended Foraging Arenas
NAN-YAO SU
Department of Entomology and Nematology, Ft. Lauderdale Research and Education Center, University of Florida,
Institute of Food and Agricultural Sciences, Ft. Lauderdale, FL 33314
J. Econ. Entomol. 98(6): 2143Ð2152 (2005)
ABSTRACT Distance effects of three treatments, novißumuron, Þpronil, and thiamethoxam, against
laboratory populations of the Formosan subterranean termite, Coptotermes formosanus Shiraki, were
tested in extended foraging arenas with foraging distances of 50 m. The results showed that during
the 10-wk test period, all termites were killed by novißumuron baits, whereas the nonrepellent
termiticides Þpronil and thiamethoxam divided the laboratory populations into two groups after
causing 25Ð35% worker mortality. The horizontal transfer of lethal effects of Þpronil was ⱕ5m.For
thiamethoxam, the distance of transfer was substantially shorter. Because of their dose-dependent
lethal time, the nonrepellent termiticides did not fulÞll the requirements of a liquid bait model.
KEY WORDS novißumuron, Þpronil, thiamethoxam, areawide population management
SOIL TREATMENTS WITH LIQUID insecticides have been the
major tool in subterranean termite control since the
1950s. In 2002, for example, soil termiticide applica-
tions accounted for 77% market share of the subter-
ranean termite control business in the United States
(Anonymous 2002). The application of soil termiti-
cides beneath a structure creates a barrier to exclude
soil-borne termites. Because of the extensive foraging
range of an underground colony, soil termiticide
treatments usually do not impact the overall pop-
ulation of subterranean termites (Su and Scheffrahn
1988a). The surviving colony continues to produce
foragers and alates that further infest nearby areas.
The termite control industryÕs reliance on soil ter-
miticide barriers is one of the contributing factors for
the continuing expansion of the Formosan subterra-
nean termite, Coptotermes formosanus Shiraki, in the
United States (Su 2003).
In recent years, nonrepellent termiticides have be-
come popular alternatives for termite control indus-
try. Approximately 60% of the termiticides used in
2002 were one of the nonrepellent termiticides
such as Þpronil (Termidor, BASF Corp., Research
Triangle Park, NC), imidacloprid (Premise, Bayer En-
vironmental Service, Montvale, NJ), or chlorfenapyr
(Phantom, BASF Corp.) (Anonymous 2002). Because
of their nonrepellency and apparent delayed action, it
has been suggested that, unlike the repellent pyre-
throid or fast-acting organic phosphate termiticides,
these new termiticides may impact the subterranean
termite populations (Kard 2001, Potter and Hillery
2002, Wagner 2003, Hu 2005). Laboratory study indi-
cated that termites exposed to sublethal doses of imi-
dacloprid did not show aversion to the subsequent
exposure and may continue to travel through the
treated soil, resulting in the colony suppression
(Thorne and Breisch 2001). Movement of exposed
termites also may spread the nonrepellent toxicants to
nestmates through trophallaxis and social grooming
(Ibrahim et al. 2003, Hu 2005). Shelton and Grace
(2003), however, reported that high concentration of
Þpronil (⬎10 ppm) was needed for a successful trans-
fer of lethal dose to recipients, but at such dose, the
donors may be killed too fast for a substantial toxicant
transfer to occur within the population.
Laboratory evaluation for repellency, delayed ac-
tion, or horizontal transfer of these toxicants typically
uses small groups of Þeld-collected termites for testing
(Thorne and Breisch 2001, Ibrahim et al. 2003, Shelton
and Grace 2003, Hu 2005). Such protocols may be
adequate in evaluating response of individual termites
or a small group of termites to treatments, but it is
often challenging to interpret these laboratory results
for their effects against Þeld colonies with foraging
distance of 50 Ð100 m (King and Spink 1969, Su and
Scheffrahn 1988a, Grace et al. 1989, Su et al. 1993).
Laboratory bioassay with test tubes containing treated
soil (Su and Scheffrahn 1990) and the modiÞed ver-
sions (Su et al. 1995, Gahlhoff and Koehler 2001) were
designed primarily to test termiticide barrier efÞcacy,
and it is probably inappropriate to infer results of
nonrepellent termiticide tube tests for their impact
against subterranean termite populations. None of the
laboratory protocols, including those for slow-acting
baits (Su and Scheffrahn 1993, 1996), could duplicate
the distance factor of the Þeld colonies of subterra-
0022-0493/05/2143Ð2152$04.00/0 䉷 2005 Entomological Society of America
nean termites. This is especially critical in testing the
scope of transfer of nonrepellent and slow-acting tox-
icants by the nestmates over a large area, and the goal
of the control measure is aimed at the colony level.
The objective of this study was to examine the
distance effects of two nonrepellent termiticides,
Þpronil and thiamethoxam (Syngenta Crop Protec-
tion, Greensboro, NC), and one termite bait, novißu-
muron (Recruit III, Dow AgroSciences, Indianapolis,
IN), against laboratory groups of C. formosanus in a
laboratory arena with a linear foraging distance of
⬎50 m. Thiamethoxam is a second generation neo-
nicotinoid belonging to the thianicotinyl chemical
subclass and has been under termiticide Experimental
Use Permit since 2002.
Materials and Methods
The extended foraging arena was composed of 11
small arenas (24 by 24 cm and 1.4 cm in thickness)
connected in linear series with each other by a 5-m-
long coiled Tygon tubing (0.6 cm in diameter) to form
a 50-m (excluding the length of the small arenas)
foraging distance from one end to the other (Fig. 1A).
The small arena was composed of two sheets of trans-
parent Plexiglas (24 by 24 by 0.6 cm in thickness)
separated from each other by Plexiglas laminates
(2 cm in width and 0.2 cm in thickness on three sides,
and 7 cm in width and 0.2 cm in thickness on one side)
placed between the outer margins to form a 0.2-cm gap
of 20 by 15 cm (Fig. 1B). Two pieces of 0.2-cm-thick
transparent Plexiglas (7 by 1.5 cm) were bolted be-
tween the 0.6-cm-thick Plexiglas sheets at 3.5 cm from
each other near the arena center to maintain the
0.2-cm gap of the arena. Approximately 60 g of sifted
sand (150Ð500-
m sieves) was poured in the 0.2-cm
gap and moistened with ⬇15 ml of deionized water.
The upper sheet had an access hole (0.5 cm in diam-
eter) near one corner, onto which a Plexiglas cup
(4.5 cm in diameter and 3 cm in height) with lid was
Þtted to form a termite release chamber (Fig. 1B).
Three transparent plastic tubes (4.5 cm in diameter
and 3 cm in length) containing moistened wood pieces
(Spruce, Picea spp., ⬇2.5 cm in length and ⬇0.5 cm in
diameter) were inserted in the Tygon tubing at three
positions, one at the center and the others near the
Þrst and last small arenas (Fig. 1A).
Termites were collected from three Þeld colonies of
C. formosanus by using the method described by Su
and Scheffrahn (1986). Ten thousand workers (un-
differentiated larvae of at least the third instar) and
1,500 soldiers were divided roughly into 11 groups and
placed in the arena through each release chamber,
after which more moistened wood pieces were placed
into chamber to provide additional food sources. The
arena was placed in the dark at 25 ⫾ 2⬚Cfor1wkto
allow termite movement throughout the 50-m arena,
after which a treatment chamber (4.5 cm in diameter
Fig. 1. The extended foraging arena was composed of 11 small arenas (24 by 24 cm and 1.4 cm in thickness) connected
to each other by a 5-m-long Tygon tubing (0.6 cm in diameter) to form a 50-m foraging distance (A). The small arena was
composed of two sheets of transparent Plexiglas (24 by 24 by 0.6 cm in thickness) with the 0.2-cm gap Þlled with moistened
sand (B). After placing termites through the release chambers, the treatment chamber was inserted near 0-m arena.
2144 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 98, no. 6
and 3 cm in length) was inserted at one section of
Tygon tubing ⬇30 cm from the second arena from one
end (Fig. 1A). The small arena nearest to the treat-
ment chamber was referred to as the “0-m arena,” the
arena at the left end was referred to as the “⫺5-m
arena,” and other small arenas were similarly referred
to by their distance from the treatment chamber,
namely, 5Ð 45-m arenas (Fig. 1A). For novißumuron,
the treatment chamber was Þlled with a roll of cellu-
lose bait containing 0.5% novißumuron, or Recruit III.
For Þpronil or thiamethoxam, the chamber contained
a segment of 2-cm-thick sand treated with either ter-
miticide sandwiched between two segments of 1-cm-
thick untreated sand. Water emulsions of Termidor SC
(at 0.06%) or ACTAW 25 WG (0.1%) (Syngenta Crop
Protection) were applied in sand to yield 80.8 ppm
(wt:wt) Þpronil or 135.8 ppm thiamethoxam. The
treated sand was left overnight to dry before use. The
yielded rate was equivalent to applying 4.07 liters/m
2
sand for preconstruction termiticide treatments
(NPCA 1985), assuming the penetration of the liquid
termiticide onto 2-cm-thick sand. Extended arenas
with treatment chambers Þlled with untreated sand
also were prepared as the control.
After the installation of treatments, digital images of
the small arenas were taken weekly for 10 wk, and the
numbers of live or dead workers were counted from
the images on the computer monitor. Because some,
but not all, corpses decomposed quickly and became
invisible from the digital images, the weekly dead
termite counts included freshly dead individuals as
well as some of those present in previous week. The
weekly data thus represented the observed mortality
for each small arena at the time when the digital image
was taken. The experiments were replicated three
times by using termites of three colony origins. At the
end of the 10-wk period, the arenas were disassembled
to count the numbers of surviving workers and soldiers
and their positions in the arenas. The square-root of
percentage of mortality was arcsine-transformed and
subjected to analysis of variance (ANOVA). SigniÞ-
cant differences (
␣
⫽ 0.05) among treatments were
separated by least signiÞcant difference (LSD) test
(SAS Institute 1985).
Results
Noviflumuron Baits. Termites entered the treat-
ment chamber containing novißumuron bait immedi-
ately after it was inserted in tubing between 0- and 5-m
arenas and began feeding (Fig. 2A). After 1 wk of
feeding, motionless workers enclosed with exuviae
(Fig. 2B) were observed throughout most of the 11
small arenas, with the highest number (approximately
nine workers) found in the ⫺5-m arena (Fig. 3). A
small number of exuviae-wrapped workers, possibly
during their routine molting, were occasionally found
in control arenas (Fig. 3), but the substantially larger
number of such workers in novißumuron arenas im-
plied its chitin synthesis inhibitory (CSI) effects on
termites. The “jack-knife” posture was akin to the
symptom of termites exposed to another chitin syn-
thesis inhibitor, hexaßumuron (Su and Scheffrahn
1993). The affected workers remained alive but mo-
tionless and were extensively groomed by healthy
workers. No dead termites were found in any of the 11
small arenas with novißumuron at 1 wk (Fig. 4). More
termites with CSI effects were found throughout the
50-m arenas at 2 wk (Fig. 3), and some began to die
after developing dark necrotic lesions, resulting in
5Ð30% mortalities at 2 and 3 wk mostly in small arenas
⬇5 m from novißumuron baits (Fig. 4). At 3 wk, the
number of workers with CSI effects declined, and the
peak seemed to shift from ⫺5-m arena at 1 wk to 5-m
arena at 2 wk and further away from baits toward
25-mÐ30-m arenas at 3 wk (Fig. 3). As the mortality
increased at 4 wk, fewer termites with CSI effects were
observed. Even at 4 wk when 50 Ð90% mortality was
recorded from ⫺5- to 50-m arenas, feeding of baits
continued. Akin to the distance pattern of chitin syn-
thesis inhibitory effects, initial mortality peak also was
observed near the treatment chamber (⫺5-mÑ5-m
arenas), but then mortality began to occur near 25Ð
30-m arenas at 4 to 5 wk (Fig. 4). As the mortality levels
at arenas ⫺5Ð10 m reached near 100% at 6 to 7 wk,
mortality in arenas ⬎20 m also increased. Only at 7 wk
when all termites were dead between ⫺5- and 10-m
arenas (including bait chamber), feeding of novißu-
muron baits stopped. This was probably because of the
avoidance of decomposed corpses near the treatment
chamber. No termites were found feeding the baits
after 7 wk, but by then termites throughout the arenas
apparently have had ingested lethal doses of novißu-
muron, either directly or through trophallaxis, as ev-
idenced by the increasing mortality that rippled
through arenas ⬎20 m between 6 and 10 wk (Fig. 4).
By 10 wk, no live termites were found in any of the 11
small arenas in the novißumuron treatment.
Sand Treated with Fipronil. One to 2 d after the
treatment chamber was inserted, dead termites were
found in Þpronil-treated sand (Fig. 5A). At 1 wk when
decomposition of some corpses became evident, no
live termite entered the treated sand. Almost all ter-
mites in 0-m arena for Þpronil at 1 wk were either dead
or moribund (Fig. 6). Moribund termites typically laid
on the side or back with all appendages folded to-
gether (Fig. 5B). Dead termites were often covered
with sand. Some exposed termites, either directly or
through social contact, moved to 5-m arena before
dying, indicating the “horizontal transfer” of Þpronil.
With a few occasions of dead termites found in 30-m
arena, most dead termites were observed in arenas
ⱕ5 m from treated sand throughout the 10-wk test
period (Fig. 6). Termites stopped entering treated
sand after 1 wk, and no additional mortality was ob-
served in arenas ⬎5 m from treatment. Consequently,
the initial population of 10,000 workers (plus 1,500
soldiers) was divided into two isolated groups.
Sand Treated with Thiamethoxam. As with Þpronil
treatment, dead termites were found in thiame-
thoxam-treated sand a few days into the experiment,
and no live termites were entering the treated sand
after 1 wk, when corpse decomposition became evi-
dent. Approximately 30% mortality was recorded in
December 2005 S
U:EXTENDED FORAGING ARENA 2145
the 0-m arena at 1 wk, and with the exception of
⬇5% mortality observed in 20- and 25-m arenas at
2 wk, dead termites (⬇30Ð50% mortality) were found
mostly in 0-m arenas during initial 8-wk period
(Fig. 6). At 9 wk, ⬇30% mortality also was recorded
in ⫺5- and 5-m arenas, but at the 10 wk, dead termites
were found only at the 0-m arena. Between 1 and
10 wk, a few live termites were found in 0-m arena,
whereas many dead termites were buried in sand and
isolated from the live termites (Fig. 5C). Live termites
seemed to travel into the Tygon tube connecting to
the treated chamber, but none entered the treated
sand after the second week. No or very few dead
termites were found in arenas ⱖ5 m from treated sand
throughout the 10-wk test period (Fig. 6). As with the
Þpronil test, the surviving termites in the extended
arenas were divided into two isolated groups.
Final Mortality and Positions of Surviving Ter-
mites. Almost no dead termites were observed in any
of the control arenas throughout the 10-wk experi-
ment (Fig. 4), but the Þnal count revealed a mean
worker mortality of 17.6% (Table 1). The lack of ap-
parent termite corpses in any of the small arenas for
untreated control suggested that dead workers may
have been cannibalized or removed, or some may have
molted into soldier caste. Indeed, more soldiers (1,505
and 1,537) were found in the Þnal count of two rep-
licates of control arenas than initially placed (1,500) in
Fig. 2. Termites fed on novißumuron baits up until 7 wk when 100% mortality was recorded in small arenas ⱕ10 m from
treatment (A), and workers with chitin synthesis inhibitory effects were found through out the 11 small arenas during 1Ð4 wk
(B).
2146 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 98, no. 6
the arenas, suggesting the workerÐsoldier transforma-
tion. As expected from the weekly digital image counts
of novißumuron arenas, no live termites were found in
any parts of the extended arena, resulting in 100%
mortality for both castes of all replicates (Table 1).
The Þnal mean worker mortalities for arenas contain-
ing sand treated with nonrepellent termiticides were
signiÞcantly higher (43.0 and 48.4% for Þpronil and
thiamethoxam, respectively) than untreated control
(17.6%), indicating ⬇25Ð30% of the 10,000 workers
were killed by the termiticides (Table 1). There was
no signiÞcant difference in solider mortalities be-
tween control and Þpronil or thiamethoxam treat-
ment.
The isolated group of ⬇200 workers in the ⫺5-m
arena, wood chamber, and tubing leading to 0-m arena
contained 40% soldiers at 10 wk for both Þpronil and
thiamethoxam (Fig. 7). The other groups of 5,000 Ð
5,500 workers were found in ⱖ5-m arenas (plus tubing
and wood chambers). Soldier proportion for Þpronil
arenas ranged 10 Ð30% for arenas ⬎5 m away from
treated sand, with the largest number (⬇600) found in
the 10-m arena and tubing leading to the 15-m arena.
For thiamethoxam arenas, ⬇25 workers plus ⬇17 sol-
Fig. 3. Numbers of workers enclosed in exuviae recorded in ⫺5Ð45-m arenas for control and novißumuron treatment
during the 10-wk test.
December 2005 SU:EXTENDED FORAGING ARENA 2147
diers (⬇ 40% soldier) were found in 0-m arenas, but
they were in tunnels isolated from dead termites bur-
ied in arena sand (Fig. 5C). The remaining termites
were found in arenas ⱖ5 m with more workers further
away from the treatment, but proportionally more
solders were presented near thiamethoxam-treated
sand and dead termites were in 0-m arena (Fig. 7).
Discussion
Results of this study showed that the CSI novißu-
muron eliminated 10,000 C. formosanus works (plus
1,500 soldiers) within the 50-m arena in 10 wk,
whereas the nonrepellent termiticides divided the
populations into two isolated groups after killing ⬇25Ð
30% of the workers. The results indicated that termites
exposed to Þpronil, either directly or through social
interaction with contaminated individuals, moved
⬇5 m, and those exposed to thiamethoxam moved
⬍5 m before the onset of death. Once dead and de-
composed corpses accumulated near treatments,
healthy termites no longer came in contact with treat-
ment and survived. Because termites entered sand
treated with Þpronil (80.8 ppm) or thiamethoxam
(135.8 ppm) initially, neither termiticide at the tested
concentrations was repellent, and the avoidance was
probably a response to dead termites or decomposed
corpses. It seemed that the horizontal transfer of Þpro-
nil did occur as suggested previously (Kard 2001, Pot-
ter and Hillery 2002, Ibrahim et a. 2003, Wagner 2003,
Fig. 4. Percentage of worker mortality recorded in ⫺5Ð45-m arenas for control and novißumuron treatment during the
10-wk test.
2148 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 98, no. 6
Hu 2005), but the distance of transfer was ⱕ5m.As
reported by Shelton and Grace (2003), those exposed
to high doses of Þpronil may be able to transfer lethal
doses to others, but apparently they could not move
too far away from treatment before being immobi-
lized. For thiamethoxam at 135.8 ppm, the distance of
transfer was substantially shorter.
Potter and Hillery (2002), who used in-ground
“bucket” stations to measure activity of the eastern
subterranean termite, Reticulitermes flavipes (Kollar),
reported elimination of termite activity in six of eight
sites after soil application of Þpronil. The stations
that remained active at two sites in their report were
⬎5 m away from treated soil. Their results thus do
not contradict with results of this study. A Þeld study
reported that soil application of the nonrepellent
termiticide imidacloprid did not measurably impact
nearby populations of C. formosanus (Osbrink et al.,
2005). Another Þeld study to compare termite control
measures against the invasive R. flavipes in Chile also
showed that applications of Þpronil in soil did not
affect colony activities in the vicinity of treatments
(R. Ripa, personal communication).
Numerous studies have demonstrated the elimina-
tion of all detectable subterranean termite activity by
hexaßumuron or novißumuron baits at a variety of
locations with many termite species, “to the point
where citations to the published literature are super-
ßuous” (Grace and Su 2001). Results of this laboratory
test conÞrmed previous Þeld studies that the lethal
effects of novißumuron can affect a population with a
foraging distance of 50 m, leading to colony elimina-
tion.
The main difference between the nonrepellent ter-
miticides and CSIs such as novißumuron is their lethal
time, i.e., time required to death after exposure to
lethal doses (Su and Scheffrahn 1988b). Unlike novi-
ßumuron, lethal times of Þpronil and thiamethoxam
are dose-dependent (Remmen and Su 2005). Individ-
uals contaminated (either directly or through con-
tact) with lower but lethal doses were able to move
away from treatment as evidenced by the dead ter-
mites found in arenas 5 m away from Þpronil-treated
sand. But those exposed to higher lethal doses, either
by repeatedly entering or continuously tunneling
treated sand, may die rather quickly in or near the
treatment. The large number of dead or decomposed
corpses in areas near the treatment probably acted as
the repellent to healthy termites. The avoidance of
dead and decomposed corpses (necrophobic behav-
ior) was well documented in termites such as C. for-
mosanus (Su 1982) and Pseudacanthotermes spiniger
Sjo¨ stedt (Chouvenc 2003). The effect of nonrepellent
termiticides when acquired by termites at higher
doses, therefore, is similar to that of fast-acting ter-
miticides such as chlorpyrifos, and their barrier efÞ-
cacy is because of secondary repellency (Su et al.
1982).
It remains possible that soil treatments with non-
repellent termiticides may eliminate a colony whose
entire foraging range is within 5 m (or less for thia-
methoxam) of treatment, but the treated soil must
contain the ideal (active ingredient) concentration(s)
that is both lethal and slow-acting. Treated sand used
in this study contained a homogeneous concentration,
but it is unlikely that applications of liquid termiticides
in the Þeld, either by trenching, rodding, or drilling
and injection, can produce a homogeneously lethal
and slow-acting concentration that allows contami-
Fig. 5. Dead termites were found in Þpronil-treated sand
1 to 2 d after it was inserted in the arena (A), and moribund
(those laying on the side or back with all appendages folded
together) and dead termites (covered with sand particles)
were observed in 0 m arena at 1 wk (B). Throughout the
10-wk experiment for thiamethoxam, live termites were
found in 0-m arena where corpses were buried in sand and
well isolated from live termites (C).
December 2005 SU:EXTENDED FORAGING ARENA 2149
nated termites to move away from treated soil at any
substantial distance before being immobilized. Ter-
mites exposed to soil with higher concentrations
would be killed rather quickly, those exposed to lower
nonlethal doses would survive, and only those exposed
to the intermediate range of both lethal and slow-
acting doses may move away from treated soil before
dying. Because foraging galleries of a colony may ori-
ent toward any direction; not necessary only toward
treated soil, some termites would survive and the col-
ony may be divided into more than one disconnected
populations as shown in this study.
To manage populations of subterranean termites in
a large area, especially the invasive species such as
C. formosanus, the ability of a control agent to elimi-
nate or kill the entire colony is vital. Because the
surviving colony would continue to produce alates
that further infest nearby areas, use of a soil termiti-
cide, be it nonrepellent or otherwise, would allow
Fig. 6. Percentage of worker mortality recorded in ⫺5Ð45-m arenas for Þpronil and thiamethoxam treatment during the
10-wk test.
Table 1. Mortality (% ⴞ SE) of C. formosanus workers and
soldiers in the extended foraging arenas containing noviflumuron
bait or sand treated with fipronil or thiamethoxam during a 10-wk
experiment
Treatment Worker Soldier
Control 17.6 ⫾ 2.1a 13.8 ⫾ 13.8a
Novißumuron 100.0 ⫾ 100.0b 100.0 ⫾ 100.0b
Fipronil 43.0 ⫾ 0.4c 27.6 ⫾ 4.4a
Thiamethoxam 48.4 ⫾ 7.0c 20.4 ⫾ 6.1a
Means followed by the same letter within a column are not signiÞ-
cantly different (
␣
⫽ 0.05) according to LSD test (SAS Institute 1985).
2150 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 98, no. 6
further establishment and spread of C. formosanus. As
reported by Osbrink et al. (2005), the nonrepellent
termiticides apparently did not fulÞll the require-
ments of a liquid bait model. If C. formosanus popu-
lations are to be managed, there is a need to use a
control measure that can kill the entire colony.
Acknowledgments
I thank R. Pepin (University of Florida) and A. Mullins
(New Orleans Mosquito and Termite Control Board) for
technical assistance, and R. H. Scheffrahn and B. Cabrera
(University of Florida) for review of the manuscript. This
research was supported by the Florida Agricultural Experi-
ment Station and a grant from USDAÐARS under the grant
agreement no. 58-6435-2Ð 0023. Additional funding was pro-
vided by Syngenta Crop Protection. This article was ap-
proved for publication as Journal Series No. R-10895.
References Cited
Anonymous. 2002. State of the industry 2002. Pest Control
70: S1ÐS23.
Chouvenc, T. 2003. Corpse burial behavior of the growing-
fungus termite, Pseudacanthotermes spiniger (Termitidae,
Macrotermitidae): necrophoric behavior and its induc-
tion. M.S. thesis, University of Burgundy, Dijon, France.
Gahlhoff, J. E., and P. G. Koehler. 2001. Penetration of the
eastern subterranean termite into soil treated at various
thicknesses and concentrations of Dursban TC and
Premise 75. J. Econ. Entomol. 94: 486Ð491.
Grace, J. K., and N.-Y. Su. 2001. Evidence supporting the use
of termite baiting system for long-term structural pro-
tection (Isoptera). Sociobiology 37: 301Ð310.
Grace, J. K., A. Abdallay, and K. R. Farr. 1989. Eastern sub-
terranean termite (Isoptera: Rhinotermitidae) foraging
territories and populations in Toronto. Can. Entomol. 121:
551Ð556.
Hu, X. P. 2005. Evaluation of efÞcacy and nonrepellency of
indoxacarb and Þpronil-treated soil at various concentra-
tions and thickness against two subterranean termites
(Isoptera: Rhinotermitidae). J. Econ. Entomol. 98: 509Ð
517.
Ibrahim, S. A., G. Henderson, and H. Fei. 2003. Toxicity,
repellency, and horizontal transmission of Þpronil in the
Formosan subterranean termite (Isoptera: Rhinotermiti-
dae). J. Econ. Entomol. 96: 461Ð 467.
Fig. 7. Distribution of surviving workers and percentage of soldiers in the extended arena 10 wk after Þpronil- or
thiamethoxam-treated sand was placed in the arena. Data for ⫺5 m included the isolated population in ⫺5-m arena plus those
in the Tygon tubing (and the wood chamber) between 0- and ⫺5-m arenas. Data for 0 m included those found in the 0-m
arena and treatment chamber only. The 5-m data included those in the 5-m small area and in the portion of tubing between
the treatment chamber and the arena. For the rest, the data included those found in the small arenas plus in the 5-m tubing
toward treatment chamber.
December 2005 SU:EXTENDED FORAGING ARENA 2151
Kard, B. 2001. Gulfport studies stay the course. Pest Control
69: 30 Ð33.
King, E. G., and W. T. Spink. 1969. Foraging galleries of the
Formosan termite, Coptotermes formosanus, in Louisiana.
Ann. Entomol. Soc. Am. 62: 537Ð542.
[NPCA] National Pest Control Association. 1985. Ap-
proved reference procedures for subterranean termite
control. NPCA Inc., Dunn Loring, VA.
Osbrink, W.L.A., M. L. Cornelius, and A. Lax. 2005. Effect
of imidacloprid soil treatments on the occurrence of For-
mosan subterranean termite (Isoptera: Rhinotermitidae)
in independent monitors. J. Econ. Entomol. 98: 2160 Ð
2168.
Potter, M. H., and A. E. Hillery. 2002. Exterior-targeted
liquid termiticides: an alternative approach to managing
subterranean termites (Isoptera: Rhinotermitidae) in
buildings. Sociobiology 39: 373Ð 405.
Remmen, L. N., and N.-Y. Su. 2005. Time trends in mortality
for thiamethoxam and Þpronil against Formosan subter-
ranean termites and eastern subterranean termites
(Isoptera: Rhinotermitidae). J. Econ. Entomol. 98: 911Ð
915.
SAS Institute. 1985. SAS userÕs guide: statistics, version 5 ed.
SAS Institute, Cary, NC.
Shelton, T. G., and J. K. Grace. 2003. Effects of exposure
duration on transfer on nonrepellent termiticides among
workers of Coptotermes formosanus Shiraki (Isoptera:
Rhinotermitidae). J. Econ. Entomol. 96: 456 Ð 460.
Su, N.-Y. 1982. An ethological approach to the remedial con-
trol of the Formosan subterranean termite, Coptotermes
formosanus Shiraki. Ph.D. dissertation, University of Ha-
waii, Honolulu.
Su, N.-Y. 2003. Overview of the global distribution and con-
trol of the Formosan subterranean termite. Sociobiology
41: 7Ð16.
Su, N.-Y., and R. H. Scheffrahn. 1986. A method to access,
trap, and monitor Þeld populations of the Formosan sub-
terranean termite (Isoptera: Rhinotermitidae) in the ur-
ban environment. Sociobiology 12: 299Ð304.
Su, N.-Y., and R. H. Scheffrahn. 1988a. Foraging population
and territory of the Formosan subterranean termite
(Isoptera: Rhinotermitidae) in an urban environment.
Sociobiology 14: 353Ð359.
Su, N.-Y., and R. H. Scheffrahn. 1988b. A review of subter-
ranean termite control practices and prospects for inte-
grated pest management programs. Integrated Pest Man-
agement Reviews. 3: 1Ð13.
Su, N.-Y., and R. H. Scheffrahn. 1990. Comparison of eleven
soil termiticides against the Formosan and eastern sub-
terranean termites (Isoptera: Rhinotermitidae). J. Econ.
Entomol. 83: 1918Ð1924.
Su, N.-Y., and R. H. Scheffrahn. 1993. Laboratory evaluation
of two chitin synthesis inhibitors, hexaßumuron and di-
ßubenzuron, as bait toxicants against Formosan and east-
ern subterranean termites (Isoptera: Rhinotermitidae).
J. Econ. Entomol. 86: 1453Ð1457.
Su, N.-Y., and R. H. Scheffrahn. 1996. Comparative effects of
two chitin synthesis inhibitors, hexaßumuron and lufenu-
ron, in a bait matrix against subterranean termites
(Isoptera: Rhinotermitidae). J. Econ. Entomol. 89: 1156 Ð
1160.
Su, N.-Y., M. Tamashiro, J. R. Yates, and M. I. Haverty. 1982.
Effects of behavior on the evaluation of insecticides for
prevention of or remedial control of the Formosan sub-
terranean termite. J. Econ. Entomol. 75: 188Ð193.
Su, N.-Y., P. M. Ban, and R. H. Scheffrahn. 1993. Foraging
populations and territories of the eastern subterranean
termite (Isoptera: Rhinotermitidae) in southeastern Flor-
ida. Environ. Entomol. 22: 1113Ð1117.
Su, N.-Y., G. S. Wheeler, and R. H. Scheffrahn. 1995. Sub-
terranean termite penetration into sand treated with ter-
miticides at various thicknesses. J. Econ. Entomol. 88:
1690Ð1694.
Thorne, B. L., and N. L. Breisch. 2001. Effects of sublethal
exposure to imidacloprid on subsequent behavior of sub-
terranean termite Reticulitermes virginicus (Isoptera: Rhi-
notermitidae). J. Econ. Entomol. 94: 492Ð 498.
Wagner, T. 2003. U.S. Forest Service termiticide tests. So-
ciobiology 41: 131Ð141.
Received 27 May 2005; accepted 9 August 2005.
2152 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 98, no. 6
