|
Adamczyk & Gore: Activity of CrylAc and CrylF Against Armyworms
LABORATORY AND FIELD PERFORMANCE OF COTTON
CONTAINING CRY1AC, CRY1F, AND BOTH CRYIAC
AND CRY1F (WIDESTRIKE) AGAINST BEET ARMYWORM
AND FALL ARMYWORM LARVAE (LEPIDOPTERA: NOCTUIDAE)
J. J. ADAMCZYK, JR. AND J. GORE
USDA, ARS, Southern Insect Management Research Unit, P.O. Box 346, Stoneville, MS 38776
ABSTRACT
The efficacy of transgenic cotton genotypes containing CrylAc, CrylF, and CrylAc stacked
with CrylF (WideStrike, Dow Agrosciences, Indianapolis, IN) were investigated during
2001-2003 against the beet armyworm, Spodoptera exigua (Htibner) (=BAW), and the fall ar-
myworm, Spodoptera frugiperda (J. E. Smith) (=FAW), in laboratory bioassays and small ex-
perimental field plots. In all experiments, cotton containing CrylF was more toxic to BAW
and FAW larvae compared to cotton containing only CrylAc. In the majority of experiments,
the addition of CrylAc to the CrylF genotype had no increased effect on efficacy and certain
biological parameters against BAW and FAW larvae compared to cotton containing only
CrylF. Furthermore, the presence or absence of an additive, synergistic, or antagonistic ef-
fect between CrylAc and CrylF was not observed in these field and laboratory experiments.
Key Words: cotton, Cryl genes, transgenic cotton, beet armyworm, fall armyworm,
Spodoptera spp.
RESUME
La eficacia de los genotipos de algod6n transg6nicos que tienen CrylAc, CrylF, y CrylAc
combinados con CrylF (WideStrike, Dow Agrosciences, Indianapolis, IN) fueron investiga-
dos durante 2001-2003 contra del gusano trozador (BAW), Spodoptera exigua (Htibner), y el
gusano cogollero (FAW), Spodoptera frugiperda (J. E. Smith), en bioensayos de laboratorio y
en experiments en parcelas pequeias de campo. En todos los experiments, el algod6n que
tenia CrylF fue el mas t6xico a las larvas de BAW y de FAW en comparaci6n al algod6n que
tenia solamente CrylAc. En la mayoria de los experiments, la adici6n de CrylAc al genotipo
CrylF no tuvo un incremento sobre la eficacia y ciertas parametros biol6gicos contra las lar-
vas de BAW y FAW comparado al algod6n que tenia solamente CrylF. Ademas, no se observe
la presencia o ausensia de un efecto aditivo, sinergistico o antag6nico entire el CrylAc y
CrylF en estos experiments de campo y de laboratorio.
Since the first CrylAc Bacillus thuringiensis
Berliner (Bt) cotton variety was commercialized in
1996 (Bollgard, Monsanto Ag. Co., St. Louis,
MO), there have been numerous advancements for
insect control with transgenic technology. Current
and experimental cotton varieties can contain
CrylAc alone or they can be stacked with Cry2Ab
(Bollgard II, Monsanto Ag. Co.) or CrylF (Wide-
Strike, Dow Agrosciences, Indianapolis, IN).
Furthermore, a novel exotoxin from B. thuringien-
sis also is currently in development (VipCot,
Syngenta Crop Protection, Greensboro, NC).
The beet armyworm, Spodoptera exigua (Hiib-
ner) (=BAW), is an occasional but serious pest of
various vegetable and row crops in the mid-south-
ern United States of America (=Mid-South). Com-
pared to other North American armyworm
species, knowledge of the ecology of this pest in
the Mid-South is limited. This pest has no known
photoperiod or temperature induced diapause
mechanism (Kim & Kim 1997), and it is able to
overwinter by continuous generations in southern
Florida and Texas. Therefore, initial populations
of beet armyworms found throughout the Mid-
South are believed to be the result of immigration
from those areas (Mitchell 1979; Hendricks et al.
1995). Populations in the Mid-South are typically
found in cotton after July 1, with higher popula-
tions on various wild hosts in the fall months (Ad-
amczyk et al. 2003). Although larval feeding on
cotton is primarily concentrated on foliage, larvae
can cause devastating losses in yield (Hardee &
Herzog 1997; Adamczyk et al. 1998).
The fall armyworm, Spodoptera frugiperda (J.
E. Smith) (=FAW), also is a destructive migratory
pest of many crops in the Western Hemisphere
(Sparks 1979; Young 1979). Like the BAW, this
pest has the potential to damage both conven-
tional cotton bolls and Bollgard cotton bolls (Ad-
amczyk et al. 1998).
Although certain lepidopterous pests of cotton
are controlled by Bollgard cotton [e.g., tobacco
budworms and pink bollworms, Pectinophora gos-
sypiella (Saunders)], the CrylAc 6-endotoxin in
Florida Entomologist 87(4)
Bollgard cotton is ineffective for control of BAW
and FAW (Adamczyk et al. 1998; Henneberry et
al. 2001). Consequently, outbreaks of BAW and
FAW on Bollgard@ often need full application
rates of foliar insecticide treatments to keep these
populations below economic injury levels (Hood
1997; Smith 1997). Efficacy data for BAW and
FAW feeding on Bollgard II@ is mainly limited to
laboratory bioassays and small experimental field
plots. However, the addition of Cry2Ab along with
CrylAc appears to have improved the efficacy of
Bollgard II@ against both BAW and FAW (Adam-
czyk et al. 2001; Stewart et al. 2001). The purpose
of the study was to examine the efficacy of
CrylAc, CrylF, and CrylAc stacked with CrylF
against BAW and FAW in laboratory bioassays
and small experimental field plots.
MATERIALS AND METHODS
Field Plots
From 2001-2003, experimental transgenic cot-
ton varieties (Dow Agrosciences, Indianapolis,
IN) were planted in research plots near Eliza-
beth, MS under a yearly Experimental Use Per-
mit (EUP) (Table 1). In 2001, cotton was planted
on May 23 and plots consisted of 2 rows (1.0 m
centers) x 10.67 m. In 2002 and 2003, cotton was
planted on May 13 and June 10, respectively, and
plots consisted of 4 rows (1.0 m centers) x 12.20
m. All plots were arranged in a randomized com-
plete block design with each variety replicated 8
times (twice in each block). Plots were irrigated
once in 2002 and twice in 2001 and 2003. Only in-
secticides not active on Lepidoptera were applied
to all plots throughout the season as dictated by
local management practices.
Insects
Colonies of fall armyworms (FAW) and beet ar-
myworms (BAW) were established from local mi-
gratory populations found in the Mississippi
Delta. In May 2001, larvae (ca. 500) of FAW were
collected from whorl-stage field corn near Stonev-
TABLE 1. EXPERIMENTAL TRANSGENIC COTTON GENO-
TYPES EVALUATED IN 2001-2003.
Genotype Cry genes Years evaluated
MXB-7 1Ac 2001, 2002
(CrylAc)
MXB-9 1F 2001, 2002
(CrylF)
MXB-13 lAc and IF 2001-2003
(CrylAc/CrylF or Widestrike)
PSC355 None 2001-2003
(conventional cotton isoline)
ille, MS. FAW egg masses (ca. 20) were collected
in August 2003 from royal paulownia, Paulownia
tomentosa (thumb.) (Sieb. & Zucc. ex Steud.).
BAW larvae (ca. 500 each year) were collected
from redroot pigweed, Amaranthus retroflexus L.,
in June 2001 and July 2002. Both species were
reared for one complete generation in the labora-
tory as described by Adamczyk et al. (1998), and
the subsequent generation was utilized in either
bioassays or field inoculations.
Field Experiments
Inoculations of BAW egg masses to plants for
all varieties were conducted in 2001 and 2002. In
the laboratory, egg masses were deposited on ny-
lon cloth placed on the top of adult rearing cages
(3.79-liter cardboard containers). For each inocu-
lation, an egg mass of equal size (ca. 200-300
eggs/2.54-cm2 cloth sample) was stapled to the un-
derside of a mature leaf in all plots. Egg masses
were spaced ca. 0.5 m from each other. Each plot
received a total of 42 egg masses on July 10-12,
2001 and 56 egg masses on August 1-2, 2002.
Eight days after inoculations (DAI), BAW popula-
tions were estimated with a standard 1.2-m drop
cloth placed in the center row per plot (3 samples/
plot). All recovered larvae within a plot were
placed in 232-ml plastic containers and trans-
ported to the laboratory and weighed within 1 h
after arrival. Prior to analysis, the coefficient of
variation for the mean weights and number of lar-
vae among genotypes was substantially improved
by a log transformation. Both mean weights and
numbers of BAW were analyzed by REML-
ANOVA and were separated according to Fisher's
Protected LSD (Littell et al. 1996; PROC MIXED,
SAS Institute 2001).
In 2001, a minimum of 15 leaves (i.e., egg
masses)/plot that showed evidence of successfully
hatched neonates were visually examined after 9
DAI for BAW damage. Leaf damage was esti-
mated with a categorical rating scale where 0%
indicated no leaf damage while evidence of leaf
consumption was given a value of 10%, 25%, or
50%. Prior to analysis, damage ratings appeared
to be normally distributed, and a square-root
transformation did not improve the coefficient of
variation significantly among genotypes. There-
fore, no transformation on this categorical data
set was needed. Mean damage ratings were ana-
lyzed by REML-ANOVA, and means were sepa-
rated according to Fisher's Protected LSD (Littell
et al. 1996, PROC MIXED, SAS Institute 2001).
Bioassays
First-position flower buds (=squares) contain-
ing various transgenes (Table 1) were assayed in
2001 for bioactivity against BAW and FAW neo-
nates. Individual squares were placed into a
December 2004
Adamczyk & Gore: Activity of CrylAc and CrylF Against Armyworms
Tight-Fit Lid sealing Petri dish (50 x 9 mm, BD
Falcons #351006, VWR International). For BAW,
3 larvae were placed in a dish containing a single
square (5 dishes/plot) for a total of 120 larvae/gen-
otype. For FAW, a single larva was placed in a
dish containing a single square (10 dishes/plot)
for a total of 40 larvae/genotype. Four days after
exposure (DAE), larvae were prodded with a
camel-hair brush and considered alive if coordi-
nated movement was observed. Percent survival
of neonates was analyzed by REML-ANOVA, and
means were separated according to Fisher's Pro-
tected LSD (Littell et al. 1996; PROC MIXED,
SAS Institute 2001).
In 2002, terminal leaves containing various
transgenes (Table 1) were assayed for bioactivity
against BAW neonates. Individual leaves were
placed into a Tight-Fit Lid sealing Petri dish (50 x
9 mm, BD Falcon #351006, VWR International).
Three larvae were placed in a dish containing a
single terminal leaf (5 dishes/plot) for a total of
120 larvae/genotype. At 3, 6, and 8 DAE, larvae
were prodded with a camel-hair brush and consid-
ered alive if coordinated movement was observed.
Percent survival of neonates was analyzed by
REML-ANOVA, and means were separated ac-
cording to Fisher's Protected LSD (Littell et al.
1996; PROC MIXED, SAS Institute 2001).
In 2003, terminal leaves containing only the
CrylAc/CrylF genotype (Table 1) were assayed
for bioactivity against FAW neonates. A single
larva was placed in a dish containing a single
square or single terminal leaf (20 dishes/plot) for
a total of 160 larvae/genotype. At 4 and 7 DAE,
larvae were prodded with a camel-hair brush and
considered alive if coordinated movement was ob-
served. Percent survival of neonates was analyzed
by REML-ANOVA, and means were separated ac-
cording to Fisher's Protected LSD (Littell et al.
1996; PROC MIXED, SAS Institute 2001).
RESULTS AND DISCUSSION
Field Experiments
Experimental cotton genotypes had differen-
tial affects on the survival of BAW larvae (Table
2). In 2001, the mean number of BAW larvae was
significantly reduced in plots containing CrylF
and CrylF stacked with CrylAc compared to the
plots containing CrylAc alone or conventional
cotton (PSC 355). The mean number of BAW lar-
vae found in plots containing only CrylAc was not
significantly different from the mean number of
larvae found in conventional cotton. Therefore,
the addition of CrylAc to cotton containing CrylF
had no significant effect in reducing the number
of BAW larvae found in plots containing both
these transgenes. However, in 2002 all transgenic
cotton genotypes had significantly reduced mean
BAW larval numbers as compared to the mean
TABLE 2. NUMBER OF BAW LARVAE FOUND ON VARIOUS
TRANSGENIC GENOTYPES 8 DAYS AFTER EGG IN-
OCULATIONS IN 2001 AND 2002.
Mean number of larvae SEM
Genotype 2001 2002
PSC 355 37.8 13.16 a 17.8 9.44 a
CrylAc 31.0 10.73 a 5.3 1.70 b
CrylF 4.8 1.11 b 1.5 0.50 c
CrylAc/CrylF 6.3 1.75 b 3.0 0.71 bc
df 3,9 3,9
F value 16.20 13.43
(P > F) ANOVA <0.001 0.001
Means in a column followed by the same letter are not sig-
nificantly different (a = 0.05); Fisher's Protected LSD option of
PROC MIXED (Littell et al. 1996; SAS Institute 2001). Means
were log-transformed prior to analysis.
number of BAW larvae found in conventional cot-
ton. Unlike 2001, cotton containing only CrylAc
had significantly reduced BAW larval numbers as
compared to the mean number of BAW larvae
found in conventional cotton. In an adjacent ex-
periment, higher parasitism rates of larvae from
Cotesia marginiventris (Cresson) were observed
in 2002 compared to 2001 (Adamczyk & Hardee
2002). This may have contributed to the lower
numbers of recovered larvae in all plots in 2002
than in 2001. Sublethal affects of CrylAc (Hen-
neberry et al. 2001) combined with increased par-
asitism may have resulted in the increased BAW
mortality found in the CrylAc genotype in 2002
as compared to 2001. Another possibility is that
different agronomic conditions (e.g., higher nitro-
gen levels) may have increased the level of
CrylAc expressed in the plants which translated
into the increased efficacy against BAW found in
2002 (Pierce et al. 1999).
BAW larvae collected from all transgenic cotton
genotypes had significantly lower mean weights as
compared to BAW larvae collected from conven-
tional cotton in 2001 and 2002 (Table 3). In addi-
tion, mean weights of BAW larvae were
significantly lower when collected on cotton con-
taining CrylF as compared to BAW larvae col-
lected from cotton containing only CrylAc. As with
BAW larval numbers (Table 2), mean weights of
BAW larvae collected from cotton containing both
CrylAc and CrylF were not significantly different
from mean weights of BAW larvae collected from
cotton containing only CrylF for both years.
Leaf damage caused by BAW larvae was signif-
icantly lower in all transgenic cotton genotypes
compared to conventional cotton in 2001 (Table
4). As with BAW larval weights (Table 3), larval
damage was significantly lower in cotton contain-
ing CrylF compared to cotton containing only
CrylAc. Furthermore, BAW larval damage was
Florida Entomologist 87(4)
TABLE 3. WEIGHT OF BAW LARVAE 8 DAYS AFTER EGG INOCULATIONS ON VARIOUS TRANSGENIC GENOTYPES IN 2001
AND 2002.
Mean weight of larvae (mg) SEM
Genotype No. weighed 2001 No. weighed 2002
PSC 355 94 17.4 4.07 a 45 7.4 1.40 a
CrylAc 114 6.5 0.71 b 38 2.6 0.36 b
CrylF 19 3.3 0.21 c 16 0.6 0.23 c
CrylAc/CrylF 21 2.3 0.27 c 19 1.2 0.18 c
df 3,12 3,9
F value 43.28 34.93
(P > F) ANOVA <0.001 <0.001
Means in a column followed by the same letter are not significantly different (a = 0.05); Fisher's Protected LSD option of PROC
MIXED (Littell et al. 1996; SAS Institute 2001). Means were log-transformed prior to analysis.
not significantly different among the genotypes
with CrylF alone and CrylF stacked with
CrylAc.
Bioassays
Similar trends in larval survival on the vari-
ous cotton structures from the different geno-
types (Table 1) were found in the laboratory. The
addition of CrylAc to the CrylF genotype did not
significantly reduce BAW or FAW larval survivor-
ship when fed cotton squares compared to
squares containing only CrylF (Table 5). For both
BAW and FAW, only squares that contained
CrylF significantly decreased larval survivorship
compared to conventional cotton. This same trend
was observed for BAW neonates fed cotton termi-
nal leaves (Table 6). Survival of FAW larvae in
2003 was significantly lower when fed squares
and terminal leaves of cotton containing both
CrylAc and CrylF than conventional cotton (note
that CrylAc and CrylF alone genotypes were not
tested in 2003) (Table 7).
TABLE 4. LEAF DAMAGE CAUSED BY BAW LARVAE 9 DAYS
AFTER EGG INOCULATIONS ON VARIOUS TRANS-
GENIC GENOTYPES IN 2001.
No. Mean %
Genotype evaluated leaf damage SEM
PSC 355 91 39.0 1.79 a
CrylAc 95 32.2 3.21 b
CrylF 74 17.4 1.76 c
CrylAc/CrylF 76 18.5 4.06 c
df 3,9
F value 31.65
(P > F) ANOVA <0.001
Means in a column followed by the same letter are not sig-
nificantly different (a = 0.05); Fisher's Protected LSD option of
PROC MIXED (Littell et al. 1996;SAS Institute 2001).
The CrylF transgene apparently contributed
more to total toxicity than the crylAc transgene
in the CrylAc/CrylF genotype (=WideStrike@)
against BAW and FAW larvae. Luo et al. (1999)
used a diet bioassay containing purified CrylAc
or CrylF and showed that CrylF was >10x more
toxic than CrylAc against BAW and FAW larvae.
In Bollgard II, the cry2Ab transgene is ex-
pressed throughout the plant at much higher lev-
els than CrylAc (Greenplate et al. 2003).
However, on an equal dose basis, Cry2Ab is more
toxic against the soybean looper, Pseudoplusia in-
cludens (Walker), than CrylAc, while CrylAc is
more toxic against the bollworm, H. zea (Sims
1997). Therefore, differences between toxicity of
crylAc and CrylF transgenes to BAW and FAW
could be due to a higher titer of CrylF in the plant
compared to CrylAc, inherent toxicity differences
between the two transgenes, or possibly a combi-
nation of both. However, the presence or absence
of an additive, synergistic, or antagonistic effect
between CrylAc and CrylF was not observed in
TABLE 5. SURVIVAL OF BAW AND FAW LARVAE AT 4 DAE
WHEN FED COTTON SQUARES FROM VARIOUS
TRANSGENIC GENOTYPES IN 2001.
Mean % survival SEM
Genotype BAW FAW
PSC 355 76.7 6.78 a 52.5 4.79 a
CrylAc 70.0 4.88 a 55.0 8.66 a
CrylF 44.2 8.06 b 20.0 10.80 b
CrylAc/CrylF 43.3 7.77 b 25.0 9.57 b
df 3,21 3, 12
F value 9.95 4.32
(P > F) ANOVA <0.001 0.028
Means in a column followed by the same letter are not sig-
nificantly different (a = 0.05); Fisher's Protected LSD option of
PROC MIXED (Littell et al. 1996; SAS Institute 2001).
December 2004
Adamczyk & Gore: Activity of CrylAc and CrylF Against Armyworms
TABLE 6. SURVIVAL OF BAW NEONATES WHEN FED COTTON TERMINAL LEAVES FROM VARIOUS TRANSGENIC GENOTYPES
IN 2002.
Mean % survival SEM
Genotype 3 DAE' 6 DAE 8 DAE
PSC 355 83.3 + 0.06 a 81.7 0.08 a 78.3 0.09 a
CrylAc 90.0 0.04 a 86.7 0.05 a 78.3 0.03 a
CrylF 80.0 + 0.05 a 56.7 + 0.10 b 48.3 0.10 b
CrylAc/CrylF 80.0 0.03 a 55.0 0.02 b 53.3 0.03 b
df 3,9 3,9 3,9
F value 0.92 7.35 6.90
(P > F) ANOVA 0.468 0.010 0.010
Means in a column followed by the same letter are not significantly different (a = 0.05); Fisher's Protected LSD option of PROC
MIXED (Littell et al. 1996; SAS Institute 2001).
'Days after exposure to leaves.
TABLE 7. SURVIVAL OF FAW NEONATES WHEN FED COTTON SQUARES AND TERMINAL LEAVES FROM VARIOUS TRANS-
GENIC GENOTYPES IN 2003.
Mean % survival SEM
Squares Leaves
Genotype 4 DAE1 7 DAE 4 DAE 7 DAE
PSC 355 61.9 + 3.44 a 50.0 4.79 a 75.0 7.14 a 71.9 5.34 a
CrylAc/CrylF 25.0 3.54 b 8.1 + 1.57 b 65.0 5.68 b 25.6 3.71 b
df 1,6 1,3 1,3 1,3
F value 55.84 147.99 19.20 50.70
(P > F) ANOVA <0.001 0.001 0.022 0.006
Means in a column followed by the same letter are not significantly different (a = 0.05); Fisher's Protected LSD option of PROC
MIXED (Littell et al. 1996; SAS Institute 2001).
Days after exposure to squares or leaves.
these field and laboratory experiments. Insuring
that both transgenes provide dual protection
against key lepidopterous pests is crucial for re-
sistance management to transgenic crops (Gould
& Tabashnik 1998).
DISCLAIMER
Mention of trade names or commercial products in
this publication is solely for the purpose of providing
specific information and does not imply recommendation
or endorsement by the U.S. Department of Agriculture.
REFERENCES CITED
ADAMCZYK, J. J., JR., V. J. MASCARENHAS, G. E.
CHURCH, B. R. LEONARD, AND J. B. GRAVES. 1998.
Susceptibility of conventional and transgenic cotton
bolls expressing the Bacillus thuringiensis CrylAc 6-
endotoxin to fall armyworm (Lepidoptera: Noctu-
idae) and beet armyworm (Lepidoptera: Noctuidae)
injury. J. Agric. Entomol. 15: 163-171.
ADAMCZYK, JR., J. J., L. C. ADAMS, AND D. D. HARDEE.
2001. Field efficacy and seasonal expression profiles for
terminal leases of single and double Bt toxin cotton
genotypes. J. Econ. Entomol. 94: 1589-1593. 2001.
ADAMCZYK, JR., J. J., AND D. D. HARDEE. 2002. Insect-
Resistant transgenic Crops, pp. 23-37. In K. Rajas-
ekaren, T. Jacks, and J. Finley [eds.], Genetic Engi-
neering to Improve Attributes of Crop Plants.
American Chemical Society, Washington, DC.
ADAMCZYK, J. J., JR., M. R. WILLIAMS, J. T. REED, D. W.
HUBBARD, AND D. D. HARDEE. 2003. Spatial and
temporal occurrence of beet armyworm (Lepi-
doptera: Noctuidae) moths in Mississippi. Fla. Ento-
mol. 86: 229-232.
GOULD, F. L., AND B. TABASHNIK. 1998. Bt-cotton resis-
tance management. Chapter 4, pp. 65-105 In M. Mel-
lon and J. Rissler [eds.], Now or Never: Serious New
Plans to Save a Natural Pest Control. Union of Con-
cerned Scientists. Cambridge, MA.
GREENPLATE, J. T., J. W. MULLINS, S. R. PENN, A.
DAHM, B. J. REICH, J. A. OSBORN, P. R. RAHN, L.
RUSCHKE, AND Z. W. SHAPPLEY. 2003. Partial char-
acterization of cotton plants expressing two toxin
proteins from Bacillus thuringiensis: relative toxin
contribution, toxin interaction, and resistance man-
agement. J. Appl. Ent. 127: 340-347.
HARDEE, D. D, AND G. A. HERZOG. 1997. 50th Annual
conference report on cotton insect research and con-
trol, pp. 809-818 In Proc. Beltwide Cotton Conf., Na-
tional Cotton Council, Memphis, TN.
HENDRICKS, D. E., D. W. HUBBARD, AND D. D. HARDEE.
1995. Occurrence of beet armyworm moths in the
lower Mississippi river delta as indicated by num-
bers caught in traps in 1994. Southwestern Entomol.
20: 157-164.
HENNEBERRY, T. J., L. FORLOW JECH, AND T. DE LA
TORRE. 2001. Effects oftransgenic cotton on cabbage
looper, tobacco budworm, and beet armyworm (Lepi-
doptera: Noctuidae) larval mortality and develop-
ment and foliage consumption in the laboratory.
Southwestern Entomol. 26: 325-338.
HOOD, E. 1997. The fall armyworm: and I thought I had
it made, pp. 1223-1224. In Proc. Beltwide Cotton
Conf., National Cotton Council, Memphis, TN.
KIM, Y., AND N. KIM. 1997. Cold hardiness in
Spodoptera exigua (Lepidoptera: Noctuidae). Envi-
ron. Entomol. 26: 1117-1123.
LITTELL, R. C., G. A. MILLIKEN, W. W. STROUP, AND R.
D. WOLFINGER. 1996. SAS System for Mixed Models.
SAS Institute, Cary, NC.
LUo, K., D. BANKS, AND M. J. ADANG. Toxicity, binding,
and permeability analyses of four Bacillus thurin-
giensis Cryl 6-Endotoxins using brush border mem-
brane vesicles of Spodoptera exigua and Spodoptera
frugiperda. Applied and Environ. Microbiol. 65: 457-
464.
December 2004
MITCHELL, E. R. 1979. Migration of Spodoptera exigua
and S. frugiperda, North American style, pp. 386-
393. In R. L. Rabb and G. G. Kennedy [eds.] Move-
ment of Highly Mobile Insect: Concepts and Method-
ology in Research. North Carolina State University.
PIERCE, J. P., C. D. ELLERS-KIRK, R. P. FLYNN, AND C.
FRENCH. 1999. Variation in beet armyworm suscep-
tibility and plant resistance in selected Bt cotton va-
rieties. Southwestern Entomol. 24: 123-131.
SAS INSTITUTE. 2001. Proprietary Software Release 8.2,
Cary, NC.
SIMS, S. R. 1997. Host activity spectrum of the CryIIA
Bacillus thuringiensis subsp. kurstaki protein: ef-
fects on Lepidoptera, Diptera, and non-target ar-
thropods. Southwestern Entomol. 22: 395-404.
SMITH, R. H. 1997. An extension entomologist's 1996 ob-
servations of Bollgard (Bt) technology, pp. 856-857.
In Proc. Beltwide Cotton Conf, National Cotton
Council, Memphis, TN.
SPARKS, A. N. 1979. A review of the biology of the fall ar-
myworm. Florida Entomol. 62: 82-87.
STEWART, S. D., J. J. ADAMCZYK, JR., K. S. KNIGHTEN,
AND F. M. DAVIS. 2001. Impact of Bt cottons express-
ing one or two insecticidal proteins of Bacillus thur-
ingiensis Berliner on growth and survival of noctuid
(Lepidoptera) larvae. J. Econ. Entomol. 94: 752-760.
YOUNG, J. R. 1979. Fall armyworm: control with insec-
ticides. Florida Entomol. 62: 130-133.
Florida Entomologist 87(4)
Cook et al.: Performance of Novel Insecticides Against Armyworms
FIELD AND LABORATORY PERFORMANCE OF NOVEL INSECTICIDES
AGAINST ARMYWORMS (LEPIDOPTERA: NOCTUIDAE)
D. R. COOK1, B. R. LEONARD2 AND J. GORE3
'LSU AgCenter, Northeast Research Station, St. Joseph, LA
2LSU AgCenter, Macon Ridge Research Station, Winnsboro, LA
3USDA-ARS SIMRU, Stoneville, MS
ABSTRACT
Beet armyworm, Spodoptera exigua (Htibner), and fall armyworm, Spodoptera frugiperda
(J. E. Smith), are occasional pests of cotton, Gossypium hirsutum (L.), and soybean, Glycine
max (L.) Merrill. These insects can be difficult to control due to insecticide resistance and
larval behavior on plants. The objectives of these studies were to determine the efficacy of se-
lected insecticides against native infestations of beet armyworm in cotton and soybean and
to generate baseline dose-mortality responses for beet armyworm and fall armyworm adults
to indoxacarb and pyridalyl in the adult vial test. Indoxacarb, pyridalyl, spinosad, methox-
yfenozide, and emamectin benzoate controlled beet armyworm infestations up to 10 d after
treatment compared to the non-treated control. Thiodicarb reduced beet armyworm densi-
ties up to three d after treatment. The LC,, values of indoxacarb and pyridalyl for beet ar-
myworm and fall armyworm exceeded the highest concentrations tested (100-200 pg/vial) in
the adult vial test. Dose-mortality values of indoxacarb and pyridalyl were higher than dis-
criminating concentrations of cypermethrin, methomyl, profenofos, and endosulfan used in
the adult vial test for monitoring tobacco budworm, Heliothis virescens (F.), and bollworm,
Helicoverpa zea (Boddie), susceptibility in Louisiana and Texas. These results indicate that
the adult vial test may not be the most efficient test method for indoxacarb and pyridalyl in
insecticide susceptibility monitoring programs.
Key Words: Beet armyworm, Spodoptera exigua, fall armyworm, Spodoptera frugiperda, cot-
ton, Gossypium hirsutum, soybean, Glycine max, insecticides.
RESUME
El gusano trozador, Spodoptera exigua (Htibner), y el gusano cogollero, Spodoptera fru-
giperda (J. E. Smith), son plagas ocasionales de algod6n, Gossypium hirsutum (L.), y soya,
Glycine max (L.) Merrill. Estos insects son dificiles de controlar debido a la resistencia ha-
cia el insecticide y el comportamiento de las larvas en las plants. Los objectives de estos es-
tudios fueron para determinar la eficacia de los insecticides seleccionados contra las
infestaciones nativas del gusano trozador en algod6n y soya y para obtener respuestas de do-
sis-mortalidad basicas para los adults de gusano trozador y de gusano cogollero al indoxa-
carb y pyridalyl en pruebas de adults en viales de prueba. El indoxacarb, pyridalyl,
spinosad, methoxyfenozide, y emamectin benzoate controlaron las infestaciones del gusano
trozador hasta 10 dias despu6s del tratamiento comparado al control sin tratamiento. El
Thiodicarb reduj6 las densidades del gusano trozador hasta 3 dias despu6s del tratamiento.
Los valores del CL,, de indoxacarb y pyridalyl para el gusano trozador y el gusano cogollero
excedieron las concentraciones mas altas probadas (100-200 pg/vial) en la prueba de los
adults en viales de prueba. Los valores de dosis-mortalidad de indoxacarb y pyridalyl fue-
ron mas altas que las concentraciones discriminantes del cypermethrin, methomyl, profeno-
fos, y endosulfan usados en pruebas de adults en viales para monitorear la susceptibilidad
del gusano del brote de tobacco, Heliothis virescens (F.), y el gusano del elote del maiz, Heli-
coverpa zea (Boddie), en Louisiana y Texas. Estos resultados indican que la prueba de los
adults en viales posiblemente no es el m6todo de prueba mis eficaz para indoxacarb y pyri-
dalyl en programs para monitorear la susceptibilidad de insecticides.
Beet armyworm, Spodoptera exigua (Hiibner), and soybean (Baldwin 1994). Beet armyworm can
and fall armyworm, Spodoptera frugiperda (J. E. be difficult to control, and for many years the only
Smith), are occasional pests of cotton, Gossypium effective insecticides were thiodicarb and chlorpy-
hirsutum (L.), and soybean, Glycine max (L.) Mer- rifos. Their performance has varied considerably
rill, in the mid-southern and southeastern United against beet armyworm in the Mid-South and
States. Beet armyworm larvae feed primarily on southeastern United States. Thiodicarb and chlor-
foliage in cotton (Smith 1989, Leser et al. 1996) pyrifos provided >65% control of beet armyworm
Florida Entomologist 87(4)
larvae in cotton in Texas (Smith 1985). In South
Carolina, thiodicarb also provided 90% control of
beet armyworm in cotton (Sullivan et al. 1991).
Reed et al. (1994) reported <50% control of beet
armyworm with both thiodicarb and chlorpyrifos
in Mississippi cotton. In Louisiana, control of beet
armyworm with thiodicarb and chlorpyrifos in
cotton has been inconsistent (Burrs et al. 1994;
Graves et al. 1995; Mascarenhas et al. 1996). Spi-
nosad (Dow Agrosciences, Indianapolis, IN),
indoxacarb (E. I. DuPont de Nemours and Co.,
Wilmington, DE), and pyridalyl (Valent USA
Corp., Walnut Creek, CA) are novel compounds
that have demonstrated efficacy against many
lepidopteran pests of cotton and soybean.
Fall armyworm larvae feed primarily on soy-
bean foliage and are readily exposed to foliar in-
secticide applications (Baldwin 1994). In cotton,
early instar fall armyworms occur in the lower
portion of the plant canopy and feed on foliage
(Ali et al. 1989, 1990). Therefore, control of early
instar fall armyworms in cotton can be difficult
because foliar insecticide applications generally
do not penetrate the canopy sufficiently to reach
the larvae. Older larvae move within the plant
canopy to fruiting structures (Ali et al. 1990).
These larger larvae feed inside fruiting struc-
tures which minimize their exposure to foliar in-
secticide applications. Fall armyworm larvae also
become more tolerant to insecticides as larval size
increases (Yu 1983; Mink & Luttrell 1989) mak-
ing control even more difficult to achieve.
Insecticide resistance in key insect pests has
become a significant problem in crop production.
Surveying insect populations for changes in sus-
ceptibility to insecticides is an integral component
of insecticide resistance management. Monitoring
efforts should be initiated before a compound is
widely used and while the frequency of resistant
individuals is low (ffrench-Constant & Roush
1990). Determining the range of initial resistance
frequencies among insect populations facilitates
early detection of changes in susceptibility to an
insecticide. Therefore, early establishment of re-
sistance baselines are critical for successful imple-
mentation of insecticide resistance management
strategies before field control failures become
widespread. Baseline responses for laboratory
and field strains of insects to novel compounds
should be established to develop discriminating
concentrations for monitoring programs and for
historical reference values. Numerous states have
implemented insecticide resistance monitoring
programs for bollworm, Helicoverpa zea (Boddie),
and tobacco budworm, Heliothis virescens (F.), in
cotton. However, coordinated insecticide resis-
tance monitoring programs in cotton have not
been developed for beet armyworm and fall army-
worm in the United States due to the sporadic oc-
currence of these pests in the Mid-South and
southeastern United States.
The objectives of these studies were to evalu-
ate the efficacy of selected insecticides against na-
tive infestations of beet armyworm in cotton and
soybean, and to generate baseline dose-mortality
responses for beet armyworm and fall armyworm
adults to indoxacarb and pyridalyl in the adult
vial test. These data will support insecticide rec-
ommendations and provide reference dose-mor-
tality data for future monitoring programs.
MATERIALS AND METHODS
Field Experiments
Field trials were conducted during 1998 and
2000 at the LSU Ag Center Macon Ridge Research
Station (Franklin Parish, LA). Trials 1998 and
2000-B were conducted in cotton, while trial 2000-
A was conducted in soybean. Plots were planted to
the cotton varieties 'Stoneville LA 887' (Stoneville
Pedigree Seed Co., Memphis, TN) and 'Phytogen
355' (Dow AgroSciences LLC, Indianapolis, IN) in
1998 and trial 2000-B, respectively. The soybean
variety 'Pioneer 9631' (Pioneer Hi-Bred Interna-
tional, Inc., Des Moines, IA) was used in trial
2000-A. Plots were planted on 11 June 1998, 30
May in trial 2000-A, and on 28 June in trial 2000-
B. Plots consisted of four rows on 1-m centers and
15.2 m long. Treatments were arranged in a ran-
domized complete block design with four replica-
tions. Cultural practices recommended by the
LSU AgCenter were followed to maintain plots in
a consistent manner within each trial.
Insecticide treatments included the following:
emamectin benzoate (Denim 0.16 Emulsifiable
Concentrate (EC), 2.15% ai wt/wt, Syngenta Crop
Protection, Greensboro, NC), indoxacarb (Stew-
ard 1.25 Suspension Concentrate (SC), 14.5% ai
wt/wt, E. I. Du Pont de Nemours and Company,
Wilmington, DE), methoxyfenozide (Intrepid 80
Wettable Powder (WP), 80% ai wt/wt, Dow Agro-
Sciences LLC, Indianapolis, IN), pyridalyl (S-
1812 4EC, 45% ai wt/wt, Valent USA Corporation,
Walnut Creek, CA), Spinosad (Tracer 4SC, Dow
AgroSciences LLC, Indianapolis, IN), and thiodi-
carb (Larvin 3.2 Flowable (F), 34% ai wt/wt,
Bayer CropScience, Research Triangle Park, NC).
In 1998, treatments were applied on 14 and 17
August with a high-clearance sprayer and a CO2-
charged spray system calibrated to deliver 56.1 L
per ha through TX-8 hollow cone nozzles (Spray-
ing Systems Company, Wheaton, IL) (two per
row) at 338 kPa. Treatments were applied on 11
and 14 Aug in trial 2000-A and trial 2000-B, re-
spectively, with a high-clearance sprayer and a
CO2-charged spray system calibrated to deliver
56.1 L per ha through TX-8 hollow cone nozzles
(two per row) at 359 kPa.
Treatment efficacy was determined 10 d after
treatment (DAT), three and seven DAT, and two
and seven DAT, respectively, in trials 1998, 2000-
December 2004
Cook et al.: Performance of Novel Insecticides Against Armyworms
A, and 2000-B. Larval density data were collected
with a standard (38.1 cm) sweep net (25 sweeps
per plot). Data for each trial were subjected to
analysis of variance procedures and means sepa-
rated according to Fisher's Protected Least Sig-
nificant Difference (SAS Institute, 1990).
Laboratory Experiments
Insects tested were obtained from susceptible
laboratory colonies maintained at the Louisiana
State University Department of Entomology, Ba-
ton Rouge, LA. The beet armyworm colony was
obtained from Ecogen, Inc. (Langhorne, PA) dur-
ing 1994. This colony was originally established
at the USDA-ARS Southern Insect Management
Laboratory at Stoneville, MS before 1983. The fall
armyworm colony was established in 1997 from
collections in field corn, Zea mays L., and supple-
mented with additional individuals collected from
field corn in 1999.
Larvae were fed an artificial wheat-germ and
soybean protein diet described by King & Hartley
(1985). Rearing conditions consisted of a 14:10
light-dark photoperiod, 23.9 to 29.4C, and 80%
relative humidity.
Samples of technical grade indoxacarb (E.I.
DuPont de Nemours and Company, Wilmington,
DE), pyridalyl (Valent USA Corporation, Walnut
Creek, CA), spinosad (Dow AgroSciences, India-
napolis, IN), and cypermethrin (Chem Service,
West Chester, PA) were used in adult vial tests.
Procedures similar to those described by Plapp et
al. (1987) for the adult vial test were used to eval-
uate the toxicity of indoxacarb, pyridalyl, and spi-
nosad to beet armyworm and fall armyworm.
Stock solutions of each compound were developed
by dissolving technical grade insecticide in ace-
tone. Dilutions from each stock solution were
used to yield the desired concentrations. The inte-
rior surface of20-ml scintillation vials was coated
with insecticide by pipetting 0.5 ml of the appro-
priate insecticide solution into the vials. These vi-
als were placed on a modified hot dog roller
(heating element disconnected) until all of the ac-
etone had evaporated. Vials were stored in a dark
environment at ambient temperature (approxi-
mately 23.9C) no longer than 21 d before being
used in assays. All assays were conducted at am-
bient temperature (approximately 23.9C).
Washed (clean) non-treated vials were used as
controls. Previous tests (J. B. Graves & B. R. Le-
onard, unpublished data) indicated no differences
in mortality between washed non-treated vials
and vials treated with acetone only.
Prior to testing, insects were segregated by sex
based on dimorphic pupal characters. Pupae were
placed into 3.78-L cardboard cartons containing a
thin layer of vermiculite on the bottom. Newly
closed adults were removed daily and placed into
polypropylene cages (29.97 x 29.97 x 29.97 cm)
(BugDorm, Megaview Science Education Services
CO. Ltd., Taichung, Taiwan). Male and female
moths were held in separate cages for ca. 24 h and
provided 10% sugar water as a food source. Moths
were placed into insecticide treated vials or con-
trol vials (1 moth per vial) and mortality was de-
termined after 24 h of exposure. Moths were
considered dead if they were incapable of sus-
tained flight for at least 1.0 meter. The number
and range of concentrations for each colony and
compound combination are detailed in Table 1.
Data were corrected for mortality in control vials
(Abbott 1925) and analyzed by probit analysis
with Polo PC (LeOra Software, Berkeley, CA). Dif-
ferences were considered to be significant based
upon non-overlap of the 95% confidence limits.
Separate data analyses were conducted for males
and females of the respective insect species for
each compound. Data for males and females of the
respective insect species for individual com-
pounds were pooled when no significant differ-
ences were detected between sexes or LC,0 or LC90
values exceeded the highest concentration tested.
Data from adult vial tests are reported as pg of in-
secticide per vial.
TABLE 1. RANGE OF CONCENTRATIONS OF INDOXACARB, PYRIDALYL, SPINOSAD, AND CYPERMETHRIN TESTED AGAINST
LABORATORY COLONIES OF INSECTS IN THE ADULT VIAL TEST.
Concentration range
Compound Insect species No. of concentrations (pg/vial)
Indoxacarb Beet amnyworm 7 10-200
Fall armyworm 4 25-100
Pyridalyl Beet armyworm 7 10-200
Fall armyworm 4 25-100
Spinosad Beet armyworm 15 1-100
Fall armyworm 10 5-100
Cypermethrin Beet armyworm 7 0.5-100
Fall armyworm 5 5-100
Florida Entomologist 87(4)
RESULTS AND DISCUSSION
Field Experiments
In trial 1998 (cotton), all insecticide treat-
ments significantly reduced beet armyworm den-
sities compared to the non-treated control at 10
DAT (Table 2). Beet armyworm densities were re-
duced by 5.6-fold, 5.7-fold, and 21.2-fold in plots
treated with spinosad, indoxacarb, or pyridalyl,
respectively, compared to those observed in the
non-treated plots. In trial 2000-A (soybean), all
insecticide treated plots had significantly lower
densities of beet armyworm larvae compared to
the non-treated plots at three DAT (Table 3).
Plots treated with methoxyfenozide or emamec-
tin benzoate had significantly fewer larvae com-
pared to plots treated with indoxacarb, spinosad,
or thiodicarb. Beet armyworm densities were 5.5-
fold, 6.5-fold, and 17.9-fold lower in the indoxac-
arb, emamectin benzoate, and methoxyfenozide
treated plots, respectively, compared to beet ar-
myworm densities in the non-treated plots. At
seven DAT, all insecticide treatments, except
thiodicarb, significantly reduced beet armyworm
densities compared to those observed in the non-
treated control. In treated plots spinosad reduced
beet armyworm densities 2.6-fold and indoxacarb
reduced beet armyworms 4.8-fold compared to
non-treated plots. No larvae were collected in
plots treated with methoxyfenozide. In trial
2000-B (cotton), all insecticide treatments signif-
icantly reduced beet armyworm densities com-
pared to those observed in the non-treated
control at two DAT (Table 4). Plots treated with
indoxacarb, spinosad, or emamectin benzoate
had significantly fewer larvae than plots treated
with methoxyfenozide. Beet armyworm densities
in the spinosad and indoxacarb treated plots
were 10.5-fold and 6.5-fold lower, respectively,
compared to those in the non-treated plots. At
seven DAT, all insecticide treatments signifi-
cantly reduced larval densities compared to the
non-treated control. Beet armyworm densities in
the spinosad, indoxacarb, and methoxyfenozide
treated plots were 30.2- fold, 11.9-fold, and 49.1-
fold lower, respectively, compared to those in the
non-treated plots.
Results of these studies are similar to those
from Fitzpatrick et al. (1996); Teran-Vargas et al.
(1997); Gore et al. (1999); Torrey et al. (1999) in
which indoxacarb, spinosad, methoxyfenozide,
and emamectin benzoate provided excellent con-
trol of beet armyworm infestations. Thiodicarb
significantly reduced beet armyworm larval densi-
ties compared to the non-treated control at three
DAT. At seven DAT, however, larval densities in
the thiodicarb treated plots were not significantly
different from those in the non-treated plots.
These results for thiodicarb are similar to those
reported by Mascarenhas et al. (1996) in which
the performance of thiodicarb was inconsistent.
Thiodicarb is no longer recommended for beet ar-
myworm control in cotton in Mississippi and Lou-
isiana (Bagwell et al. 2003; Layton 2004). The
inconsistent performance of thiodicarb in these
studies further supports its removal from insecti-
cide recommendations for control of beet army-
worm in cotton in Louisiana. Thiodicarb is
recommended for use in soybeans against beet ar-
myworm in Mississippi and Louisiana (Anony-
mous 2003; Baldwin et al. 2003) and for use in
cotton and soybeans against beet armyworm in
Arkansas (Johnson et al. 2002; Lorenz et al. 2002).
Laboratory Experiments
The LC,0 values of indoxacarb and pyridalyl ex-
ceeded the highest concentration tested (200 pg
per vial for indoxacarb and pyridalyl) for beet ar-
myworm adults (Table 5). The LC,0 values of spi-
nosad (45.6 pg per vial) and cypermethrin (37.1 pg
per vial) were not significantly different from each
other. The LC,0 values of indoxacarb, pyridalyl,
spinosad, and cypermethrin for beet armyworm
adults exceeded the highest concentrations tested
TABLE 2. EFFICACY OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM IN COTTON, 1998.
No. beet armyworm larvae per 25 sweeps
Treatment Rate per ha (kg AI) 10 DAT ( SE)
Indoxacarb 0.101 6.3 b + 0.9
Pyridalyl 0.14 1.8 b 1.7
Spinosad 0.073 6.8 b 1.6
Emamectin Benzoate 0.011 4.0 b 3.3
Non-Treated 38.3 a 5.2
F 87.1
df 4,12
P>F <0.01
Means followed by a common letter are not significantly different (P < 0.05 Fisher's Protected Least Significant Difference).
December 2004
Cook et al.: Performance of Novel Insecticides Against Armyworms
TABLE 3. EFFICACY OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM IN SOYBEAN (TRIAL 2000-A).
No. beet armyworm larvae per 25 sweeps
Treatment Rate per ha (kg AI) 3 DAT ( SE) 7 DAT ( SE)
Indoxacarb 0.101 5.2 bc 3.0 2.4 bc 1.5
Spinosad 0.045 8.0 bc 1.9 4.4 b 2.1
Methoxyfenozide 0.224 1.6 c 1.5 0.0 c 0.0
Thiodicarb 0.504 12.2 b 10.3 8.4 a 4.1
Emamectin Benzoate 0.011 4.4 c 2.1 2.2 be 1.5
Non-Treated 28.6 a 11.1 11.4 a 6.2
F 14.1 10.3
df 5,20 5,20
P>F <0.01 <0.01
Means within columns followed by a common letter are not significantly different (P > 0.05 Fisher's Protected Least Significant
Difference).
(200 pg per vial for indoxacarb and pyridalyl, 100
pg per vial for spinosad and cypermethrin). The
LC50 values of indoxacarb and pyridalyl for fall ar-
myworm adults exceeded 100 pg per vial (highest
concentration tested) (Table 6). The LC,0 value of
cypermethrin (31.0 pg per vial) was significantly
lower than that of spinosad (69.3 pg per vial) for
fall armyworm adults. The LC,0 values of indoxa-
carb, pyridalyl, spinosad, and cypermethrin for
fall armyworm adults exceeded 100 pg per vial
(highest concentration tested).
In these studies, the LC50 values of indoxacarb
and pyridalyl for beet armyworm and fall army-
worm from laboratory colonies exceeded 100 pg
per vial. These values were significantly higher
compared to the discriminating concentrations of
cypermethrin (5-10 pg per vial) (Plapp et al. 1987;
Graves et al. 1989), methomyl (2.5-10 pg per vial),
profenofos (10-40 pg per vial), and endosulfan (3-
10 pg per vial) (Kanga et al. 1995; Graves et al.
1994) for tobacco budworm and bollworm. Andal-
oro et al. (2000) reported LC50 values >100 ppm
for bollworm, tobacco budworm, and beet army-
worm larvae exposed to glass surfaces treated
with indoxacarb indicating that contact exposure
to residues is not a primary route of intoxication
for indoxacarb. Additionally, Wing et al. (2000) re-
ported that indoxacarb is inactive and is metabol-
ically activated into toxic metabolites. These
metabolites are extremely active and block so-
dium channels in the insect nervous system. In-
formation regarding the route of intoxication of
pyridalyl has not been released.
These data compare the relative toxicity of in-
doxacarb and pyridalyl to that of other common
insecticides against two Spodoptera species and
comprise initial efforts to develop baseline data.
These data also demonstrate that the adult vial
test is not an efficient test procedure for use with
indoxacarb and pyridalyl in resistance monitor-
ing efforts, as opposed to pyrethroids and spi-
nosad, which generally perform well in the adult
vial test. These studies indicate that discriminat-
ing concentrations for indoxacarb and pyridalyl
for use in the adult vial test would be extremely
high. Coordinated resistance monitoring efforts
TABLE 4. EFFICACY OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM IN COTTON (TRIAL 2000-B).
No. beet armyworm larvae per 25 sweeps
Treatment Rate per ha (kg AI) 2 DAT ( SE) 7 DAT ( SE)
Indoxacarb 0.101 6.5 c 3.4 3.3 b + 3.3
Spinosad 0.101 4.0 c 3.6 1.3 b + 1.3
Methoxyfenozide 0.168 23.5 b 8.3 0.8 b 1.0
Emamectin Benzoate 0.008 8.0 c 1.6 5.8b 2.5
Non-Treated 42.0 a 18.3 39.3 a 10.5
F 13.2 43.1
df 4,12 4,12
P>F <0.01 <0.01
Means within columns followed by a common letter are not significantly different (P > 0.05 Fisher's Protected Least Significant
Difference).
Florida Entomologist 87(4)
December 2004
TABLE 5. RESPONSES OF LABORATORY REARED BEET ARMYWORM ADULTS TO INDOXACARB, PYRIDALYL, SPINOSAD, AND
CYPERMETHRIN IN THE ADULT VIAL TEST.
Regression
N Slope SE LC,, 95% C.L. LC,, 95% C.L. X2, df equations
Indoxacarb 350 0.72 0.22 >2001 NA2 >2001 NA2 4.08,5 Y = 0.72x + -0.21
Pyridalyl 350 0.10 0.25 >2003 NA2 >2003 NA2 21.15,56 Y = 0.10x + -1.87
Spinosad 893 1.87 0.17 45.6 39.9-52.5 >1004 NA2 10.06,136 Y = 1.87x + -3.11
Cypermethrin 176 1.14 0.21 37.1 17.7-124.5 >1005 NA2 8.52,5 Y = 1.14x + -1.79
Concentrations expressed in pg insecticide per vial.
Values exceeded 200 pg per vial (highest concentration tested), 200 pg per vial concentration resulted in 39.0% mortality.
'Confidence limits could not be calculated.
Values exceeded 200 pg per vial (highest concentration tested), 200 pg per vial concentration resulted in 16.0% mortality.
Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 68.3% mortality.
5Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 65.0% mortality.
'Significant %' (P = 0.05).
TABLE 6. RESPONSES OF LABORATORY REARED FALL ARMYWORM ADULTS TO INDOXACARB, PYRIDALYL, SPINOSAD, AND
CYPERMETHRIN IN THE ADULT VIAL TEST.
Regression
N Slope SE LC5s 95% C.L. LC90 95% C.L. X2, df equations
Indoxacarb 329 0.63 0.39 >100' NA2 >100' NA2 5.99,2 Y = 0.63x + -2.10
Pyridalyl 166 2.26 0.72 >1003 NA2 >1003 NA2 3.57,2 Y = 2.26x + -5.17
Spinosad 859 1.70 0.22 69.3 44.8-134.0 NA4 NA2 35.79,86 Y = 1.70x + -3.13
Cypermethrin 174 2.09 0.35 31.0 24.6-42.4 NA5 NA2 2.45,3 Y = 2.09x + -3.11
Concentrations expressed in pg insecticide per vial.
Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 27.3% mortality.
'Confidence limits could not be calculated.
Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 25.7% mortality.
Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 70.0% mortality.
5Values exceeded 100 pg per vial (highest concentration tested), 100 pg per vial concentration resulted in 88.9% mortality.
'Significant %' (P = 0.05).
generally test hundreds to thousands of insects of
a particular species annually. The high discrimi-
nating concentrations of indoxacarb and pyrida-
lyl in the adult vial test would dramatically
increase the cost of monitoring efforts.
ACKNOWLEDGMENTS
The authors thank Andrea Huval, Lauren Ca-
vanaugh, Jana Smith, Josh Temple, and Rhett Gable for
their assistance with these studies. Also, the authors
thank the LSU AgCenter, Cotton Incorporated, and
Louisiana's cotton producers for their financial support.
This article was approved for publication by the Direc-
tor of the Louisiana Agricultural Experiment Station as
manuscript No. 04-26-0304.
REFERENCES CITED
ABBOTT, W. S. 1925. A method of computing the effective-
ness of an insecticide. J. Econ. Entomol. 18:265-267.
ALI, A., R. G. LUTTRELL, H. N. PITRE, AND F. M. DAVIS.
1989. Distribution of fall armyworm (Lepidoptera:
Noctuidae) egg masses on cotton. Environ. Entomol.
18: 881-885.
ALI, A., R. G. LUTTRELL, AND H. N. PITRE. 1990. Feeding
sites and distribution of fall armyworm (Lepidop-
tera: Noctuidae) larvae in cotton. Environ. Entomol.
19: 1060-1067.
ANDALORO, J. T., K. D. WING, J. H. GREEN, AND E. B.
LANG. 2000. StewardTM dispersion and cotton leaf in-
teractions: impact on cotton insect pests and safety
to beneficial arthropods, pp. 939-940 In Proc. 2000
Beltwide Cotton Conf., National Cotton Council,
Memphis, TN.
ANONYMOUS. 2003. Soybean Insect Management Guide.
Mississippi State University Cooperative Extension
Service Pub. 17 pp. Mississippi State University.
http://msucares.com/insects/soybeans/pubs/insect
management.pdf.
BAGWELL, R. D., B. R. LEONARD, G. BURRIS, S. STEWART,
J. FAIRCLOTH, S. KELLY, C. PINNELL-ALISON, T. ER-
WIN, M. FARRIS, AND S. MICINSKI. 2003. Cotton Insect
Control 2003. Louisiana Cooperative Extension Ser-
vice Pub. 1083. LSU AgCenter, Baton Rouge, LA. 8 pp.
BALDWIN, J. L. 1994. Armyworms, pp. 38-41 In L. G.
Higley and D. J. Boethel [eds.] Handbook of Soybean
Insect Pests. Entomological Society of America, Lan-
ham, MD.
BALDWIN, J. L., M. BAUR, AND B. FITZPATRICK. 2003.
Control Soybean Insects 2003. Louisiana Coopera-
tive Extension Service Pub. 2211. LSU AgCenter,
Baton Rouge, LA. 8 pp.
BURRIS, E., J. B. GRAVES, B. R. LEONARD, AND C. A.
WHITE. 1994. Beet armyworms (Lepidoptera: Noctu-
Cook et al.: Performance of Novel Insecticides Against Armyworms
idae) in Northeast Louisiana: observations on an un-
common pest. Florida Entomol. 77: 454-459.
FFRENCH-CONSTANT, R. H., AND R. T. ROUSH. 1990. Re-
sistance detection and documentation: the relative
roles of pesticidal and biochemical assays, pp. 4-38
In R. T. Roush and B. E. Tabashnik [eds.], Pesticide
Resistance in Arthropods. Chapman and Hall, New
York, NY.
FITZPATRICK, B. J., R. N. MASCARENHAS, M. L. BOYD, D.
J. BOETHEL, E. BURRIS, AND D. COOK. 1996. Soybean
looper and beet armyworm control on soybean in
Louisiana, 1995. Arthropod Management Tests 21:
289-291.
GORE, J., J. H. FIFE, AND B. R. LEONARD. 1999. Toxicity
of selected insecticides to beet armyworms and soy-
bean loopers in cotton, 1998. Arthropod Manage-
ment Tests 24: F58, http://www.entsoc.org/Protected/
AMT/amt1999/F58.html.
GRAVES, J. B., B. R. LEONARD, AND A. M. PAVLOFF.
1989. An update on pyrethroid resistance in tobacco
budworm and bollworm in Louisiana, pp. 343-346 In
Proceedings 1989 Beltwide Cotton Conf., National
Cotton Council, Memphis, TN.
GRAVES, J. B., B. R. LEONARD, E. BURRIS, S. MICINSKI,
S. H. MARTIN, C. A. WHITE, AND J. L. BALDWIN. 1994.
Status of insecticide resistance in tobacco budworm
and bollworm in Louisiana, pp. 769-774 In Proceed-
ings 1994 Beltwide Cotton Conf., National Cotton
Council, Memphis, TN.
GRAVES, J. B., B. R. LEONARD, AND C. A. WHITE. 1995.
Efficacy of selected commercial and experimental in-
secticides against late season populations of beet ar-
myworm, soybean looper, and tobacco budworm,
1994. Arthropod Management Tests 20: 197-198.
JOHNSON, D. R., G. E. STUDEBAKER, G. LORENZ, AND J.
K. GREENE. 2002. Cotton Insect Management. Uni-
versity of Arkansas Cooperative Extension Service
Pub. FSA-2065. University of Arkansas, Fay-
etteville, AR. 6 pp.
KANGA, L. H. B., F. W. PLAPP, JR., G. W. ELZEN, M. L.
WALL, AND J. D. LOPEZ, JR. 1995. Monitoring for re-
sistance to organophosphorus, carbamate, and cyclo-
diene insecticides in the tobacco budworm adults. J.
Econ. Entomol. 88: 1144-1149.
KING, E. G., AND G. G. HARTLEY. 1985.Diatraea saccha-
ralis, pp. 265-270 In P. Singh and R. F. Moore [eds.],
Handbook of Insect Rearing, Vol. 2. Elsevier, Ams-
terdam.
LAYTON, B. 2004. 2004 Cotton Insect Control Guide. Mis-
sissippi State University Cooperative Extension Ser-
vice Pub. 343. Mississippi State University. 36 pp.
LESER, J. F., M. A. KARNER, C. R. WARD, AND J. K.
WALKER 1996. Insect and mite pest management in
the southwest, pp. 695-739 In E. G. King, J. R. Phil-
lips, and R. J. Coleman [eds.] Cotton Insects and
Mites: Characterization and Management. The Cot-
ton Foundation, Memphis, TN.
LORENZ, G., D. R. JOHNSON, G. E. STUDEBAKER, AND J.
K GREENE. 2002. Control Insects on Soybeans. Uni-
versity of Arkansas Cooperative Extension Service
Pub. FSA-2067. University of Arkansas, Fayette-
ville, AR. 6 pp.
MASCARENHAS, V. J., B. R. LEONARD, E. BURRIS, AND J.
B. GRAVES. 1996. Beet armyworm (Lepidoptera:
Noctuidae) control on cotton in Louisiana. Florida
Entomol. 79: 336-343.
MINK, J. S., AND R. G. LUTTRELL. 1989. Mortality of fall
armyworm (Lepidoptera: Noctuidae) eggs, larvae,
and adults exposed to several insecticides on cotton.
J. Entomol. Sci. 24: 563-571.
PLAPP, F. W., G. M. MCWHORTER, AND W. H. VANCE.
1987. Monitoring for pyrethroid resistance in the to-
bacco budworm in Texas-1986, pp. 324-326 In Pro-
ceedings 1987 Beltwide Cotton Conf., National
Cotton Council, Memphis, TN.
REED, J. T., B. LAYTON, AND C. S. JACKSON. 1994. Eval-
uation of insecticides and insecticide combinations
for beet armyworm control. Arthropod Management
Tests 19: 237-238.
SAS INSTITUTE. 1990. SAS/STAT Users Guide, version
6, SAS Institute, Cary, NC.
SMITH, R. H. 1985. Fall and beet armyworm control, pp.
134-136 In Proceedings 1985 Beltwide Cotton Res.
Prod. Conf. National Cotton Council, Memphis, TN.
SMITH, R. H. 1989. Experiences with beet armyworm
control in cotton in 1988, pp. 273-275 In Proceedings
1989 Beltwide Cotton Res. Prod. Conf. National Cot-
ton Council, Memphis, TN.
SULLIVAN, M. J., S. G. TURNIPSEED, AND T. W. SMITH.
1991. Beet armyworm control in South Carolina, p.
777 In Proceedings 1991 Beltwide Cotton Conf. Na-
tional Cotton Council, Memphis, TN.
TERAN-VARGAS, A. P, E. GARZA-URBINA, C. A. BLANCO-
MONTERO, G. PEREZ-CARMONA, AND J. M. PELLE-
GAUD-RABAGO. 1997. Efficacy of new insecticides to
control beet armyworm in northeastern Mexico, pp.
1030-1031 In Proceedings 1997 Beltwide Cotton
Conf., National Cotton Council, Memphis, TN.
TORREY, K. D., B. R. LEONARD, AND J. GORE. 1999. Con-
trol of soybean looper and beet armyworm, 1998. Ar-
thropod Management Tests 24: F115, http://www.
entsoc.org/Protected/AMT/amtl999/F115.html.
WING, K. D., M. SACHER, Y. KAGAYA, Y. TSURUBUCHI, L.
MULDERIG, M. CONNAIR, AND M. SCHNEE. 2000. Bio-
activation and mode of action of the oxadiazine in-
doxacarb in insects. Crop Protection 19: 537-545.
YU, S. J. 1983. Age variation in insecticide susceptibility
and detoxification capacity of fall armyworm (Lepi-
doptera: Noctuidae) larvae. J. Econ. Entomol. 76:
219-222.
Florida Entomologist 87(4)
December 2004
BEHAVIOR AND DISTRIBUTION OF THE TWO
FALL ARMYWORM HOST STRAINS IN FLORIDA
ROD N. NAGOSHI AND ROBERT L. MEAGHER
Center for Medical, Agricultural and Veterinary Entomology, Agricultural Research Service
U.S. Department of Agriculture, Gainesville, FL 32608
ABSTRACT
Fall armyworm is a significant agricultural pest in the United States, affecting most notably
sweet corn and turf grass. While infesting much of North America, fall armyworms invading
the eastern United States arise from annual migrations of populations wintering in south-
ern Florida. It has long been noted that this seasonal geographical localization represents an
opportunity for controlling this pest prior to its annual migration. However, such efforts
have been hindered by the presence of two genetically distinct but morphologically identical
strains that differ physiologically and behaviorally. The biology of the host strains is poorly
understood and this lack of knowledge precludes accurate predictions of fall armyworm pop-
ulation behavior in the field. This paper reviews recent studies examining strain behavior
and discusses the potential relevance of these results to the development of effective re-
gional management strategies that can be used proactively to mitigate the economic impact
of this pest.
Key Words: Spodoptera frugiperda, area wide management, corn, turf grass.
RESUME
El gusano cogollero es una plaga significativa para la agriculture de los Estados Unidos, que
afecta mas notablemente el maiz dulce y pastos. Mientras que infestan la mayor parte de
America del Norte, los gusanos cogolleros que invaden el este de los Estados Unidos provie-
nen de las emigraciones anuales de poblaciones que pasan el inverno en el sur de Florida.
Por un largo tiempo se ha notado que esta localizaci6n geografica por estaciones climaticas
represent una oportunidad para controlar esta plaga antes de la emigraci6n annual. Sin em-
bargo, estos esfuerzos han sido paralizados por la presencia de dos cepas gen6ticamente dis-
tintas pero morfol6gicamente identicas con una fisiologia y un comportamiento diferentes.
La biologia de las cepas con respect hospederos es pobremente conocida y es esta falta de
conocimiento la que impide la predicci6n precisa del comportamiento de la poblaci6n del gu-
sano cogollero en el campo. Este articulo revisa los studios recientes que han estudiado el
comportamiento de las cepas y discute el potential pertinente de estos resultados al desarro-
1lo de estrategias de manejo regionales efectivos que pueden ser usados pro-activamente
para mitigar el impact econ6mico de esta plaga.
The fall armyworm, Spodoptera frugiperda
(J. E. Smith) is a periodic and significant eco-
nomic pest in most of the continental United
States, capable of causing substantial losses in
maize, sorghum, forage grasses, turf grasses, rice,
cotton, and peanut production (Luginbill 1928;
Sparks 1979). Because fall armyworm do not sur-
vive conditions of prolonged freezing, most of the
infestations in the continental U.S. derive from
annual migrations of populations that over winter
in southern Florida and southern Texas (Barfield
et al. 1980). This localization of winter popula-
tions theoretically provides an opportunity to dra-
matically reduce the migratory population, as
previously noted in a quote from E. F. Knipling:
"The fall armyworm could serve as a model spe-
cies for developing the concept of managing highly
mobile pests by an organized attack on popula-
tions at a strategic time and place for the purpose
of protecting crops in other and perhaps much
larger areas at some later time in the seasonal cy-
cle" (Knipling 1980). Unfortunately, the biological
information necessary to develop an area-wide
management strategy for this pest has been slow
in coming, in large part because two morphologi-
cally identical but physiologically distinct host
strains have complicated efforts to understand
and predict fall armyworm behavior in the field.
This paper reviews recent studies describing new
methods of strain identification that greatly en-
hance our capacity to investigate and understand
fall armyworm population biology. Preliminary
results suggest that at least one strain, whose pri-
mary target is corn, might be particularly amena-
ble to a regional management program.
Characteristics of the Two Host Strains
The existence of two strains was originally pos-
tulated after comparisons of electrophoretic pro-
Nagoshi & Meagher: Fall Armyworm Host Strains
tein variants from the wild identified genetically
distinct subpopulations that were preferentially
associated with either large grasses (designated
corn-strain), such as corn and sorghum, or smaller
grasses (designated rice-strain), such as rice and
bermudagrass (Pashley 1986; Pashley 1988a; Pa-
shley et al. 1985; Pashley et al. 1987a). This host
plant specificity reflects nutritional adaptation, as
rice-strain larvae feeding on corn displayed a
slower rate of weight gain, longer developmental
time, lower pupal weight, and reduced survival
than when reared on bermudagrass (Pashley
1988b; Pashley et al. 1995; Veenstra et al. 1995).
Whitford et al. (1988) also reported reductions in
larval and pupal weight but did not observe differ-
ences in developmental time or survival. The ef-
fect on larval growth rate correlated with higher
levels of mixed-function oxidase (an enzyme fam-
ily involved in detoxification pathways) in the
corn-strain. In contrast, the same set of studies
showed that rearing corn-strain larvae on rice or
bermudagrass had no consistent negative effect
on larval development or fitness.
However, the observed nutritional variations
are unlikely to completely account for the plant
host bias exhibited by the strains. In behavioral
preference tests, first instars of both strains
strongly preferred corn to turf grass (Pashley et al.
1995; unpublished data). In addition, although
both strains developed equally when reared on rice
or turf grass, corn-strain larvae were rarely found
on these plants in the field, present in only 3% of
larvae collected from wild grass (McMichael &
Prowell 1999). In contrast, and despite their adap-
tation to rice and bermudagrass, rice-strain larvae
made up as much as 16% of the samples collected
from corn plants. These results suggest that habi-
tat specificity in strain distribution is probably due
to adult behavior, the most obvious candidate be-
ing ovipositional host choice. However, because in-
consistent results were obtained in the one major
study testing this possibility, the biological basis
for the plant host bias exhibited by the two strains
remains unexplained (Whitford et al. 1988).
An important consideration for the manage-
ment of this pest is that differences were found in
the response of the two strains to chemical and
biological agents. Rice-strain larvae were more
susceptible than the corn-strain to several insec-
ticides, including diazinon and carbaryl, while
the reverse was true for carbofuran (Adamczyk et
al. 1997; Pashley et al. 1987b). Similarly, the rice-
strain was more susceptible than the corn-strain
to transgenic Bacillus thuringiensis Berliner (Bt)
cotton (Adamczyk et al. 1997). In addition, some
bermudagrass cultivars bred for fall armyworm
resistance showed differential effectiveness with
respect to the two strains, with rice-strain larvae
generally able to gain more weight and consume
more plant material than their corn-strain coun-
terparts (Jamjanya et al. 1990; Leuck et al. 1968;
Lynch et al. 1983; Pashley et al. 1987a; Quisen-
berry & Whitford 1988). Clearly strain-identity
must be taken into consideration when evaluat-
ing the effectiveness of new insecticides and "re-
sistant" plant cultivars.
DNA Markers of Strain Identity
The fall armyworm strains are morphologi-
cally identical, making an unambiguous determi-
nation of strain identity difficult and largely
limited to molecular methods. Restriction Frag-
ment Length Polymorphisms (RFLPs) were iden-
tified in genomic DNA and formed patterns that
could be segregated into two distinct groups gen-
erally consistent with the rice-strain and corn-
strain populations derived from allozyme compar-
isons (Lu et al. 1992). Similarly, dendrograms
produced by amplified fragment-length polymor-
phism analysis revealed two assemblages that
were over 90% consistent with strain assignments
based on host plant (McMichael & Prowell 1999).
The same two groups could be identified by com-
paring variations in mitochondrial DNA (mtDNA)
sequences, which was modestly more accurate
than allozyme analysis at distinguishing strains
(Lu & Adang 1996; Pashley 1989). In particular,
an Mspl restriction enzyme polymorphism was
identified that was diagnostic of strain identity
and for which a PCR-based detection method was
developed (Levy et al. 2002; Lu & Adang 1996).
These techniques allowed detection of the strain-
specific RFLP from individuals exposed to outdoor
conditions for up to two weeks after death
(Meagher & Gallo-Meagher 2003; Fig. la).
The potential usefulness of this methodology
to assay populations was demonstrated by the ex-
amination of adult males captured in pheromone
traps. Specimens from cornfields during the
spring growing season were tested by PCR with
72% (21/29) shown to carry the corn-strain
marker (mtC). This contrasted with 39% (15/39)
when the same traps were tested a few weeks af-
ter harvest. We compared these findings to collec-
tions made during the same time periods from
traps placed in a pasture habitat containing pri-
marily small grass species. During the pre-har-
vest and post-harvest periods, over 90% (18/19
and 17/18, respectively) carried the rice-strain
(mtR) marker. Therefore, the distribution of the
COI polymorphism in adult males correlates with
the expected behavior of fall armyworm strains
with respect to the local plant population.
Another genetic marker for strain identity is
FR (for Fall armyworm Rice strain), a tandem-re-
peat sequence present in large clusters only in the
rice-strain genome (Lu et al. 1994). We developed
a PCR-based method for detecting FR sequences
that allowed analysis from single individuals
(Nagoshi & Meagher 2003b). When FR clusters
are present in the template DNA, PCR amplifica-
Florida Entomologist 87(4)
tion produces a "ladder" of
from the synthesis of different
peated sequence, a consequen
peat organization that all
amplification alternatives (F
son, amplification of genomic
strain produces between 0-3 b
absence of large clusters. Thr
netic crosses we unambiguou
ters to the sex chromosomes (
2003a). This was consistent w
sexual dimorphism in the nun
present in the rice-strain, sug
magnitude more copies on ti
chromosome (Lu et al. 1994).
Interstrain Mating Behaviour
The persistence of geneti
differences between the host
gests barriers to matings bet
ever, a comparison of stra
allozymes and mtDNA polym
that interstrain hybridizatio
wild populations (Prowell 199
A. Mitochondnal CO gene
Fig. 1. Agarose gels displaying
specific DNA markers used to
strains. Genomic DNA from in
were individually amplified by P(
mitochondrial COI gene was am
with Mspl, a single band (569 bp
of the strain-specific Mspl site,
rice-strain (mt) (R). Two smaller
bp) are produced if the MspI site
corn-strain (mtc) (C) identity. (b).
viduals and one corn-strain indivi
PCR for the presence of FR dust
duces a DNA ladder with an up
smear in FR' specimens and 0-3
samples. The gels were stained
and photographed under ultravic
are in base pairs.
fragments resulting of individuals carrying an allozyme marker of one
t multiples of the re- strain had an mtDNA genotype of the other. One
ice of the tandem re- laboratory study indicated that this interbreeding
ows a variety of may be limited in nature (Pashley & Martin
ig. Ib). In compari- 1987). When corn-strain females were mated to
DNA from the corn- rice-strain males (C X R), no progeny were pro-
)ands, indicating the duced and no spermatophores were transferred.
rough a series of ge- In contrast, the reciprocal mating of rice-strain fe-
sly mapped FR clus- males to corn-strain males (R X C) had fertility
Nagoshi & Meagher equal to control (within-strain) crosses. However,
ith earlier reports of the hybrid R X C daughters produced failed to
nbers of FR clusters mate with males from either strain but were able
ggestive of a order of to mate with their hybrid brothers, although with
he Y than on the X reduced fertility. In comparison, R X C hybrid
males could fertilize females of either strain, al-
though again fertility was somewhat reduced.
These results suggest significant strain-specific
mate selection, such that corn-strain females
c and physiological have a strong preference to males of the same
strains strongly sug- strain or to hybrids, while rice-strain females are
;ween strains. How- more promiscuous. However, this interpretation
in-specific esterase must be tempered by the failure of two subse-
orphisms suggested quent studies to find similar directional and re-
*n was occurring in stricted interstrain mating behavior (Quisenberry
)8). Between 11-16% 1991; Whitford et al. 1988). Instead, normal fertil-
ity was observed in both directions of interstrain
crosses. We were also able to obtain fertile prog-
R R R C eny from the mating of corn-strain females to rice-
strain males in the laboratory, with no obvious dif-
ferences in fecundity from within-strain crosses
(unpublished results). The discrepancy between
the mating experiments performed by different
laboratories is unexplained, but suggests that lab-
oratory culturing and conditions may easily con-
-loo found strain-specific mate selection.
Support for the existence of an assortative
S mating mechanism came from field studies in
-0 which virgin females of each strain were used to
-0oo attract and capture males in the wild. Males of
-300 both strains exhibited a (60-75%) preference to fe-
males of the same strain, suggesting that phero-
-200 mone differences might have a role in mate choice
(Pashley 1993; Pashley et al. 1992). However, a
more substantial difference was observed in stud-
B. FRelement ies examining the temporal partitioning of noc-
turnal mating activities (Pashley et al. 1992).
the diagnostic strain- Corn-strain females began calling (releasing
distinguish the host pheromone) earlier in the scotophase than the
Individual adult males
R. (a). A portion of the rice-strain. Even more significant was the obser-
plified. After digestion ovation that strain-specific matings occurred at op-
) indicates the absence posite times of the night with little overlap. The
a characteristic of the corn-strain mated during the first two-thirds of
bands (497 bp and 72 the evening while the rice-strain mated in the
is present, indicating a last third. Hence assortative mating might reflect
Three rice-strain indi- divergence in the timing of strain-specific mating
idual were analyzed by activity, with additional contributions coming
ers. PCR analysis pro- from differences in pheromone attraction.
pper molecular weight
distinct bands in FRO To measure the degree to which interstrain
vith ethidium bromide matings occur in the wild, we made use of the fact
let illumination. Sizes that our two strain-specific genetic markers un-
dergo different but predictable inheritance pat-
December 2004
Nagoshi & Meagher: Fall Armyworm Host Strains
terns. We designated the parental rice-stain
marker configuration as mt FR' and the corn-
strain as mtc FRO. With interstrain crosses, the
mitochondrial marker should display a strictly
maternal inheritance pattern while FR will ex-
hibit sex linkage. This was confirmed by genetic
crosses performed in the laboratory in which each
set of parents was sacrificed and tested for their
molecular genotypes after mating and oviposition
(Nagoshi & Meagher 2003a). As predicted the mtc
FR' hybrid configuration was produced in males
from a single cross between strains, while the re-
ciprocal mt FROhybrid combination was produced
in males in two generations (Fig. 2). Therefore, if
interstrain matings are common in the wild and
there are no substantial fitness differences be-
tween genotypes, a random sampling of field-col-
lected males should show substantial numbers of
all four combinations. In particular, the number of
mtc FR' should approximate the number of mtc
FRO males. In contrast, if mating within the pa-
rental strains is highly preferred or substantially
more productive, then the parental marker config-
urations (mtc FRO and mtR FR') should predomi-
nate. After a random sampling of field collected
males from several southern Florida locations, we
found 56% (199/356) of mt individuals had the
hybrid configuration (mtR FRO), consistent with
substantial R X C interstrain matings (Nagoshi &
Meagher 2003a). In contrast, only 3% (4/144) of
RXC
mtc males were FR', the reciprocal hybrid pat-
tern. These data strongly suggests that produc-
tive matings between corn-strain females and
rice-strain males rarely occur in the wild, pre-
venting the introduction ofFR carrying X chromo-
somes into a corn-strain maternal background.
If the mtR FRO field population is derived from
interstrain hybridization, then we might expect
to see some deviation in plant host preference
from their "pure strain" mt FR' siblings. To ex-
amine this possibility, we compared the propor-
tion of mt FRO individuals in the rice-strain pool
collected from different habitats (Nagoshi &
Meagher 2003a). A "primarily corn habitat" was
arbitrarily defined as one where in which over
70% of the fall armyworm collected were mtc, in-
dicating that the plants in this time and area
were attractive primarily to the corn-strain. Sim-
ilarly, traps containing over 70% mt males were
defined as "primarily grass habitats." The re-
mainder formed the "mixed habitat" category. We
found that the proportion of FRO males within the
rice-strain pool increased to 83% (15/18) of the to-
tal rice-strain population trapped in corn habi-
tats, compared to 61% (39/64) in areas biased to
the rice-strain and 51% (23/45) in mixed habitats.
Although the differences are only of borderline
statistical significance, the trend is consistent
with mtR FRO males being less specific in their
plant host preference than mt FR' males.
CXR
R-stramn
p X -FR tR
Y-FR9 mtR X
C-strail
X-FRO C
X-FRo mt1
F l X-FRO0 m X-FROR
Y-FR+ 9mtR X X-FRO Omt1'
C-strain
P X-FR 0
Y-FR0 r mtC
X-FR+ O
F I Y-FR0 mtC
R-strami
X X-FR+ Omt
X-FR+ dmtR
FR+ mtc males
S X-FRO ()
F2 Y-FR+ f mtR
X-FRo
X-FRo UmtR
FRO mtR males
Fig. 2. Fall armyworm interstrain crosses designed to produce progeny with a hybrid marker configuration (Na-
goshi & Meagher 2003b). FR clusters are on the sex chromosomes (X and Y) while the mitochondrial haplotype is
maternally inherited. In Lepidopterans the female is the heterogametic sex. In the R X C mating, rice-strain fe-
males (mt FR') are mated to corn-strain males (mt' FRO). The hybrid female progeny are backcrossed to corn-strain
males. All the resulting F2 males will have the hybrid configuration (mtR FRO). The C X R mating represents the re-
ciprocal cross. All F, males will be heterozygous for FR clusters and therefore identified as FR' by PCR. All will also
be mtc because of their corn-strain mother.
Florida Entomologist 87(4)
Undeveloped Areas do not Support High Fall Armyworm
Populations
Critical to the development of an area-wide
management strategy for fall armyworm is an un-
derstanding of the relative contributions of the
various host plants to the overwintering and
migrating populations of each strain (Knipling
1980). While several studies have examined
sweet corn growing agricultural areas in Florida,
other types of environments have not been exam-
ined in detail. To address this issue, we surveyed
four different habitats that reflect some of the
principal environments present in southern Flor-
ida (Meagher & Nagoshi 2004; Nagoshi &
Meagher 2004). These are naturalized grassy
wetlands and three types of developed areas: ag-
ricultural fields, managed turf grass (a sod farm
and golf course), and urban.
We found that traps placed in geographically
dispersed natural areas all showed very low cap-
ture rates throughout the year (Meagher & Na-
goshi 2004). Over the 18-month test period the
average for these traps was 0.4 captures/night/
trap. Aside from unusually high numbers in two
collections in February 2002, the average cap-
tures/night/trap never surpassed 5 for any two-
week collection period (from March 2002 to July
2003). This occurred even though there was a
high density and variety of short and tall grass
species in the areas adjacent to these traps. Ap-
parently these undeveloped habitats do not serve
as high density refuges for fall armyworm at any
time of the year, and are therefore unlikely to con-
tribute substantially to the northward migrating
populations in the spring or be a major source of
14
12
Captures/night/trap 10
(2/2002-7/2003)
R-strain 8
C-strain
6
4
2
(616)
0
the reinfestation of Florida agricultural areas in
the fall and winter.
In contrast, 5-10-fold higher adult captures oc-
curred in traps in a turf grass sod farm (average 8
captures/night/trap), agricultural fields (average
17 captures/night/trap), and urban developments
(average 5 captures/night/trap). These results
suggest that several types of human activities can
lead to increases in the local fall armyworm popu-
lation. It may be that unmanaged habitats are
supportive of natural enemies that effectively con-
trol fall armyworm infestation or that the higher
diversity of plant types in some way inhibits the
establishment of high populations. Alternatively,
developed areas may differ in the type, density, or
quality of plant growth. In any case, identifying
the environmental factors that make certain hab-
itats unattractive to fall armyworm could have
important benefits to the development of new con-
trol methods for this important agricultural pest.
The Corn-Strain is Primarily Found in Agricultural Areas
We used the strain-specific molecular markers
combined with extensive pheromone trapping to
examine the distribution of the strains in the dif-
ferent habitats (Meagher & Nagoshi 2004; Na-
goshi & Meagher 2004). The rice-strain was found
to be present in substantial proportions in all ar-
eas examined. This was particularly the case in
the natural habitats and the sod farm, where over
90% of the captured males were of the rice-strain,
though in the former the overall rate of capture
was low (Fig. 3). In terms of numbers, the largest
contributions came from agricultural areas and,
not surprisingly, the turf grass-rich sod farm.
(335)
-
(500)
_a
natural agricultural sod urban
Fig. 3. Adult males captured per night in pheromone traps in four different habitats in southern Florida from
February 2002 to July 2003. Fall armyworm numbers tested for strain identity at each habitat is in parenthesis.
Data include those reported in Meagher & Nagoshi (2004) and Nagoshi & Meagher (2004).
December 2004
Nagoshi & Meagher: Fall Armyworm Host Strains
By far the highest numbers of corn-strain cap-
tures occurred in the agricultural areas, where an
average of nearly 6 corn-strain males was cap-
tured/night/trap, representing 32% of the total col-
lection population at this site. Surprisingly, we
also found a high proportion of the males trapped
in urban areas were of the corn-strain, represent-
ing about half of the capture population. This was
unexpected given that these sites did not have
plant hosts in the vicinity known to be attractive
to the corn-strain. However, the urban site aver-
aged only a little over 2 corn-strain captures/night/
trap, less than half of that observed in agricultural
areas but still significantly (>95%) greater than
observed in natural areas (Meagher & Nagoshi
2004). In comparison, although the overall fall ar-
myworm population in the sod farm was relatively
high (>8 captures/night/trap; Fig. 3), it contributed
only modest amounts of corn-strain, averaging 1
capture/night/trap. Corn-strain captures in natu-
ral areas were inconsequential.
A critical question with respect to controlling
fall armyworm infestation is to determine where
the build up of strain populations occurs prior to
the annual northward migration. This movement
is at least in part dependent upon the timing of fa-
vorable weather patterns (Luginbill 1928; Mitchell
et al. 1991); hence, a significant reduction in popu-
lation numbers prior to the storm fronts could sub-
stantially reduce fall armyworm infestation in the
eastern U.S. Population surveys in southern Flor-
ida cornfields typically show a rise in the overall
fall armyworm population in the spring, followed
by a rapid and prolonged decline during the sum-
mer months that presumably reflects the north-
ward migration (Mitchell et al. 1991; Nagoshi &
Meagher 2004; Pair et al. 1986). We conducted sur-
veys of the four habitat types in the spring (Febru-
ary to May) of 2002 and 2003, just prior to the early
summer migration period (Meagher & Nagoshi
2004). By far, the highest populations of corn-
strain were found in the agricultural areas, dis-
playing a 3-fold higher trap capture rate than the
next highest (urban) site (Fig. 4a). Natural areas,
the sod farm, and one season's data from a golf
course indicated only minor contributions from
these locales. In comparison, the rice-strain was
found in substantial numbers in both agricultural
and managed turf habitats, consistent with a
broader host range (Fig. 4b). These results indi-
cated that in the weeks prior to migration, the
corn-strain population that will be the source of
much of the corn damaging fall armyworm infesta-
tion in the eastern U.S. was mostly localized to the
agricultural fields of southern Florida.
Comparisons of Seasonal Changes in Strain Populations
Because both strains are detected in regions
where they are unlikely to survive the winter, it is
expected that both migrate. However, given their
different habitat preferences it would not be sur-
prising to find strain-specific variations in migra-
tion behavior. This possibility was first suggested
by studies in Louisiana that found the corn-strain
population, first detected in the spring, reached a
peak in density in early to mid-summer coinci-
dent with the maturation of the local corn crop
(Pashley et al. 1987b). In comparison, the rice-
strain population did not show substantial num-
bers until late summer, a period when the corn-
strain population was nearly absent (Pashley et
al. 1992). These differences suggest that the tim-
ing and/or magnitude of migration may not be the
same for both strains.
If there are habitat-dependent or strain-spe-
cific factors that initiate migration, then we
might expect the two strains to differ in the tim-
ing of their northward movement, which presum-
ably can be detected by the sudden decline in
capture numbers in southern Florida traps. We
tested this by the examination of fall armyworm
populations in agricultural fields predominated
by sweet corn and tomatoes and a sod farm asso-
ciated with turf grass (Meagher & Nagoshi 2004;
Nagoshi & Meagher 2004). These sites were cho-
sen because their respective plant populations al-
low clear predictions about the strain that should
be attracted. Our data indicated that the corn-
and rice-strains showed the same July-October
population trough, although there is some evi-
dence that the rice-strain decline begins earlier
than the corn-strain (Fig. 5, arrows). This was
even the case in comparisons between the rice-
strain population in the sod farm with corn-strain
in the agricultural habitat, indicating this popu-
lation pattern was not dependent on the timing of
sweet corn planting and harvesting. Instead, the
decline in population appears to be due to some
more general environmental or biological condi-
tion. Previous studies have attributed similar
changes in the fall armyworm population to vari-
ations in plant quality and quantity resulting
from the wet-dry seasonal cycle characteristic of
tropical areas (Pair et al. 1986). For example, in
studies performed in Mexico, high capture rates
tended to occur 60-90 days after rainfall peaks,
while intervals of least capture most frequently
occurred 60-90 days after periods of least rainfall
(Raulston et al. 1986). Alternatively, sharp de-
clines in capture numbers during the year may be
related to extremes in the daily minimum tem-
perature. We found in our study that fall army-
worm captures for both strains were lowest when
the daily minimum temperature rose above 20C
(July-October) or fell below 5C (January; see Na-
goshi & Meagher 2004). There is precedence for
correlations between daily temperature and field
capture numbers for other insects (Butler et al.
1999; Cammell & Knight 1992; Scott et al. 2000;
Souza & Carvalho 2002). It may be that an impor-
tant fall armyworm behavior, such as mating or
Florida Entomologist 87(4)
M Spring 2002
[ Spring 2003
-RI= M
natural agricultural
B. R-strain
captures/night/trap
urban
natural agricultural sod urban gotf
Habitats examined in the spring (Feb May)
Fig. 4. Adult males captured per night in pheromone traps in four different habitats in southern Florida prior
to the spring migration in 2002 and 2003 for corn-strain (a) and rice-strain (b) moths. Asterisk indicates only spring
2003 data available. Data include those reported in Meagher & Nagoshi (2004) and Nagoshi & Meagher (2004).
flight activity, is affected deleteriously by sea-
sonal temperature extremes.
After the summer decline, fall armyworm pop-
ulations begin increasing in the fall and winter in
agricultural areas, coincident with the late year
corn growing season. The timing of this increase
was shown to correlate with weather and wind
conditions conducive to southward migration,
leading to the suggestion of a north-to-south re-
turn movement prior to the winter freeze (Mitch-
ell et al. 1991; Pair et al. 1986; Pair et al. 1987).
Our studies on strain distributions during this
period led to the surprising observation that the
fall population peak was due entirely to increases
in rice-strain numbers (Nagoshi & Meagher
2004). Despite the presence of extensive sweet
corn plantings in the agricultural trap areas from
October to May, corn-strain numbers did not in-
crease until February (Fig. 5, dashed line). The
same pattern was observed in the fall of 2003 in
the same and in additional agricultural test sites
(Nagoshi & Meagher 2004; unpublished results).
Interestingly, the rice-strain population in the sod
farm site showed a similar population dynamic,
indicating that this behavior is not specific to the
agricultural planting and harvest cycle. Appar-
ently, the presence of its preferred host plant and
environmental conditions conducive to expansion
of the rice-strain population were not sufficient to
stimulate corn-strain increases during the fall.
The reason for this is unknown, but these obser-
vations suggest that if there is a return migration
in the fall, it is rice-strain specific.
Alternatively, there may be environmental fac-
tors that specifically prevent the expansion of ei-
ther a migrant or indigenous corn-strain
population late in the year. An intriguing, but
purely speculative, explanation would be the exist-
ence of a corn-strain specific pathogen or natural
enemy whose numbers increase during the year,
A. C -strain
captures/night/trap
December 2004
Nagoshi & Meagher: Fall Armyworm Host Strains
40
III11llIIIIIllll lllllll li tldlll|| l llllql
30
A 2 0 in /\ A
Captures/night/trap 20
(Agriculture)
A Rice-strain 10
A Corn-strain
0----------Z~6*^^ ^ !_ ---!Z------2-i--!-!-- 2
NC t C A Co In C M C0 rN Go Ca Cf CO
50
40
B.
Captures/night/trap 30 -
(Sod farm)
O Rice-strain
* Corn-strain
04 0 04 0 0 0 0 0 0 0 00 0 0 0M 0M 0
Fig. 5. Distribution of strains collected from pheromone traps in agricultural areas and a turf grass sod farm.
The strain composition of each collection was determined by PCR and used to calculate the average number of each
strain present per night per trap during each collection period. Arrows point to when the spring population begins
to decline (open: rice-strain; filled: corn-strain). Dashed line indicates period of high fall armyworm numbers in the
fall/winter associated with the rice-strain. Data include those reported in Nagoshi & Meagher (2004).
thereby suppressing reestablishment of a late sea-
son corn-strain, but not rice-strain, reinfestation.
A crash in this putative population during the win-
ter would then allow the corn-strain population
peak observed every spring. It is clear that identi-
fying the biological or environmental reasons for
the relative absence of the corn-strain in the fall
could have important ramifications for the devel-
opment of strain-specific suppression methods.
Prospects for the Area-wide Control of Fall Armyworm
Strains
Studies show that weather conditions prior to
migration and changes in agricultural practices in
southern Florida can mitigate significantly the
duration and severity of fall armyworm damage
in the eastern U.S. (Luginbill 1928; Westbrook &
Sparks 1986). These observations suggest that
changes in fall armyworm population dynamics in
the overwintering area can significantly alter the
magnitude of the northward migration. Unfortu-
nately, the large amount of acreage, the broad
host range and high mobility of the pest, the inter-
spersion of urban development in the infested
areas, and the presence of many ecologically-
sensitive habitats make area-wide suppression of
fall armyworm in southern Florida difficult. Cur-
rent technology does not provide an affordable or
environmentally benign method for the complete
suppression of fall armyworm populations under
these conditions. However, our studies show that
the corn-strain has a relatively limited seasonal
and geographical distribution in southern Flor-
ida. It may, therefore, be economically feasible to
focus efforts over a limited time and space just
prior to the normal migration period, with the ob-
jective to substantially reduce the corn-strain mi-
grant population or delay migration such that the
most vulnerable portions of the more northern
corn growing seasons are missed.
Studies on the seasonal and geographical dis-
tribution of fall armyworm strains also have un-
covered possible seasonal or habitat-specific
factors that seem very effective in controlling the
corn-strain. Both strains are largely absent in
naturalized areas despite the presence of grass
species that should be attractive to the rice-strain
and laboratory evidence that corn-strain larvae
are fully capable of normal development on these
substrates. We also observed surprisingly low
numbers of corn-strain males captured in phero-
mone traps in all habitats tested during the fall, a
period of extensive agricultural activity and sub-
stantial increases in the rice-strain capture popu-
lation. The absence of a corn-strain population
peak under conditions favorable to the rice-
strain, and when host plants normally attractive
to the corn-strain are present, suggests the exist-
ence of unknown factors effective in suppressing
population numbers.
Current and ongoing studies on the strains in-
dicate that a regional management program de-
signed to mitigate or delay fall armyworm
migration might be feasible. Still needed are
longer term and more detailed information on the
geographical and seasonal distribution of the fall
armyworm strains in southern Florida, in partic-
ular the timing of the spring and fall population
increases with respect to seasonal and environ-
mental factors. We also need a more extensive un-
derstanding of strain-specific biology, including
the mechanisms of strain-specific mating and ovi-
positional choice, and the effects ofinterstrain hy-
bridization on behavior and physiology (in
particular fertility, migration, plant host choice,
and susceptibility to chemical and biological
agents). While substantial research remains, we
believe that information and techniques are now
available that can address these issues and lead
to the development of an effective area-wide man-
agement strategy.
REFERENCES CITED
ADAMCZYK, J. J., JR., J. W. HOLLOWAY, B. R. LEONARD,
AND J. B. GRAVES. 1997. Susceptibility of fall army-
worm collected from different plant hosts to selected
insecticides and transgenic Bt cotton. J. Cotton Sci.
1: 21-28.
BARFIELD, C. S., J. L. STIMAC, AND M. A. KELLER 1980.
State-of-the-art for predicting damaging infestations
of fall armyworm. Florida Entomol. 63: 364-375.
BUTLER, L., V. KONDO, E. M. BARROWS, AND E. C. TOWN-
SEND. 1999. Effects of weather conditions and trap
types on sampling for richness and abundance of forest
macrolepidoptera. Environ. Entomol. 28: 795-811.
CAMMELL, M. E., AND J. D. KNIGHT. 1992. Effects of cli-
matic change on the population dynamics of crop
pests. Adv. Ecol. Res. 22: 117-162.
JAMJANYA, T., S. S. QUISENBERRY, S. S. CROUGHAN, AND
R. N. STORY. 1990. Comparison ofbermudagrass lines
grown in different cultural conditions and the effect
on screening for fall armyworm (Lepidoptera: Noctu-
idae) resistance. J. Econ. Entomol. 83: 585-590.
KNIPLING, E. F. 1980. Regional management of the fall
armyworm-a realistic approach? Florida Entomol.
63: 468-480.
LEUCK, D. B., C. M. TALIAFERRO, G. W. BURTON, R. L.
BURTON, AND M. C. BOWMAN. 1968. Resistance in
bermudagrass to the fall armyworm. J. Econ. Ento-
mol. 61: 1321-1322.
LEVY, H. C., A. GARCIA-MARUNIAK, AND J. E. MARU-
NIAK. 2002. Strain identification of Spodoptera fru-
December 2004
giperda (Lepidoptera: Noctuidae) insects and cell
line: PCR-RFLP of cytochrome oxidase subunit I
gene. Florida Entomol. 85: 186-190.
LU, Y., AND M. J. ADANG. 1996. Distinguishing fall ar-
myworm (Lepidoptera: Noctuidae) strains using a
diagnostic mitochondrial DNA marker. Florida En-
tomol. 79: 48-55.
LU, Y.-J., M. J. ADANG, D. J. EISENHOUR, AND G. D.
KOCHERT. 1992. RFLP analysis of genetic variation
in North American populations of the fall armyworm
moth Spodoptera frugiperda (Lepidoptera: Noctu-
idae). Molecular Ecol. 1: 199-208.
LU, Y.-J., G. D. KOCHERT, D. J. ISENHOUR, AND M. J.
ADANG. 1994. Molecular characterization of a strain-
specific repeated DNA sequence in the fall army-
worm Spodoptera frugiperda (Lepidoptera: Noctu-
idae). Insect Mol. Biol. 3: 123-30.
LUGINBILL, P. 1928. The Fall Armyworm. USDA Tech.
Bull. 34. 92 pp.
LYNCH, R. E., W. G. MONSON, B. R. WISEMAN, AND G. W.
BURTON. 1983. Bermudagrass resistance to the fall
armyworm (Lepidoptera: Noctuidae). Environ. Ento-
mol. 12: 1837-1840.
MCMICHAEL, M., AND D. P. PROWELL. 1999. Differences
in amplified fragment-length polymorphisms in fall
armyworm (Lepidoptera: Noctuidae) host strains.
Ann. Entomol. Soc. Am. 92: 175-181.
MEAGHER, R. L., JR., AND M. GALLO-MEAGHER 2003.
Identifying host strains of fall armyworm (Lepi-
doptera: Noctuidae) in Florida using mitochondrial
markers. Florida Entomol. 86: 450-455.
MEAGHER, R. L., AND R. N. NAGOSHI. 2004. Population
dynamics and occurrence of Spodoptera frugiperda
host strains in southern Florida. Ecol. Entomol. (In
press).
MITCHELL, E. R., J. N. MCNEIL, J. K. WESTBROOK, J. F.
SILVAIN, B. LALANNE-CASSOU, R. B. CHALFANT, S. D.
PAIR, V. H. WADDILL, A. SOTOMAYOR-RIOS, AND F. I.
PROSHOLD. 1991. Seasonal periodicity of fall army-
worm (Lepidoptera: Noctuidae), in the Caribbean
basin and northward to Canada. J. Entomol. Sci. 26:
39-50.
NAGOSHI, R. N., AND R. MEAGHER 2003a. Fall army-
worm FR sequences map to sex chromosomes and
their distribution in the wild indicate limitations in
interstrain mating. Insect Mol. Biol. 12: 453-458.
NAGOSHI, R. N., AND R. L. MEAGHER 2003b. FR tandem-
repeat sequence in fall armyworm (Lepidoptera:
Noctuidae) host strains. Ann. Entomol. Soc. Am. 96:
329-335.
NAGOSHI, R. N., AND R. L. MEAGHER 2004. Seasonal dis-
tribution of fall armyworm (Lepidoptera: Noctuidae)
host strains in agricultural and turf grass habitats.
Environ. Entomol. (In press).
PAIR, S. D., J. R. RAULSTON, A. N. SPARKS, J. K. WEST-
BROOK, AND G. K. DOUNCE. 1986. Fall armyworm
distribution and population dynamics in the south-
eastern states. Florida Entomol. 69: 468-487.
PAIR, S. D., J. R. RAULSTON, D. R. RUMMEL, J. K. WEST-
BROOK, W. W. WOLF, A. N. SPARKS, AND M. F.
SCHUSTER 1987. Development and production of
corn earworm and fall armyworm in the Texas high
plains: evidence for reverse fall migration. South-
west. Entomol. 12: 89-99.
PASHLEY, D. P. 1986. Host-associated genetic differenti-
ation in fall armyworm (Lepidoptera: Noctuidae): a
sibling species complex? Ann. Entomol. Soc. Am. 79:
898-904.
Florida Entomologist 87(4)
Nagoshi & Meagher: Fall Armyworm Host Strains
PASHLEY, D. P. 1988a. Current status of fall armyworm
host strains. Florida Entomol. 71: 227-234.
PASHLEY, D. P. 1988b. Quantitative genetics, develop-
ment, and physiological adaptation in host strains of
fall armyworm. Evolution 42: 93-102.
PASHLEY, D. P. 1989. Host-associated differentiation in
armyworms (Lepidoptera: Noctuidae): an allozymic
and mitochondrial DNA perspective, pp. 103-114. In H.
D. Loxdale and J. den Hollander [eds.], Electrophoretic
Studies on Agricultural Pests. Clarendon, Oxford.
PASHLEY, D. P. 1993. Causes of host-associated varia-
tion in insect herbivores: an example from fall army-
worm. pp. 351-359. In K. C. Kim and B. A. McPheron
[eds.], Evolution of Insect Pests/Patterns of Varia-
tion. Wiley, NY.
PASHLEY, D. P., AND J. A. MARTIN. 1987. Reproductive
incompatibility between host strains of the fall ar-
myworm (Lepidoptera: Noctuidae). Ann. Entomol.
Soc. Am. 80: 731-733.
PASHLEY, D. P., A. M. HAMMOND, AND T. N. HARDY.
1992. Reproductive isolating mechanisms in fall ar-
myworm host strains (Lepidoptera: Noctuidae). Ann.
Entomol. Soc. Am. 85: 400-405.
PASHLEY, D. P., T. N. HARDY, AND A. M. HAMMOND.
1995. Host effects on developmental and reproduc-
tive traits in fall armyworm strains (Lepidoptera:
Noctuidae). Ann. Entomol. Soc. Am. 88: 748-755.
PASHLEY, D. P., S. J. JOHNSON, AND A. N. SPARKS. 1985.
Genetic population structure of migratory moths:
the fall armyworm (Lepidoptera: Noctuidae). Ann.
Entomol. Soc. Am. 78: 756-762.
PASHLEY, D. P., S. S. QUISENBERRY, AND T. JAMJANYA.
1987a. Impact of fall armyworm (Lepidoptera: Noc-
tuidae) host strains on the evaluation of bermuda-
grass resistance. J. Econ. Entomol. 80: 1127-1130.
PASHLEY, D. P., T. C. SPARKS, S. S. QUISENBERRY, T.
JAMJANYA, AND P. F. DOWD. 1987b. Two fall army-
worm strains feed on corn, rice and bermudagrass.
Louisiana Ag. 30: 8-9.
PROWELL, D. P. 1998. Sex linkage and speciation in Lep-
idoptera, pp. 309-319. In D. Howard and S. Berlocher
[eds.], Endless Forms: Species and Speciation. Ox-
ford, NY.
QUISENBERRY, S. S. 1991. Fall armyworm (Lepidoptera:
Noctuidae) host strain reproductive compatibility.
Florida Entomol. 72: 194-199.
QUISENBERRY, S. S., AND F. WHITFORD. 1988. Evalua-
tion of bermudagrass resistance to fall armyworm
(Lepidoptera: Noctuidae): influence of host strain
and dietary conditioning. J. Econ. Entomol. 81: 1463-
1468.
RAULSTON, J. R., S. D. PAIR, A. N. SPARKS, J. LOERA G.,
F. A. PEDRAZA M., A. PALAMON T., A. ORTEGA, J.
RUIZ SANCHEZ M., P. MARQUEZ C., H. RULES A., J.
PEREZ M., R. RODRIGUEZ R., H. CARRILLO R., R. AR-
CHUNDIA R., AND F. HERRERA R. 1986. Fall army-
worm distribution and population dynamics in the
Texas-Mexico Gulf Coast area. Florida Entomol. 68:
686-691.
SCOTT, T. W., A. C. MORRISON, L. H. LORENZ, G. G.
CLARK, D. STRICKMAN, P. KITTAYAPONG, H. ZHOU,
AND J. D. EDMAN. 2000. Longitudinal studies of
Aedes aegypti (Diptera: Culicidae) in Thailand and
Puerto Rico: population dynamics. J. Med. Entomol.
37: 77-88.
SOUZA, B., AND C. F. CARVALHO. 2002. Population dy-
namics and seasonal occurrence of adults of
Chrysoperla externa (Hagen, 1861) (Neuroptera:
Chrysopidae) in a citrus orchard in southern Brazil.
Acta Zool. Acad. Sci. Hungaricae 48: 301-310.
SPARKS, A. N. 1979. A review of the biology of the fall ar-
myworm. Florida Entomol. 62: 82-87.
VEENSTRA, K. H., D. P. PASHLEY, AND J. A. OTTEA. 1995.
Host-plant adaptation in fall armyworm host
strains: comparison of food consumption, utilization,
and detoxication enzyme activities. Ann. Entomol.
Soc. Am. 88: 80-91.
WESTBROOK, J. K., AND A. N. SPARKS. 1986. The role of
atmospheric transport in the economic fall army-
worm (Lepidoptera: Noctuidae) infestations in the
southeastern United States in 1977. Florida Ento-
mol. 69: 492-502.
WHITFORD, F., S. S. QUISENBERRY, T. J. RILEY, AND J.
W. LEE. 1988. Oviposition preference, mating com-
patibility, and development of two fall armyworm
strains. Florida Entomol. 71: 234-243.
Florida Entomologist 87(4)
December 2004
LABORATORY SELECTION FOR BEET ARMYWORM
(LEPIDOPTERA: NOCTUIDAE) RESISTANCE TO METHOXYFENOZIDE
JEFFREY GORE AND JOHN J. ADAMCZYK, JR.
Southern Insect Management Research Unit, USDA-ARS, P.O. Box 346, Stoneville, MS 38776
ABSTRACT
Beet armyworms, Spodoptera exigua (Hiibner), were artificially selected in the laboratory for
resistance to the insect growth regulator, methoxyfenozide. A field collected beet armyworm
colony was separated into three cohorts that were independently selected with three concen-
trations (0.033 ppm, 0.064 ppm, and 0.125 ppm) of methoxyfenozide incorporated into a me-
ridic diet. These concentrations corresponded closely with the LC,1 (0.033 ppm), LC5s (0.072
ppm), and LC,, (0.161 ppm), respectively, for the original colony. After seven generations of
continuous exposure to methoxyfenozide, resistance in the colony selected at the low concen-
tration did not increase significantly. In contrast, LCs, values increased 9.7- and 9.4-fold for
the colonies selected at the moderate and high concentrations, respectively, over that of the
original colony. Crosses between resistant and susceptible individuals indicated that the re-
sistance was heritable. At 4 d after exposure, mortality of offspring from the reciprocal crosses
was intermediate between mortality for the offspring from the parental crosses. When rated
at 10 d, mortality of offspring from the reciprocal crosses was not different significantly from
offspring from the cross between susceptible parents. These data will be important for devel-
oping a management program for beet armyworm resistance to methoxyfenozide.
Key Words: IPM, Intrepid, Spodoptera exigua, Insect Growth Regulator.
RESUME
Los gusanos trozadores, Spodoptera exigua (Hiibner), fueron seleccionados artificialmente en
el laboratorio para su resistencia al regulador de crecimiento de insects, metoxifenozido. Una
colonia del gusano trozador recolectada en el campo fue separada en tres cohorts que inde-
pendientemente fueron seleccionados con tres concentraciones (0.033 ppm, 0.064 ppm, y 0.125
ppm) de metoxifenozido incorporadas en una dieta meridica. Estas concentraciones corres-
pondieron de manera cercana con las concentraciones letales CLI, (0.033 ppm), CLs5 (0.072
ppm), y CLg0(0.161 ppm), respectivemente, para la colonia original. Despu6s de siete genera-
clones de ser expuesta continuamente al metoxifenozido, la resistencia en la colonia seleccio-
nada a la concentraci6n menor no aument6 significativamente. En contrast, los valores de
CLs5 aumentaron por 9.7 y 9.4 veces para las colonies seleccionadas de concentraciones mode-
radas y altas, respectivamente, sobre las de la colonia original. Los cruzes entire los individuos
resistentes y susceptibles indicaron que la resistencia puede ser heredada. A los 4 dias des-
pu6s de exponerlos, la mortalidad de la progenie de los cruzes reciprocos fue intermedia entire
la mortalidad de la progenie de los cruzes de los padres. Cuando fue calibrada a los 10 dias, la
mortalidad de la progenie de los cruzes reciprocos no fue significativamente diferente de la
progenie de los cruzes entire los padres susceptibles. Estos datos seran importantes para de-
sarrollar un program de manejo para la resistencia del gusano trozador al metoxifenozido.
The beet armyworm, Spodoptera exigua (Hiib-
ner), is an occasional pest of cotton, Gossypium
hirsutum L., in the southern United States. In
general, beet armyworm populations remain low
due to actions of natural enemies such as the par-
asitoid Cotesia spp., and various predators and
pathogens (Mohaghegh et al. 2001; Bianchi et al.
2002). However, these beneficial insects are inad-
vertently disturbed with numerous insecticide
applications for other pests. Cotton in the south-
ern United States has numerous key pests such
as the heliothine complex (tobacco budworm, He-
liothis virescens (F.), and bollworm, Helicoverpa
zea (Boddie)); tarnished plant bug, Lygus line-
olaris Palisot de Beauvois, and numerous stink
bug species (Heteroptera: Pentatomidae) that re-
quire multiple applications of broad spectrum in-
secticides annually to prevent economic losses.
The introduction of Bollgard cotton has reduced
the numbers of insecticide applications applied
for tobacco budworm and bollworm (Williams
1996, 2001). However, Bollgard cotton is typically
treated multiple times annually with pyrethroids
to control bollworms, and organophosphates to
control tarnished plant bugs and stink bugs.
Those insecticides provide little control of beet
armyworms, but can be damaging to beneficial
arthropods. Consequently, beet armyworms gen-
erally only reach economically damaging levels
after insecticides have been applied to control
other pests. There are few insecticides currently
labeled in cotton that provide effective, economi-
Gore & Adamczyk: Beet Armyworm Resistance to Methoxyfenozide
cal control of beet armyworms. When beet army-
worm outbreaks occur, these insecticides are
applied over large areas; thereby, providing in-
tense selection for the development of resistance.
Methoxyfenozide (Intrepid 2F, Dow Agro-
Sciences, Indianapolis, IN) is one of a few insecti-
cides that provides effective control of beet
armyworms in cotton. This insecticide is an insect
growth regulator that acts as an agonist of 20-hy-
droxyecdysone, a key hormone in the molting pro-
cess (Wing 1988). Methoxyfenozide affects only
the larval stage of Lepidoptera (Wing et al. 1988).
Intoxicated larvae cease feeding soon after inges-
tion and eventually die due to a premature molt.
Although methoxyfenozide is highly effective
against beet armyworm larvae, this compound
has little impact on beneficial arthropods (Wing
et al. 1988). Therefore, this insecticide is benefi-
cial in integrated pest management systems and
preservation of the insect growth regulators is an
important concern.
Beet armyworms are inherently tolerant to
many insecticides and have a high propensity for
developing resistance to insecticides (Wolfen-
barger 2002). Much of the research reported on
beet armyworm resistance has represented com-
pounds from the organochlorine, organophos-
phate, carbamate, and pyrethroid classes of
insecticides. Little work has been conducted on
beet armyworm resistance to insect growth regula-
tors. However, Moulton et al. (2002) demonstrated
that the necessary genetic variability needed for
beet armyworms to develop resistance to the insect
growth regulators, tebufenozide and methoxy-
fenozide, was present in populations from Thai-
land, and may be present in field populations from
the United States. Because of this, proactive resis-
tance management plans should be in place to en-
sure the sustainability of this insecticide. Baseline
data are currently available for beet armyworm
susceptibility to methoxyfenozide (Mascarenhas et
al. 1998a, b). Those data are important for resis-
tance management; however, information concern-
ing the genetic heritability of resistance is not
available. Understanding the heritability of beet
armyworm resistance to methoxyfenozide will be
necessary to design appropriate resistance man-
agement strategies. The current study was con-
ducted to determine inheritance patterns of beet
armyworm resistance to methoxyfenozide as a pro-
active step in the development of an effective resis-
tance management plan.
MATERIALS AND METHODS
A colony of beet armyworms was established
from larvae collected on pigweed, Amaranthus
spp., collected during the summer of 2002. This
colony was maintained in the laboratory for five
generations before the initiation of the experi-
ment. A concentration-mortality bioassay was
conducted with the commercial formulation of In-
trepid 2F. The insecticide was incorporated into a
meridic diet to determine the susceptibility of the
colony before selection. Serial dilutions of In-
trepid 2F and distilled water were made from a
stock solution with a concentration of 1000 ppm of
methoxyfenozide active ingredient. Dilutions
ranged from 1.6 ppm to 100 ppm plus a non-
treated control for initial bioassays. One ml of
each solution was incorporated into 100 ml of me-
ridic diet to obtain eight concentrations of treated
diet that ranged from 0.016 ppm to 1.0 ppm.
Treated and non-treated diet was dispensed into
29.5-ml plastic cups in 2.5-ml aliquots with a re-
peat pipetter. A total of 60 cups for each concen-
tration was used for the bioassay. A single
neonate beet armyworm was placed in each cup,
and mortality was rated after 96 h. Mortality was
scored as inability of larvae to move after being
touched with a blunt probe. A concentration-mor-
tality curve was generated with Probit Analysis
and the LCs0 and LCg, values were determined
(PROC PROBIT, SAS Institute 1989).
The selection experiments were initiated in
November of 2002. Three colonies were indepen-
dently selected for resistance to methoxyfenozide
with three different levels of selection pressure
(low, moderate, and high). The concentrations in-
cluded 0.032 ppm, 0.064 ppm, and 0.125 ppm for
the low, moderate, and high levels of selection, re-
spectively. These concentrations corresponded
closely with LC,, (0.033 ppm), LCs, (0.072 ppm),
and LC9, (0.161 ppm) values, respectively, ob-
tained from the concentration-mortality curve of
the original colony. Larvae were exposed to the in-
secticide by incorporating formulated Intrepid 2F
into meridic diet as previously described. How-
ever, instead of using 29.5-ml plastic cups, ap-
proximately 20 ml of treated diet was dispensed
into 236 ml cardboard cups. This facilitated the
selection of a greater number of individuals on
each concentration compared to making selec-
tions to an individual larva in small cups. Approx-
imately 100 neonates were placed into each cup
and allowed to feed for 96 h. After 96 h, surviving
larvae were transferred individually onto non-
treated diet in 29.5-ml cups. Only larvae that ap-
peared to be developing normally (larvae that
molted to the second instar) were selected. Dead
and moribund larvae were discarded. Larvae
were allowed to complete development on the
non-treated diet. After pupation, beet army-
worms from each colony (i.e., selection pressure)
were mass mated (20 males and 20 females) in
3.79-L cardboard containers. At least ten mating
containers were established for each colony.
Each of the beet armyworm colonies were ex-
posed to selection at the respective concentra-
tions for seven generations. After seven
generations, reciprocal crosses were made with
individuals from colonies selected at moderate
Florida Entomologist 87(4)
and high concentrations to individuals from an-
other laboratory susceptible (LC,, = 0.08, slope =
0.72, df = 4, 2 = 1.13, P = 0.89) colony. This colony
was originally collected from pigweed during
2001. Male beet armyworms from pheromone
traps were incorporated into this colony during
the summer of 2002. Reciprocal crosses between
the laboratory selected colonies and laboratory
susceptible colony were conducted during May of
2003. At least ten mating pairs were established
for each of the parental colonies and their recipro-
cal crosses. The laboratory selected colonies were
only crossed with the laboratory susceptible col-
ony. Reciprocal crosses were not done between the
colonies selected at the moderate and high con-
centrations.
Offspring from each mating pair (90 per pair)
were exposed to a discriminating concentration
(0.125 ppm) of formulated Intrepid 2F in meridic
diet that corresponded to an LCg, for the original
colony. Data for percent mortality of offspring was
corrected for control mortality by Abbott's For-
mula (Abbott 1925) and analyzed with analysis of
variance (PROC MIXED, Littell et al. 1996).
RESULTS AND DISCUSSION
Beet armyworms exposed to moderate and
high selection pressures developed 9.7- and 9.4-
fold levels of resistance compared to the original
colony within seven generations (Fig. 1). The LC5,
value (95% fiducial limits) for the original colony
was 0.07 (0.064-0.082) ppm (slope = 1.60, df = 4, X2
= 1.18, P = 0.88). The susceptibility of the colony
at the low selection pressure did not change after
seven generations of selection (LC,, = 0.07, slope =
0.95, df = 5, X2= 8.55, P=0.13). Within seven gen-
erations, the LC5, value (95% fiducial limits) in-
creased to 0.68 (0.427-1.106) ppm for the colony
selected at the moderate selection pressure (slope
= 0.93, df = 4, X2 =10.25, P = 0.04). Similarly, the
- Lot Selct ion (G7) Parcnal Colony IGO
- Modcra Selection (G7) No Sclection (G7)
- High Sclccion (G7)
.I I
Log Concentration (ppm methoxyfenozide in diet)
Fig. 1. Results of probit analysis for concentration-
mortality curves of beet armyworm susceptibility to
methoxyfenozide before and after laboratory selections.
LC5, value (95% fiducial limits) for the colony at
the high selection pressure increased to 0.66
(0.554-0.809) ppm. The concentration-mortality
curves and corresponding LC50 values were simi-
lar between the colonies selected at the moderate
and high concentrations based on overlap of 95%
fiducial limits. This may be an indication that the
same mechanism of resistance was isolated from
the two different selection pressures. Although
the two selection pressures yielded similar LC5,
values, the colony exposed to the high selection
pressure appeared to be less variable in its re-
sponse to the methoxyfenozide treated diet than
the colony exposed to the moderate selection pres-
sure. This is evidenced by the wide range of 95%
fiducial limits observed with the colony selected
at the moderate concentration. Also, the colony at
the high selection pressure had a lower X2 value
than the colony selected at the moderate concen-
tration indicating that they were more homoge-
neous in their response to methoxyfenozide. This
would be expected because the higher selection
pressure should create a greater genetic bottle-
neck; whereas, the moderate selection pressure
may have resulted in selection of other factors
such as overall health and vigor in addition to the
resistance trait.
Based on results from crosses between individ-
uals from the colonies selected at the moderate
and high selection pressures with individuals
from the susceptible colony, the selected trait was
determined to be heritable. However, the mode of
inheritance is difficult to determine from these
data. Mortality of larvae was significantly differ-
ent among the different crosses for the colonies
selected at the moderate (F = 60.46; df = 3, 22.2;
P < 0.01) and high (F = 16.05; df = 3, 29; P < 0.01)
pressures at 4 d (Fig. 2). When rated at 4 d, mor-
tality of offspring from the parental crosses for
the susceptible and resistant colonies averaged
(SEM) 92.8 (2.53) percent and <29.0 (<8.85) per-
cent, respectively. Offspring from the reciprocal
crosses had intermediate levels of mortality.
Similar to mortality at 4 d, mortality at 10 d
was significantly different among the crosses for
the moderate (F = 12.25; df = 3, 32; P < 0.01) and
high (F = 8.16; df = 3, 29; P < 0.01) selection pres-
sures (Fig. 2). At 10 d; however, mortality was not
significantly different among offspring from the
reciprocal crosses and offspring from the parental
cross for the susceptible colony. Mortality of off-
spring from the reciprocal crosses and the paren-
tal cross for the susceptible colony ranged from
82.9 (3.67) to 99.1 (0.38) percent for the colony se-
lected at the moderate pressure. Mortality of off-
spring from the cross between resistant parents
averaged 54.3 (9.7) percent. For the colony se-
lected at the high pressure, mortality of offspring
from the reciprocal crosses and the parental cross
for the susceptible colony ranged from 88.4 (3.75)
to 99.1 (0.38) percent; while, mortality of off-
December 2004
Gore & Adamczyk: Beet Armyworm Resistance to Methoxyfenozide
Male x Female
*SxS 0RxS EISxR ORxR
G8 (Moderate) G8 (High)
a 10 a a
,-. so -
Fig. 2. Mortality of offspring from reciprocal crosses
ony averaged 61.4 (8.6) percent.
Based on results of these laboratory selections,
the potential for beet armyworms to develop re-
20
G8 (Moderate) G8 (High)
Fig. 2. Mortality of offspring from reciprocal crosses
on a discriminating concentration of methoxyfenozide does exist. Beet ar-
spring from the parental cross of the resistant col-
ony averaged 61.4 (8.6) percent.
Based on results of these laboratory selections,
the potential for beet armyworms to develop re-
sistance to methoxyfenozide does exist. Beet ar-
myworm larvae from a field-collected colony
developed approximately 10-fold level of resis-
tance in a relatively short period of time (seven
generations) in the current study. Moulton et al.
(2002) isolated a beet armyworm strain from
Thailand with 320- to 340-fold level of resistance
to methoxyfenozide compared to a laboratory
strain. In a similar study, a sex linkage was deter-
mined for a beet armyworm strain exhibiting re-
sistance to several insecticides (Wolfenbarger
2002). In that experiment, resistance to fenvaler-
ate and methomyl was associated with the two X
chromosomes of the male resistant strain and not
the Y chromosome of the female (Wolfenbarger
2002). In contrast, ratings at 4 d from crosses in
the current study suggest that the Y chromosome
of the female may be important in resistance de-
velopment. However, when mortality was rated at
10 d, sex-linkage of the trait was not apparent;
therefore, no definitive conclusions can be made
concerning inheritance of the trait selected in
these experiments. More research is necessary to
isolate strains of beet armyworms with varying
levels of resistance to methoxyfenozide so that
they can be crossed with susceptible individuals
for multiple generations to determine inheritance
of resistance. This information will be important
for proactive management of beet armyworm re-
sistance to methoxyfenozide.
ACKNOWLEDGMENTS
The authors thank Don Hubbard, Les Price, Jennifer
Holcomb, and Leslie Hughes for assistance with these
experiments. Drs. Roger Leonard, Gordon Snodgrass,
and 0. P. Perera provided critical reviews of early drafts
of this manuscript. Mention of a proprietary product is
for informational purposes only and does not constitute
endorsement by the U.S. Department of Agriculture.
REFERENCES CITED
BIANCHI, F. J. J. A., J. M. VLAK, R. RABBINGE, AND W.
VAN DER WERF. 2002. Biological control of beet ar-
myworm, Spodoptera exigua, with baculoviruses in
greenhouses: development of a comprehensive pro-
cess-based model. Biol. Control 23: 35-46.
MASCARENHAS, V. J., J. B. GRAVES, B. R. LEONARD, AND
E. BURRIS. 1998a. Susceptibility of field populations
of beet armyworm (Lepidoptera: Noctuidae) to com-
mercial and experimental insecticides. J. Econ. En-
tomol. 91: 827-833.
MASCARENHAS, V. J., J. B. GRAVES, B. R. LEONARD, AND
E. BURRIS. 1998b. Dosage-mortality responses of
third instars of beet armyworm (Lepidoptera: Noctu-
idae) to selected insecticides. J. Agric. Entomol. 15:
125-140.
MOHAGHEGH, J., P. DECLERCQ, AND L. TIRRY. 2001.
Functional response of the predators Podisus macu-
liventris (Say) and Podisus nigrispinus (Dallas)
(Heteroptera: Pentatomidae) to the beet armyworm,
Spodoptera exigua (Hubner) (Lepidoptera: Noctu-
idae): effect of temperature. J. Appl. Entomol. 125:
131-134.
MOULTON, J. K., D. A. PEPPER, R. K. JANSSON, AND T. J.
DENNEHY. 2002. Pro-active management of beet ar-
myworm (Lepidoptera: Noctuidae) resistance to
tebufenozide and methoxyfenozide: baseline moni-
toring, risk assessment, and isolation of resistance.
J. Econ. Entomol. 95: 414-424.
WILLIAMS, M. R. 1996. Cotton insect losses-1995, pp.
670-689 In P Dugger and D. A. Richter [eds.], Pro-
ceedings, 1996 Beltwide Cotton Conferences, Na-
tional Cotton Council, Memphis, TN.
WILLIAMS, M. R. 2001. Cotton insect loss estimates-
2000, pp. 774-777 In P. Dugger and D. Richter [eds.],
Proceedings, 2001 Beltwide Cotton Conferences, Na-
tional Cotton Council, Memphis, TN.
WING, K. D. 1988. RH-5849, a nonsteroidal ecdysone ag-
onist: effects on a Drosophila cell line. Science 241:
467-469.
WING, K. D., R. A. SLAWECKI, AND G. R. CARLSON. 1988.
RH-5849, a nonsteroidal ecdysone agonist: effects on
larval Lepidoptera. Science 241: 470-472.
WOLFENBARGER, D. A. 2002. Inheritance of resistance
by a strain of beet armyworm to fenvalerate, meth-
omyl, methyl parathion, and permethrin. Resistant
Pest Manag. Newsletter 12: 25-27.
Florida Entomologist 87(4)
December 2004
LARVAL DEVELOPMENT OF FALL ARMYWORM
(LEPIDOPTERA: NOCTUIDAE) ON DIFFERENT COVER CROP PLANTS
R. L. MEAGHER, R. N. NAGOSHI, C. STUHL1 AND E. R. MITCHELL2
Center for Medical, Agricultural and Veterinary Entomology, Agricultural Research Service
U.S. Department of Agriculture, Gainesville, FL 32608
1University of Florida, Entomology and Nematology Department, P. O. Box 110620, Gainesville, FL 32611-0620
2Deceased
ABSTRACT
A series of laboratory and field experiments were conducted to compare larval development,
feeding behavior, and ovipositional preference of fall armyworm (Spodoptera frugiperda) on
a standard host plant, a standard cover crop plant, and two candidate cover crop plants. The
results indicate that larvae from different rearing cultures and host strains developed com-
parably on corn and sorghum-sudangrass, but generally developed poorly on cowpeas and
sunnhemp. Larval and ovipositional experiments also suggested a preference for either corn
or sorghum-sudangrass. Field plantings of cowpeas and sunnhemp in two locations were ig-
nored by fall armyworm in favor of corn. These studies suggest that cowpeas and sunnhemp
have the potential to reduce stepping stone nursery populations of fall armyworm by length-
ening developmental time and increasing larval mortality.
Key Words: Spodoptera frugiperda, Cowpeas, Sunnhemp.
RESUME
Una series de experiments de laboratorio y de campo fueron realizados para comparar el de-
sarrollo de las larvas, el comportamiento de alimentaci6n, y la preferencia de la oviposici6n
del gusano cogollero (Spodoptera frugiperda) en una plant hospedera tipica, una plant de
cobertura tipica, y dos plants candidates para ser usadas en cobertura. Los resultados in-
dicaron que las larvas de diferentes crias y cepas de hospederos desarrollaron similarmente
en maiz y sorgo-pasto de sudan, pero en general desarrollaron pobremente en caupi y en
"sunn-hemp" (Crotalariajuncea). Los experiments sobre las larvas y la oviposici6n tambi6n
sugerieron que hay una preferencia para maiz o sorgo-pasto de sudan. Las siembras de
campo con caupi y Crotalaria en dos localidades fueron ignoradas por el gusano cogollero en
favor de maiz. Estos studios sugerieron que la caupi y Crotalaria tienen un potential para
reducir las poblaciones del gusano cogollero en los viveros de piedra escalonados por alargar
el tiempo de desarrollo y al aumentar la mortalidad de las larvas.
Fall armyworm, Spodoptera frugiperda (J. E.
Smith) is a polyphagous, migratory insect that
moves northward each season from overwintering
areas in southern Florida (Luginbill 1928; Mitch-
ell 1979; Pair et al. 1986; Westbrook & Sparks
1986; Mitchell et al. 1991) and southern Texas/
northern Mexico (Luginbill 1928; Raulston et al.
1986; Pair et al. 1991). Populations from the over-
wintering areas in southern Florida move into
secondary source areas or "stepping-stone nurser-
ies" located in northern Florida and southern
Georgia in April and May, and it is believed that
populations from these areas can increase and
add to the numbers of moths moving northward
(Mitchell 1979; Pair & Westbrook 1995).
Fall armyworm is composed of two sympatric
and morphologically identical strains that are de-
fined by their host plant preferences (Pashley et
al. 1985; Pashley 1986). One strain was identified
from populations feeding on corn and sorghum
(corn strain) and the other strain was identified
from populations feeding on rice and forage
grasses (rice strain). The two strains can be dis-
tinguished by strain-specific allozyme variants
and genetic markers (Lu et al. 1994, Lu & Adang
1996; McMichael & Prowell 1999).
Corn (Zea mays L.) and other host plants, in-
cluding vegetables and cover crops, may allow fall
armyworm populations of both strains to increase
during spring in the stepping stone nurseries. In
northeastern Florida, over 12,000 ha of vegetable
crops, primarily potatoes (Solanum tuberosum L.)
and cabbage (Brassica oleracea L. var. capitata)
are grown during January to May in three counties
(St. John's, Flagler, and Putnam) (Aerts &
Nesheim 2000; Larson Vasquez & Nesheim 2000).
After vegetable harvest, cover crops such as sor-
ghum-sudangrass (Sorghum bicolor (L.) Moench, a
sorghum/sudan hybrid) generally are planted. Sor-
ghum-sudangrass is a warm-season annual grass
Meagher et al.: Cover Crops for Fall Armyworm
hybrid that is used in Florida as a green manure
cover crop following harvest of winter vegetables
(Chambliss 2002). These plants are used to pro-
duce biomass, contribute nitrogen to the soil, in-
crease soil organic matter, and prevent soil erosion
(Chambliss et al. 2003; Rich et al. 2003). Benefits
of cover crops in regards to biomass production, ni-
trogen yield, and crop yield have been shown for
both large vegetable production systems (Creamer
& Baldwin 2000) and small subsistence farming
systems (Jeranyama et al. 2000).
In northern and southern Florida large popu-
lations of fall armyworm can develop in fields
planted to sorghum-sudangrass (ERM, unpub-
lished data; Pair & Westbrook 1995). Therefore,
alternative cover crops that are poorer host
plants for fall armyworm may reduce migrating
populations. This research was conducted to com-
pare population development and feeding behav-
ior of fall armyworm host strains on a standard
host plant (corn), a standard cover crop plant (sor-
ghum-sudangrass) (SSG), and two candidate
cover crop plants (iron-clay cowpeas and sunn-
hemp). Cowpeas [Vigna unguiculata (L.) Walpers
spp. unguiculata] is a warm-season annual le-
gume that alone or mixed with SSG can be used
as a rotation or cover crop with vegetables (Miller
et al. 1989). Sunnhemp (Crotalariajuncea L.) is a
warm-season legume that is used as a cover crop
in alternation with vegetable crops (Duke 1981;
Li et al. 2000; Rich et al. 2003).
MATERIALS AND METHODS
Larval Feeding Experiments
Larvae for this experiment were from two
sources. 'Tifton' larvae were from a laboratory
culture reared on artificial diet and shown to
carry the mitochondrial marker of corn strain
(Meagher & Gallo-Meagher 2003; Nagoshi &
Meagher 2003). This culture originated from indi-
viduals received from Dr. James Carpenter,
USDA-ARS, Tifton, GA and was reared on a pinto
bean artificial diet according to the procedures of
Guy et al. (1985).'Ona' larvae were from a culture
of individuals collected in July 2002 from the
Range Cattle Research and Education Center,
Ona, FL. This culture was reared on grasses (ber-
mudagrass, Cynodon dactylon (L.) Pers. and star-
grass, C. nlemfuensis Vanderyst var. nlemfuensis
'Florona') and has the rice strain mitochondrial
marker (Nagoshi & Meagher 2003).
Plants were grown in 3.8-1 pots in a greenhouse
at ambient temperature, and were fertilized
weekly with Miracle-Gro 15-30-15 plant food.
No pesticides were applied to the plants. New
growth leaves were selected from plants aged ca. 3
wk. old for cowpeas, SSG, and corn, and 6 wk. old
for sunnhemp. Plant foliage was placed on filter
paper discs (Whatman, 90 mm) moistened with
ca. 1 ml deionized water in a 9-cm diameter poly-
styrene petri dish (Thomas Scientific, catalog
#3488-B32). One neonate larva was placed on
plant foliage, and the petri dishes were placed in
an incubator at 23.9 2 with a 14:10 photoperiod.
The filter paper in each petri dish was moistened
daily with ca. 1 ml of deionized water for the first
10 days. Larvae were supplied with a continuous
amount of fresh plant material until time of pupa-
tion. Larval weights were measured at 10 days.
Development time (in days) from neonate to pupa,
pupal weight, and sex were recorded at pupation.
For each host plant, 15-30 larvae were arranged
in three replications on different dates, and mor-
tality on each host plant was recorded. Analysis of
variance of loglO-transformed data (PROC
MIXED, Contrasts, Littell et al. 1996) was used to
examine variation among plants.
Larval Choice Experiments
The larval choice experiments were designed
to evaluate the feeding preference of larvae on
leaf sections of four host plants. Rice strain larvae
from the Ona culture were used along with corn
strain larvae collected from a population near
Hague, FL. This culture was reared on green-
house-grown corn and was in culture for < four
generations. Plant culturing and leaf selection
was similar to that used for the larval feeding ex-
periment, except that leaves were trimmed to a
uniform size (ca. 2 x 5 cm).
Four separate experiments were completed.
First, a four-choice experiment compared sections
from the four host plants. One section of each
plant was randomly placed on filter paper discs
moistened with ca. 3 ml deionized water and sec-
tions were placed ca. 2 cm from the center along
the outer edge of the petri dish. Twenty neonate
larvae from Ona or Hague cultures were placed in
the center of each petri dish, which were then
placed in an incubator at 23.9 2 with a 14:10
photoperiod. Counts were made after 24 h by ob-
serving the number of larvae on or under each
leaf section. The second experiment was a two-
choice test comparing corn to each of the other
plants using Ona larvae.
Results from the first two experiments led us
to test two possible explanations for the reduced
plant host specificity of the rice strain culture.
The third experiment examined whether the
feeding experience of the parents influenced the
plant host preference of the next generation. Rice
strain larvae that completed development on
each host plant were mated and resulting prog-
eny were tested against all four plants. The fourth
experiment was a four-choice test that used the
progeny of larvae that selected each host plant in
Experiment 1. Therefore, larvae that selected
corn, cowpeas, SSG, or sunnhemp were reared on
that plant and their progeny used in the bioassay.
Florida Entomologist 87(4)
All experiments contained 10 replications. Analy-
sis of variance of square root-transformed data
(PROC MIXED, Contrasts, Littell et al. 1996) was
used to examine variation among plants.
Oviposition Choice Experiments
In addition to larval preference, the differen-
tial distribution of fall armyworm on different
plant types could result from selective oviposi-
tional behavior by adult females. Oviposition
choice experiments were conducted in a green-
house using plastic swimming pools as a cage.
The bioassay consisted of a plastic swimming pool
(109.2 cm d x 12 cm h) containing 22.7 kg of com-
mercially available sand. Hardware cloth (1.27
cm 1, 25.4 cm h) was curled and placed in the sand
along the inside edge of the pool. Gray window
screen was placed on top of the hardware cloth,
and a second inverted swimming pool was placed
on top of the screen to form a 'sandwich'.
All tests used our Gainesville laboratory cul-
ture reared on artificial diet and shown to carry
the mitochondrial marker of corn strain (Meagher
& Gallo-Meagher 2003). Greenhouse-grown plants
(same size and age as plants used above) were
placed outside for 3 d before being used. One plant
of each host was placed randomly in a circle inside
the unit. Ten male and female moths (<5 d old)
were placed along with a 10% honey-sugar solu-
tion feeding station in the center of the unit. The
screen and top pool were placed on top of the hard-
ware cloth enclosing the moths. In this way, moths
could not oviposit on the smooth surface of the top
or bottom pools. Plants were sampled 48 h later for
eggmasses. Experiment 1 was completed June-
August 2001 and used 10 separate cages (replica-
tions); experiment 2 was completed in October
2001 and used 8 replications. All data were trans-
formed by a square root (y + 0.5) transformation
and means were separated with the Contrasts
statement in PROC MIXED (Littell et al. 1996).
Field Experiments
Corn, cowpeas, SSG, and sunnhemp were
planted in four randomized complete blocks at a
field site in Gainesville on 19 June 2001 and at
the University of Florida Vegetable Research Sta-
tion, Hastings on 8 July 2001. Each plot was four
rows wide by 30.5 m long. An additional treat-
ment of green manure (SSG, incorporated into the
soil 60 d after planting) was added to the Hast-
ings test. Standard agronomic practices were
used except no insecticides were applied. All plots
were sampled beginning in late July and ending
in early October for evidence of plant damage and
the presence of larvae. Plants were sampled for
larvae by checking a randomly selected one-meter
row at each observation point. Analysis of vari-
ance of square root-transformed data (PROC
MIXED, Contrasts, Littell et al. 1996) was used to
examine variation among treatments.
RESULTS
Larval Feeding
Differences between Tifton (corn strain) and
Ona (rice strain) cultures were found for larval
weight, pupal weight, larval development time,
and survival, and there was a significant interac-
tion between culture and host plant (P < 0.0001)
(Table 1). Therefore results from each culture
were statistically analyzed separately. For the
Tifton culture, the largest larvae were found on
corn, where the average weight was higher than
that of larvae cultured on SSG or cowpeas (P <
0.001). Larvae grown on sunnhemp were signifi-
cantly smaller, averaging approximately 21% of
TABLE 1. LARVAL WEIGHT, PUPAL WEIGHT, LARVAL DEVELOPMENT TIME, AND SURVIVAL OF TIFTON (CORN STRAIN) AND
ONA (RICE STRAIN) FALL ARMYWORM LARVAE FED DIFFERENT HOST PLANTS. MEANS WITH THE SAME UPPER-
CASE (BETWEEN CULTURES) OR LOWERCASE (AMONG PLANTS WITHIN A CULTURE) LETTER ARE NOT SIGNIFI-
CANTLY DIFFERENT, P > 0.05.
Larval wt. Pupal wt. Development time
Culture Plant (mg) (mg) (days) % Survival
Tifton 50.4 4.1 B 182.4 + 4.2 A 20.9 0.3 B 62.2 0.07 B
corn 83.5 + 8.8 a 160.3 + 6.8 b 19.1 0.5 a 62.2 + 0.12 b
SSG 51.2 6.5 b 209.6 5.7 a 19.7 0.3 a 95.6 0.04 a
cowpeas 34.5 5.0 bc 151.9 8.9 b 23.3 0.6 b 51.1 0.06 b
sunnhemp 18.0 3.9 c 190.6 7.4 a 23.4 0.5 b 40.0 0.04 b
Ona 109.0 + 6.5 A 158.8 2.2 B 18.1 + 0.3 A 77.2 + 0.05 A
corn 122.2 6.1 b 166.8 3.0 a 15.6 0.2 b 91.1 + 0.02 a
SSG 209.6 + 12.9 a 176.2 3.1 a 14.3 0.2 a 86.7 0 ab
cowpeas 62.5 6.4 c 151.0 4.7 b 20.5 0.5 c 67.8 0.04 b
sunnhemp 41.9 4.6 c 140.6 4.1 c 22.1 + 0.4 d 63.3 + 0.13 b
December 2004
Meagher et al.: Cover Crops for Fall Armyworm
the weight from those on corn (Table 1). A some-
what different result was observed with pupal
weights. The heaviest pupae were those develop-
ing on SSG and sunnhemp, with those from corn
and cowpeas being about 20-25% smaller. Larvae
feeding on cowpeas and sunnhemp took longer to
develop than those on corn and SSG, and had
lower survival than those on SSG.
The rice strain was similar to the corn strain in
that the smallest larvae, longest development
time, and highest mortality were associated with
development on cowpeas and sunnhemp (Table
1). There was unusually high early larval growth
on SSG, with an average weight about 70%
greater than found on corn. However this early
growth on SSG did not lead to an exceptionally
high pupal weight.
Larval Choice
Larvae from the more recently derived corn
strain culture (Hague) selected corn over SSG
and were rarely found on cowpeas (P < 0.0001, Ta-
ble 2). No larvae were found on sunnhemp. In
comparison, rice strain larvae from the Ona cul-
ture displayed an equal preference for corn and
SSG compared to sunnhemp and cowpeas (Table
2). The two-choice bioassays with rice strain lar-
vae comparing corn with the other plants showed
only a difference between corn and cowpeas (Ta-
ble 3).
Ona larvae whose parents were reared on the
different host plants were generally found more on
corn and SSG and less on cowpeas, no matter what
the host plant was of the parents (Table 4). The ex-
ception was larvae from SSG-reared parents,
which showed equal preference for all plants. Ona
larvae whose parents selected different host plants
were generally found more on corn or SSG and less
on cowpeas (Table 5). Larvae from parents that se-
lected sunnhemp were an exception, choosing that
plant at a similar frequency as corn or SSG.
TABLE 2. NEONATE FALL ARMYWORM LARVAE FROM
HAGUE OR ONA CULTURES TESTED IN CHOICE
BIOASSAYS COMPARING FOUR HOST PLANTS.
MEANS (AMONG PLANTS WITHIN A CULTURE)
WITH THE SAME LETTER ARE NOT SIGNIFI-
CANTLY DIFFERENT, P > 0.05.
Culture Plant Larvae per section
Hague corn 14.6 1.3 a
SSG 4.9 1.3 b
sunnhemp 0 0 c
cowpeas 0.2 0.1 c
Ona corn 8.3 0.8 a
SSG 6.3 0.8 a
sunnhemp 2.6 0.8 b
cowpeas 1.8 0.7 b
TABLE 3. TWO-CHOICE BIOASSAYS OF ONA FALL ARMY-
WORM NEONATES COMPARING CORN TO SSG,
SUNNHEMP, OR COWPEAS. MEANS IN EACH COM-
PARISON WITH THE SAME LETTER ARE NOT SIG-
NIFICANTLY DIFFERENT, P > 0.05.
Plant Larvae per section
Corn 3.2 0.7 a
SSG 4.2 1.9 a
Corn 8.6 0.8 a
Sunnhemp 5.2 1.0 a
Corn 11.6 1.5 a
Cowpeas 2.6 0.5 b
Oviposition Choice
The first study performed during the summer
resulted in few eggmasses and no significant dif-
ference among plants (Fig. 1). Oviposition during
October was higher, with more eggmasses found
on corn and SSG plants and fewer on cowpeas or
sunnhemp. These results indicate that at least
under greenhouse conditions and during the au-
tumn season, females can show a preference for
corn and SSG over the other plants.
Field Experiment
In Hastings, larval populations were found
primarily on corn plants (12.9 2.6 larvae per 1
m-row) compared to SSG (2.1 0.5) or SSG with
TABLE 4. FOUR-CHOICE BIOASSAYS WITH ONA FALL ARMY-
WORM NEONATES WHOSE PARENTS WERE REARED
ON CORN, SSG, SUNNHEMP OR COWPEAS. MEANS
(WITHIN A GROUPING) WITH THE SAME LETTER
ARE NOT SIGNIFICANTLY DIFFERENT, P > 0.05.
Plant parents
reared on Plant Larvae per section
corn corn 4.6 1.0 a
SSG 6.9 1.3 a
sunnhemp 3.6 0.8 b
cowpeas 2.8 0.5 b
SSG corn 5.9 1.1 a
SSG 5.2 0.9 a
sunnhemp 3.7 0.8 a
cowpeas 3.4 1.4 a
sunnhemp corn 8.0 0.8 a
SSG 7.8 1.1 a
sunnhemp 1.4 0.5 b
cowpeas 1.6 0.6 b
cowpeas corn 6.7 1.5 a
SSG 5.4 0.9 a
sunnhemp 3.6 + 0.6 ab
cowpeas 1.8 0.7 b
458 Florida Entc
TABLE 5. FOUR-CHOICE BIOASSAYS WITH F, ONA FALL AR-
MYWORM NEONATES WHOSE PARENTS SE-
LECTED CORN, SSG, SUNNHEMP OR COWPEAS
IN FIRST TEST. MEANS (WITHIN A GROUPING)
WITH THE SAME LETTER ARE NOT SIGNIFI-
CANTLY DIFFERENT, P > 0.05.
Plant parents
selected Plant Larvae per section
corn corn 5.4 0.5 a
SSG 5.4 0.6 a
sunnhemp 6.6 0.9 a
cowpeas 0.9 0.3 b
SSG corn 7.1 1.0 a
SSG 3.8 0.8 b
sunnhemp 5.7 0.8 ab
cowpeas 1.9 0.4 c
sunnhemp corn 4.7 1.0 ab
SSG 6.2 0.6 a
sunnhemp 5.0 0.8 ab
cowpeas 3.1 0.6 b
cowpeas corn 6.7 0.6 a
SSG 5.5 0.8 ab
sunnhemp 3.2 0.7 bc
cowpeas 2.8 0.7 c
green manure (1.9 0.5) (Fig. 2). No fall army-
worm larvae were found on cowpeas or sun-
nhemp. The Gainesville plots provided the same
information, as more larvae were collected on
corn (26.0 + 3.8) compared to SSG (3.5 + 1.0). Very
few larvae were collected on sunnhemp (0.17 +
0.1) or cowpeas (0.08 0.06). It was surprising
that few larvae were found on SSG, but the re-
sults suggest that when both corn and SSG are
present in the same area, fall armyworm will
show a strong bias to the former.
a
2
r -
E
ua
a b
a [
cam SSG cowpea sunnhemp
Host Plant
Fig. 1. Adult moths tested in swimming pool bioassays
comparing oviposition on four host plants in summer and
October experiments. For each test, columns with the
same letter are not significantly different, P > 0.05.
mologist 87(4)
48
34
4S
0
18
o
-J
HI
40
30
20
December 2004
0 T-----V A.-
31JUI iSp 17SOp ZS op o10C1 OCl
Date
Fig. 2. Number of fall armyworm larvae collected
from 1-m row of host plant in experiments conducted at
Hastings and Gainesville, FL, 2001.
DISCUSSION
Previous studies have compared the develop-
ment of the two host strains on different food
sources, most notably corn and turf grass (Pash-
ley 1988; Whitford et al. 1988; Pashley et al. 1995;
Veenstra et al. 1995). These showed several
strain-specific differences with respect to larval
and pupal weights, consumption rates, develop-
ment times, and mortality. It was surprising that
only 62.2% of the Tifton larvae feeding on corn
survived to the pupal stage. Although the Tifton
culture shows the mitochondrial marker for corn
strain, it was a laboratory culture that histori-
cally developed on meridic diet and may have ex-
hibited feeding behavior effects of colonization
(Mason 1987).
In general, rice strain larvae were larger and
the development period shorter than the corn
strain on all four host plants. This differs from
previous reports indicating that when grown on
corn, corn strain larvae were larger (Pashley et al.
1995). Apparently substantial variation can occur
Gainesville
"-- corn
-- SSG
-d- *unnhlmp
.-*- uOwP
w- C.
Meagher et al.: Cover Crops for Fall Armyworm
with different cultures and experimental condi-
tions. Despite these potential difficulties, our lar-
val feeding studies consistently indicated that the
two strains find cowpeas and sunnhemp to be a
less favorable food source than corn or SSG.
Results from the larval choice experiments
suggest strain differences in the preference of lar-
vae to particular plant hosts. While both strains
were attracted to corn, the corn strain preference
appeared to be stronger and more specific. The re-
sult with corn strain larvae is consistent with a
previous study that showed that larvae demon-
strated a strong preference for corn over turf
grass (Pashley et al. 1995).
For rice strain larvae, parental feeding on corn
or SSG did not lead to increased bias of their prog-
eny for these plants. Similarly, the larvae from
parents raised on cowpeas or sunnhemp showed
no increased preference to these plants. These re-
sults indicate that parental feeding history does
not significantly influence the feeding preference
of the progeny. Experiment 4 tested whether the
variation in the rice strain culture was due to ge-
netic polymorphisms, e.g., the presence of subpop-
ulations with heritable biases to different plants.
If this were the case, then larvae from parents
that selected a particular host should be predis-
posed to the same bias, resulting in an increased
proportion of that generation choosing the same
plant. Results showed no clear evidence of strong
heritability for plant preference. Therefore, it ap-
pears that the reduced specificity of rice strain
larvae to corn compared to the corn strain repre-
sents an innate strain-specific difference.
Fall armyworm oviposition has been described
on corn (Thomson & All 1983) and cotton (Ali et
al. 1989), with differential preferences reported
both within and among crops (Combs & Valerio
1980; Pitre et al. 1983). Only one previous study
examined oviposition preference between host
strains, with results showing a strong preference
for corn strain moths to oviposit on corn and rice
strain moths to oviposit on bermudagrass (Whit-
ford et al. 1988). Recent research suggests that
tactile cues may be more important than plant
volatiles (Rojas et al. 2003), consistent with obser-
vations of egg masses on many non-plant objects
such as vinyl flags (Thomson & All 1982), plastic
bucket traps, and vehicle mirrors (RLM, unpub-
lished). Our results indicate that a seasonal and/
or environmental context must be taken into ac-
count when interpreting ovipositional studies.
A previous study described fall armyworm in-
festation in a forage sorghum field in southern
Florida (Pair & Westbrook 1995). With 64% of the
200-ha field infested, the authors calculated hypo-
thetically that the field could have produced 320
million adults/ha if 50% of the larvae survived to
adulthood. The amount of SSG grown in Florida is
not known, but almost 122,000 ha of vegetables
were planted in 2002 (Anonymous 2003). If 50% of
the vegetable land is planted to a SSG cover crop,
over 19 trillion adult moths are potentially pro-
duced (320 million adults per ha x 61,000 ha).
Therefore, large numbers of moths are potentially
produced that could migrate to northerly areas or
remain in local areas to later infest future crops.
Our experiments suggest that cowpeas and
sunnhemp have the potential to reduce these pop-
ulations of fall armyworm by lengthening devel-
opmental time and increasing larval mortality.
Furthermore, these alternative cover crops are
much less attractive to fall armyworm larvae and
adults, and so may limit the concentration of this
pest when corn is unavailable. Reducing the en-
dogenous fall armyworm population in cover
crops could both mitigate the level and delay the
timing of infestation during the subsequent corn-
growing season while simultaneously interfering
with migration. While a potentially effective and
economical means of cultural control, substantial
additional research is needed to determine the in-
fluence of cover crops on the population dynamics
of fall armyworm and thereby the potential effect
of changes in cover crop choice. In particular, we
need to determine the extent to which different
cover crops serve as refuges for fall armyworm
populations throughout the year and in particu-
lar during periods when corn is unavailable.
ACKNOWLEDGMENTS
We thank H. Burnside, W. Copeland, C. Dillard, R.
Furlong, and J. Leach for technical assistance with the
experiments, and D. P. Weingartner (Univ. of Florida,
Plant Pathology Department) for assistance in the field
experiment in Hastings. We thank J. Carpenter (USDA-
ARS, Tifton, GA) for the Tifton fall armyworm culture.
We thank R. L. Bugg (Univ. of California-Davis) and S.
Pair (USDA-ARS, Lane, OK) for critical review of an
early manuscript. The use of trade, firm, or corporation
names in this publication is for the information and con-
venience of the reader. Such use does not constitute an
official endorsement or approval by the United States
Department of Agriculture or the Agricultural Research
Service of any product or service to the exclusion of oth-
ers that may be suitable.
REFERENCES CITED
AERTS, M. J., AND O. N. NESHEIM. 2000. Florida crop/
pest management profiles: potatoes. Univ. of Florida,
IFAS, Florida Coop. Ext. Serv. Circ. 1237.
ALI, A. R. G. LUTTRELL, H. N. PITRE, AND F. M. DAVIS.
1989. Distribution of fall armyworm (Lepidoptera:
Noctuidae) egg masses on cotton. Environ. Entomol.
18: 881-885.
ANONYMOUS. 2003. Florida agricultural facts. Florida
Agricultural Statistics Service, Orlando, FL.
Chambliss, C. G. 2002. Producing millets and sorghums.
Univ. of Florida, IFAS, Florida Coop. Ext. Serv. SS-
AGR-89.
CHAMBLISS, C. G., R. M. MUCHOVEJ, AND J. J. MULLA-
HEY. 2003. Cover crops. Univ. of Florida, IFAS, Flor-
ida Coop. Ext. Serv. SS-AGR-66.
COMBS, R. L., JR., AND J. R. VALERIO. 1980. Oviposition
by the fall armyworm on four varieties of bermuda-
grass. J. Georgia Entomol. Soc. 15: 164-167.
CREAMER, N. G., AND K. R. BALDWIN. 2000. An evalua-
tion of summer cover crops for use in vegetable pro-
duction systems in North Carolina. HortScience 35:
600-603.
DUKE, J. A. 1981. Handbook of Legumes of World Eco-
nomic Importance. Plenum, New York. 345 pp.
GuY, R. N., N. C. LEPPLA, J. R. RYE, C. W. GREEN, S. L.
BARETTE, AND K. A. HOLLIEN. 1985. Trichoplusia ni,
pp. 487-494 In P. Sing and R. F. Moore [eds.], Hand-
book of Insect Rearing, vol. 2. Elsevier, Amsterdam.
JERANYAMA, P., O. B. HESTERMAN, S. R. WADDINGTON,
AND R. R. HARWOOD. 2000. Relay-intercropping of
sunnhemp and cowpea into a smallholder maize sys-
tem in Zimbabwe. Agron. J. 92: 239-244.
LARSON VASQUEZ, B. C., AND O. N. NESHEIM. 2000. Flor-
ida crop/pest management profiles: cabbage. Univ. of
Florida, IFAS, Florida Coop. Ext. Serv. Circ. 1256.
LI, Y., H. BRYAN, AND T. OLCZYK. 2000. Sunn hemp a
cover crop in Florida. Univ. of Florida, IFAS, Florida
Coop. Ext. Serv. SS-AGR-96.
LITTELL, R. C., G. A. MILLIKEN, W. W. STROUP, AND R.
D. WOLFINGER. 1996. SAS System for Mixed Models.
SAS Institute Inc., Cary, NC.
Lu, Y. J., AND M. J. ADANG. 1996. Distinguishing fall ar-
myworm (Lepidoptera: Noctuidae) strains using a
diagnostic mitochondrial DNA marker. Florida En-
tomol. 79: 48-55.
Lu, Y. J., G. D. KOCHERT, D. J. ISENHOUR, AND M. J.
ADANG. 1994. Molecular characterization of a strain-
specific repeated DNA sequence in the fall army-
worm Spodoptera frugiperda (Lepidoptera: Noctu-
idae). Insect Mol. Biol. 3: 123-130.
LUGINBILL, P. 1928. The fall armyworm. U.S. Dept. Ag-
ric. Tech. Bull. 34: 1-91.
MASON, L. J. 1987. The laboratory as an altered habitat:
phenotypic and genetic consequences of colonization.
Florida Entomol. 70: 49-58.
MCMICHAEL, M., AND D. P. PROWELL. 1999. Differences
in amplified fragment-length polymorphisms in fall
armyworm (Lepidoptera: Noctuidae) host strains.
Ann. Entomol. Soc. Am. 92: 175-181.
MEAGHER, R. L., AND M. GALLO-MEAGHER 2003. Identi-
fying host strains of fall armyworm (Lepidoptera:
Noctuidae) in Florida using mitochondrial markers.
Florida Entomol. 86: 450-455.
MILLER, P. R., W. L. GRAVES, W. A. WILLIAMS, AND B. A.
MADSON. 1989. Covercrops for California agricul-
ture. Univ. of Calif. Div. Agric. Nat. Res. leaflet
21471. 24 pp.
MITCHELL, E. R. 1979. Migration by Spodoptera exigua
and Spodoptera frugiperda, North America style, pp.
386-393 In R. L. Rabb & G. G. Kennedy [eds.], Move-
ment of Highly Mobile Insects: Concepts and Meth-
odology in Research. Raleigh, NC.
MITCHELL, E. R., J. N. MCNEIL, J. K. WESTBROOK, J. F.
SILVAIN, B. LALANNE-CASSOU, R. B. CHALFANT, S. D.
PAIR, V. H. WADDILL, A. SOTOMAYOR-RIOS, AND F. I.
PROSHOLD. 1991. Seasonal periodicity of fall army-
worm, (Lepidoptera: Noctuidae) in the Caribbean
basin and northward to Canada. J. Entomol. Soc. 26:
39-50.
NAGOSHI, R. N., AND R. L. MEAGHER 2003. FR tandem-
repeat sequence in fall armyworm (Lepidoptera:
Noctuidae) host strains. Ann. Entomol. Soc. Am. 96:
329-335.
December 2004
PAIR, S. D., AND J. K. WESTBROOK. 1995. Agro-ecological
and climatological factors potentially influencing ar-
myworm populations and their movement in the
southeastern United States. Southwestern Entomol.
18: 101-118.
PAIR, S. D., J. R. RAULSTON, A. N. SPARKS, J. K. WEST-
BROOK, AND G. K. DOUCE. 1986. Fall armyworm dis-
tribution and population dynamics in the
southeastern states. Florida Entomol. 69: 468-487.
PAIR, S. D., J. R. RAULSTON, J. K. WESTBROOK, W. W.
WOLF, AND S. D. ADAMS. 1991. Fall armyworm (Lepi-
doptera: Noctuidae) outbreak originating in the lower
Rio Grande Valley, 1989. Florida Entomol. 74:200-213.
PASHLEY, D. P. 1986. Host-associated genetic differenti-
ation in fall armyworm (Lepidoptera: Noctuidae): a
sibling species complex? Ann. Entomol. Soc. Am. 79:
898-904.
PASHLEY, D. P. 1988. Quantitative genetics, develop-
ment, and physiological adaptation in host strains of
fall armyworm. Evolution 42: 93-102.
PASHLEY, D. P., T. N. HARDY, AND A. M. HAMMOND.
1995. Host effects on developmental and reproduc-
tive traits in fall armyworm strains (Lepidoptera:
Noctuidae). Ann. Entomol. Soc. Am. 88: 748-755.
PASHLEY, D. P., S. J. JOHNSON, AND A. N. SPARKS. 1985.
Genetic population structure of migratory moths:
the fall armyworm (Lepidoptera: Noctuidae). Ann.
Entomol. Soc. Am. 78: 756-762.
PITRE, H. N., J. E. MULROONEY, AND D. B. HOGG. 1983.
Fall armyworm (Lepidoptera: Noctuidae) oviposi-
tion: crop preferences and egg distribution on plants.
J. Econ. Entomol. 76: 463-466.
RAULSTON, J. R., S. D. PAIR, A. N. SPARKS, J. LOERA G.,
F. A. PEDRAZA M., A. PALAMON T., A. ORTEGA, J.
RuIZ SANCHEZ M., P. MARQUEZ C., H. RULES A., J.
PEREZ M., R. RODRIGUEZ R., H. CARRILLO R., R. AR-
CHUNDIA R., AND F. HERRERA R. 1986. Fall army-
worm distribution and population dynamics in the
Texas-Mexico Gulf Coast area. Florida Entomol. 69:
455-468.
RICH, J., D. WRIGHT, J. MAROIS, AND D. SPRENKEL. 2003.
Selected legumes used as summer cover crops. Univ. of
Florida, IFAS, Florida Coop. Ext. Serv. ENY-688.
ROJAS, J. C., A. VIRGEN, AND L. CRUZ-LOPEZ. 2003.
Chemical and tactile cues influencing oviposition of a
generalist moth, Spodoptera frugiperda (Lepidoptera:
Noctuidae). Environ. Entomol. 32: 1386-1392.
THOMSON, M. S., AND J. N. ALL. 1982. Oviposition by the
fall armyworm onto stake flags and the influence of
flag color and height. J. Georgia Entomol. Soc. 17:
206-210.
THOMSON, M. S., AND J. N. ALL. 1983. Distribution of
fall armyworm egg masses on sweet corn. J. Georgia
Entomol. Soc. 18: 219-224.
VEENSTRA, K. H., D. P. PASHLEY, AND J. A. OTTEA. 1995.
Host-plant adaptation in fall armyworm host
strains: comparison of food consumption, utilization,
and detoxication enzyme activities. Ann. Entomol.
Soc. Am. 88: 80-91.
WESTBROOK, J. K., AND A. N. SPARKS. 1986. The role of
atmospheric transport in the economic fall army-
worm (Lepidoptera: Noctuidae) infestations in the
southeastern United States in 1977. Florida Ento-
mol. 69: 492-502.
WHITFORD, F., S. S. QUISENBERRY, T. J. RILEY, AND J.
W. LEE. 1988. Oviposition preference, mating com-
patibility, and development of two fall armyworm
strains. Florida Entomol. 71: 234-243.
Florida Entomologist 87(4)
Molina-Ochoa et al.: Hymenopteran Parasitoids of Spodoptera frugiperda Larva 461
NATURAL DISTRIBUTION OF HYMENOPTERAN PARASITOIDS
OF SPODOPTERA FRUGIPERDA (LEPIDOPTERA: NOCTUIDAE)
LARVAE IN MEXICO
JAIME MOLINA-OCHOA1, JAMES E. CARPENTER2, ROBERTO LEZAMA-GUTIERREZ', JOHN E. FOSTER3,
MARTIN GONZALEZ-RAMiREZ1, CESAR ANDRES ANGEL-SAHAGUN' AND JAVIER FARiAS-LARIOS'
'Universidad de Colima, Facultad de Ciencias Biol6gicas y Agropecuarias
Apartado postal 36, Tecoman, Colima 28100, M6xico
2United States Department of Agriculture, Agricultural Research Service
Crop Protection & Management Research Laboratory, P.O. Box 748, Tifton, GA 31793-0748, USA
3University of Nebraska Lincoln, Department of Entomology
312F Plant Industry Building, Lincoln, NE 68583-0816, USA
ABSTRACT
A survey of parasitoids of fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith), larvae
was conducted in six Mexican states during August and September 2000. Thirteen genera of
hymenopteran parasitoids were recovered representing the following 3 families, Braconidae:
Aleoides, Chelonus, Cotesia, Glyptapanteles, Homolobus, and Meteorus; Ichneumonidae:
Campoletis, Eiphosoma, Ophion, and Pristomerus; and Eulophidae: Aprostocetus, Euplec-
trus, and Horismenus. Out of 5591 FAW larvae collected, 772 produced parasitoids, for a par-
asitism rate of 13.8%. The highest rate of parasitism from a single collection was 42.2%,
representing three species of parasitoids in Michoacan. Chelonus insularis Cresson was the
most widely distributed species occurring in 45.3% of the locations. Pristomerus spinator (F.),
and Meteorus laphygmae (Viereck), exhibited the highest rates of parasitism for a single col-
lection with 22.2% and 22.1%, in Sinaloa, and Michoacan, respectively. The results supported
the hypothesis that natural distribution and rates of parasitism of FAW larvae may be re-
lated to more diverse habitats with more forests, orchards, and pastures near to cornfields.
Key Words: fall armyworm, Chelonus, Pristomerus, Meteorus, Ophion, Campoletis, corn, survey.
RESUME
Se llev6 a cabo un inventario de parasitoides de larvas del gusano cogollero, Spodoptera fru-
giperda (J. E. Smith) (FAW) colectadas principalmente de maizales en estado de verticilio en
seis estados mexicanos durante Agosto y Septiembre de 2000. Trece g6neros de parasitoides
himen6pteros fueron recuperados, representando a tres families, Braconidae: Aleoides, Che-
lonus, Cotesia, Glyptapanteles, Homolobus, y Meteorus); Ichneumonidae: Campoletis, Eipho-
soma, Ophion, y Pristomerus; y Eulophidae: Aprostocetus, Euplectrus, y Horismenus. De un
total de 5591 larvas colectadas, 772 produjeron parasitoides, para una tasa de parasitismo
de 13.8%. La tasa de parasitismo mas alta para una colecta simple fu6 de 42.2%, repre-
sentando a tres species de parasitoides in Michoacan. La especie mas ampliamente distri-
buida fu6 Chelonus insularis Cresson, presentandose en 45.3% de las localidades
inventariadas. Pristomerus spinator (F.), y Meteorus laphygmae (Viereck), mostraron las
tasa mas altas de parasitismo para una colecta simple con 22.2% y 22.1%, en Sinaloa, y Mi-
choacan, respectivamente. Los resultados apoyan la hip6tesis de que la distribuci6n natural
y las tasas de parasitismo pueden estar relacionadas a lo divers de los habitat con la cer-
cania de mas bosques, huertas y pastizales a los maizales.
Translation provided by the authors.
The therapeutic approach of killing pest or- ternatives to chemical control of insect pests
ganisms with toxic chemicals has prevailed as a based on health, environmental, wild life, and
pest control strategy for over 50 years (Lewis et economic concerns (Johnson et al. 1998; Mattsson
al. 1997). In the 1950s environmental effects of et al. 2000; Solomon & Schettler 2000).
persistent organochlorine insecticides such as Native insects and pathogens are normal parts
DDT began to be observed. Currently, in agricul- of functioning agro-ecosystems and can pro-
tural pest control, the adverse effects of the use of foundly influence the agricultural structure, spe-
insecticides are leading scientists to search for al- cies composition, and diversity. Agro-ecosystems
Florida Entomologist 87(4)
exhibit high biodiversity, mainly influenced by
crops, weeds, microorganisms, and arthropods,
but these factors are also influenced by geograph-
ical location, soil, and climatic characteristics, as
well as human factors. Scientific evidence sug-
gests that biodiversity can be used for improved
pest management (Altieri 1991). The increased
use of beneficial insects and interference with the
colonization of fall armyworm in multiple crop-
ping systems have prevented outbreaks in Latin
America (Altieri 1994).
The fall armyworm (FAW), Spodoptera fru-
giperda (J. E. Smith), is a voracious pest inflicting
damage to a multiplicity of annual crops in the
Americas, and it is commonly controlled with syn-
thetic insecticides, although insecticide resistance
has been observed and is a concern (Yu 1991,
1992). Moreover, two strains of FAW have been
identified according to their host preference, a
corn-associated strain that feeds principally on
corn, and a rice-associated strain that feeds prima-
rily on forage grasses and rice (Pashley et al. 1987).
Both FAW strains exhibited differences in resis-
tance to chemical and biological insecticides (Ad-
amczyk et al. 1997; L6pez-Edwards et al. 1999),
and have differences in their genetic population
structure and population ecology (Pashley 1988;
Lu & Adang 1996; Bossart & Prowell 1998; Levy et
al. 2002; Meagher & Gallo-Meagher 2003; Nagoshi
& Meagher 2003). These differences between FAW
strains complicate the management of this pest.
Biological control is a highly desirable alterna-
tive to insecticides for controlling FAW infesta-
tions (Gross & Pair 1986). The value of
parasitoids in reducing larval populations of this
noctuid has long been recognized (Luginbill 1928;
Vickery 1929). In order to develop a better under-
standing of the natural distribution of the FAW
parasitoid complex and natural enemies, surveys
have been carried out in different regions of Mex-
ico (Carrillo 1980; Lezama-Gutierrez et al. 2001;
Molina-Ochoa et al. 2001, 2003a).
Here, we report the natural distribution of par-
asitoids of FAW larvae collected from whorl-stage
corn, grain sorghum, forage sorghum, and Sudan
grass fields from five Mexican states in the Pacific
coast and one state in the Gulf of Mexico, during
the summer of 2000.
MATERIALS AND METHODS
During August and September of 2000, S. fru-
giperda larvae were collected from whorl-stage
corn, grain and forage sorghum, and Sudan grass
fields in 64 locations in the Mexican Pacific coast
states of Sinaloa, Nayarit, Jalisco, Colima, and
Michoacan, and in the Gulf of Mexico state of Ve-
racruz. Egg masses and pupae were not collected.
FAW larvae were individually placed into 30-
cc plastic cups with pinto bean diet (Burton &
Perkins 1989), and held in the laboratory (Labo-
ratory of Biological Control, Universidad de
Colima, Facultad de Ciencias Biol6gicas y
Agropecuarias, Tecoman, Colima, Mexico) for
emergence of parasitoids (Molina-Ochoa et al.
2001). Adult parasitoids were placed in 70% eth-
anol and then submitted to the USDA/ARS Sys-
tematic Entomology Laboratory, Beltsville, MD
for identification. Collection size ranged from 33
to 119 FAW larvae. The number collected was cor-
rected by subtracting the number that died from
injury or unknown causes during the first few
days after collection before calculating percent
parasitism. Mortality due to pathogens and para-
sitic nematodes has been previously reported
(Molina-Ochoa et al. 2003a).
Collection dates, geographic location, altitude,
crop, sample size and total parasitism of FAW lar-
vae in six Mexican states are presented in the Ta-
ble 1. A Garmin GPS III Plus- was used for
obtaining the coordinates and altitude data.
RESULTS AND DISCUSSION
Out of 5591 FAW larvae collected, 772 pro-
duced parasitoids, for a parasitism rate of 13.8%.
These parasitoids represented 13 genera from
three families of Hymenoptera: six Braconidae,
four Ichneumonidae, and three Eulophidae. Nine
of the 64 collections produced no parasitoids, six
of 12 collections from whorl-stage corn in Micho-
acan, two of 13 in Jalisco, and only one of 11 in
Colima. The highest rates of parasitism in each
state were found in C4 (33.3%) in Colima, J12
(21.1%)in Jalisco, M12 (14.4%) in Michoacan, N9
(18.9%) in Nayarit, S5 (27.4%) in Sinaloa, and V4
(11.5%) in Veracruz (Table 1). The most diverse
collections of parasitoids were found in the loca-
tions C5, J12, and N9 with 5, 4, and 4 species, re-
spectively, (Tables 2 and 3). The collection from S5
produced the highest rate of parasitism for a sin-
gle species with 22.1%; the braconid Meteorus la-
phygmae Viereck was the most common
parasitoid collected from Sudangrass. Other par-
asitoids in that collection were the eulophid Eu-
plectrus plathypenae Howard (2 individuals), and
the ichneumonid Ophion flavidus Brulle (1 indi-
vidual). The braconid C. insularis occurred in 29
of the 64 collections from the six states, and it was
the most widely distributed parasitoid. Another
important braconid was M. laphygmae, occurring
in 21 of the 64 collections. The ichneumonid para-
sitoids, 0. flavidus, and Pristomerus spinator F.,
occurred in 18, and 17 of the 64 collections, re-
spectively. E. plathypenae was the most impor-
tant and widely distributed eulophid, occurring in
16 of the 64 collections (Tables 2 and 3).
Chelonus insularis was the most widely dis-
tributed parasitoid of FAW larvae in this survey,
occurring in all the six Mexican States, and it was
the braconid species with the second highest par-
asitism rate per location with 16.7%. Thus, C. in-
December 2004
Molina-Ochoa et al.: Hymenopteran Parasitoids ofSpodoptera frugiperda Larva 463
TABLE 1. GEOGRAPHIC LOCATION, DATE, ALTITUDE, CROP (*), SAMPLE SIZE (N), AND TOTAL PERCENT SPODOPTERA FRU-
GIPERDA LARVAE PARASITIZED IN SIX MEXICAN STATES (**) DURING 2000.
Code
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
Cll
J1
J2
J3
J4
J5
J6
J7
J8
J9
J10
Jll
J12
J13
M1
M2
M3
M4
Date
08/04
08/04
08/04
08/06
08/06
08/06
08/06
08/06
08/06
08/07
08/07
08/08
08/15
08/15
08/15
08/17
08/17
08/17
08/17
08/17
08/17
08/18
08/18
08/23
08/09
08/09
08/09
08/10
Location
El poblado, Coquimatlin
Pueblo Juarez, Coquimatlin
Amachico, Coquimatlin
Los mezcales, Comala
El remate, Comala
Carrizalillo, Queseria
Queseria
Villa de Alvarez
Juluapan, Villa de Alvarez
Tepames, Colima
Estapilla, Colima
Ciudad Guzman
Los pinitos, Tonila
Pialla, Tuxpan
Atenquique, Tuxpan
Canoas, Zapotiltic
Apastepe
Teocuitatlan
Zacoalco de Torres
Acatlan de Juarez
Tlajomulco de Zifuiga
Zapopan
Magdalena
Crucero de Magdalena
Totolin
Santa In6s Tocumbo
Peribin
Cointzio
Coordinates
193.698'N
10347.722'W
1910.752'N
10354.634'W
1910.667'N
103 56.351'W
1920.811'N
10347.176'W
1924.825'N
10347.639'W
1925.389'N
10341.000'W
1923.362'N
10334.882'W
1917.201'N
10347.030'W
1918.890'N
103 49.611'W
1908.231'N
10337.996'W
1859.549'N
103 31.140'W
19 40.011'N
10328.830'W
1925.343'N
10332.447'W
1927.293'N
10328.514'W
1931.778'N
10327.851'W
1934.073'N
10327.324'W
1938.060'N
10330.950'W
2007.035'N
10332.704'W
2011.988'N
10333.806'W
2025.362'N
10333.406'W
2029.396'N
103 28.298'W
2043.129'N
10329.041'W
2053.008'N
10355.477'W
2056.300'N
10402.509'W
1958.890'N
10240.183'W
1944.502'N
10234.967'W
1933.106'N
102 26.586'W
1941.609'N
10116.398'W
'Corn (c), gran sorghum (gs), forage sorghum (fs), and Sudan grass (sg).
*Colima (C), Jalisco (J), Michoacan (M) Nayarit (N), Sinaloa (S), and Veracruz (V).
Alt (m)
422
279
328
608
817
1550
1304
515
539
519
304
1557
1326
1079
1338
1391
1709
1369
1425
1575
1607
1670
1496
1386
1590
1630
1475
1932
Percentage
* N parasitized
C 90 17.8
C 90 4.4
C 90 12.2
C 90 33.3
C 90 13.3
C 90 1.1
c 90 10.0
c 90 4.4
c 90 4.4
c 90 0.0
c 90 21.1
c 90 0.0
c 90 2.2
c 90 0.0
c 90 1.1
c 90 3.3
c 90 1.1
c 90 10.0
c 90 4.4
c 96 2.1
c 92 4.3
c 90 4.4
c 93 21.5
c 92 2.2
c 90 0.0
c 90 1.1
c 90 1.1
c 90 0.0
Florida Entomologist 87(4)
December 2004
TABLE 1. (CONTINUED) GEOGRAPHIC LOCATION, DATE, ALTITUDE, CROP (*), SAMPLE SIZE (N), AND TOTAL PERCENT
SPODOPTERA FRUGIPERDA LARVAE PARASITIZED IN SIX MEXICAN STATES (**) DURING 2000.
Code
M5
M6
M7
M8
M9
M10
M1l
M12
N1
N2
N3
N4
N5
N6
N7
N8
N9
N10
S1
S2
S3
S4
S5
S6
S7
S8
V1
V2
Date
08/10
08/11
08/11
08/11
08/11
08/12
08/12
08/12
08/18
08/18
08/18
08/19
08/19
08/19
08/19
08/20
08/23
08/23
08/21
08/21
08/21
08/21
08/22
08/22
08/22
08/22
09/02
09/02
Location
Cerro "La Esperanza"
Tejabin
Carretera a Nueva Italia
Presa de Zicuirin
El cenidor, Nueva Italia
La Guadalupe Paracuaro
Las yeguas Paracuaro
El cirian, Nueva Italia
Santa Maria del Oro
El rinc6n, Tepic
El pich6n, Tepic
Xalisco
El refili6n, Xalisco
Compostela
La presa, Compostela
Las lumbres, Acaponeta
Seboruco
Ahuacatlin
Bacurimi, Culiacan
La campana, Culiacan
Pericos, Mocorito
Rancho viejo, Mocorito
Aguapepito, Mocorito
Comanito, Mocorito
La poma, Badiraguato
La majada, Badiraguato
Seis de Enero, Xalapa
Altolucero, Almolonga
Coordinates
1941.223'N
10118.980'W
1913.342'N
10153.714'W
1903.290'N
10202.458'W
1856.191'N
10154.650'W
1859.651'N
10211.577'W
1907.472'N
10212.519'W
1857.308'N
10216.733'W
1853.661'N
10207.483'W
2120.121'N
10440.174'W
2132.472'N
10456.123'W
2133.479'N
10456.937'W
2119.601'N
10455.060'W
2119.407'N
10455.323'W
2117.858'N
10454.044'W
2113.714'N
10452.162'W
2220.795'N
105 18.141'W
2120.850'N
104 40.749'W
2106.331'N
10427.427'W
2451.668'N
10729.478'W
2458.415'N
10733.517'W
2503.574'N
10739.547'W
2506.033'N
10743.165'W
2503.861'N
10739.547'W
2509.006'N
10739.645'W
2515.749'N
107 40.739'W
2514.076'N
107 39.781'W
1934.115'N
9650.207'W
1935.063'N
9647.384'W
Alt (m)
1998
587
442
292
350
540
359
255
1160
849
774
1042
964
920
928
48
1134
1120
70
143
80
89
68
91
157
145
950
908
"Corn (c), gran sorghum (gs), forage sorghum (fs), and Sudan grass (sg).
*Colima (C), Jalisco (J), Michoacan (M) Nayarit (N), Sinaloa (S), and Veracruz (V).
N
c 90
c 90
c 90
c 63
c 57
fs 90
fs 90
c 90
c 90
c 96
c 95
c 107
c 90
c 93
c 90
C&gs 60
c 90
c 90
gs 97
gs 100
gs 95
gs 98
sg 95
gs 95
c 100
c 92
c 91
c 33
Percentage
parasitized
1.1
0.0
0.0
0.0
1.8
1.1
1.1
42.2
3.3
10.4
4.2
2.8
8.9
1.1
1.1
5.0
18.9
5.6
4.1
5.0
9.5
13.3
27.4
3.2
13.0
7.6
6.6
12.1
Molina-Ochoa et al.: Hymenopteran Parasitoids ofSpodoptera frugiperda Larva 465
TABLE 1. (CONTINUED) GEOGRAPHIC LOCATION, DATE, ALTITUDE, CROP (*), SAMPLE SIZE (N), AND TOTAL PERCENT
SPODOPTERA FRUGIPERDA LARVAE PARASITIZED IN SIX MEXICAN STATES (**) DURING 2000.
Coordinates Alt (m)
Percentage
* N parasitized
V3 09/02 Actopan
V4 09/02 Los Gonzalez, Actopan
V5 09/02 Bocana, Actopan
V6 09/03 El volador, Coatepec
V7 09/03 Palmillas
V8 09/03 Tierra Colorada
V9 09/04 Cerro gordo
V10 09/04 La cumbre
1934.623'N
9648.589'W
1931.894'N
9641.294'W
1924.416'N
9636.731'W
1921.594'N
9651.037'W
1912.293'N
9646.221'W
1913.255'N
9621.916'W
1925.252'N
9639.566'W
1923.320'N
9638.807'W
775 c 64 3.1
432 c 113
311 c 119 4.2
709 c 90 3.3
702 c 59 6.8
46 c 45 4.4
443 c 45 8.9
366 c 66 6.1
*Corn (c), gran sorghum (gs), forage sorghum (fs), and Sudan grass (sg).
**Colima (C), Jalisco (J), Michoacan (M) Nayarit (N), Sinaloa (S), and Veracruz (V).
sularis is one of the most abundant natural
enemies of fall armyworm larvae in the Western
Coast and Gulf of Mexico. Chelonus insularis has
been reported as an important parasitoid control-
ling FAW populations in the US (Lugingill 1928;
Vickery 1929). Ashley (1986) and Andrews (1988)
listed C. insularis occurring in Central America
and the US, highlighting its role as parasitoid of
FAW in southern Florida where 63% of the FAW
larvae were attacked. Recently, Molina-Ochoa et
al., (2003b) reported C. insularis syn. C. texanus
as the braconid with the broadest distribution in
Latin America, including South America (Uru-
guay and Venezuela), the Caribbean Basin (Trin-
idad and Puerto Rico), and the US. In that
inventory Chelonus sp. is also reported in Brazil,
Mexico, and Peru. Lewis and Nordlund (1980)
emphasized its role considering it as an excellent
candidate for the following augmentative ap-
proaches: a) release throughout its overwintering
zone; b) early-season colonization, and c) direct
therapeutic release on target crops.
In a previous survey, Molina-Ochoa et al.
(2001) commented on the importance and need of
more study in Mexico on the taxonomy of the ge-
nus Chelonus (P. M. Marsh, pers. comm.).
Meteorus laphygmae occurred in 21 of the 64
collections. The highest rate of parasitism for a
single location was obtained in S5 with 22.1%.
This parasitoid occurred in all of the collections
from Sinaloa, and the rate of parasitism ranged
from 2.1 to 22.1%. Meteorus laphygmae was also
collected in Colima, Nayarit, Michoacan, Jalisco,
and Veracruz occurring in 45.5%, 30%, 25%, 10%,
and 8.3% of the collections, respectively. This bra-
conid was reported by Ashley (1986) occurring in
the Continental US, exhibiting its greatest im-
pact on FAW collected from grass. Other reports
were made by Alvarado-Rodriguez (1987) in Si-
naloa, Mexico attacking Spodoptera exigua (Hiib-
ner) infesting tomatoes with a parasitism rate of
9.0%. A similar rate of parasitism was reported
by Molina-Ochoa et al. (2001) in a single collec-
tion of FAW larvae made in El Mante, Tamauli-
pas with 10.3%. Molina-Ochoa et al. (2003b)
listed several reports from countries of Central
and South America, such as Honduras, Nicara-
gua, Mexico, Chile, Colombia, and Suriname,
where M. laphygmae was collected from other
crops such as maize, rice, cotton, sorghum, pea-
nuts, and Bermudagrass, and was one of the most
prevalent parasitoids in South America.
Low rates of occurrence and parasitization of
Cotesia sp. probably marginiventris (Cresson),
Glyptapanteles sp. probably militaris (Walsh),
Aleiodes sp., and Homolobus sp. probably mellea
(Cresson) were recorded. They were found in 5, 2,
1, and 1 of the 64 collections, respectively.
Cotesia sp. occurred in Colima, Jalisco, Na-
yarit with lower parasitization rates than 2.3%.
Similar rates were reported by Molina-Ochoa et
al. (2001) in a previous survey conducted in four
Mexican States. This parasitoid is reported at-
tacking FAW larvae in Argentina, Brazil, Chile,
Honduras, Lesser Antilles, Mexico, Nicaragua,
Puerto Rico, Suriname (Molina-Ochoa et al.
2003b), but it has been often reported as a parasi-
toid of FAW in the US (Ashley 1986) with parasit-
ization rates of 6.3% on FAW larvae collected
from maize (Riggin et al. 1993) and from less than
1% to 40% collected from maize and Bermuda-
grass, respectively (Ashley et al. 1983).
Code
Date
Location
Florida Entomologist 87(4)
December 2004
TABLE 2. PERCENTAGE OF SPODOPTERA FRUGIPERDA LARVAE PARASITIZED BY EACH SPECIES OF BRACONIDAE AT EACH
LOCATION.
Braconidae
Code* Aleoides Chelonus Cotesia C. ....r,....r .. Homolobus Meteorus
Aleiodes sp., Chelonus sp. Probably insularis Cresson, Cotesia sp. probably marginiventris Cresson, Glyptapanteles sp. probably
militaris Walsh, Homolobus sp. probably mellea Cresson, Meteorus sp. probably laphygmae Viereck.
Molina-Ochoa et al.: Hymenopteran Parasitoids ofSpodoptera frugiperda Larva 467
TABLE 2. (CONTINUED) PERCENTAGE OF SPODOPTERA FRUGIPERDA LARVAE PARASITIZED BY EACH SPECIES OF BRA-
CONIDAE AT EACH LOCATION.
Braconidae
Code* Aleoides Chelonus Cotesia Ci. ,..i.., .... .. Homolobus Meteorus
S7 0.0 2.0 0.0 0.0 0.0 10.0
S8 0.0 0.0 0.0 0.0 0.0 6.5
V1 0.0 3.3 0.0 0.0 0.0 1.1
V2 0.0 0.0 0.0 0.0 0.0 0.0
V3 0.0 0.0 0.0 0.0 0.0 0.0
V4 0.0 0.0 0.0 0.0 0.0 0.0
V5 0.0 0.0 0.0 0.0 0.0 0.0
V6 0.0 1.1 0.0 0.0 0.0 0.0
V7 0.0 0.0 0.0 1.7 0.0 0.0
V8 0.0 0.0 0.0 0.0 0.0 0.0
V9 0.0 0.0 0.0 0.0 0.0 0.0
V10 0.0 0.0 0.0 0.0 0.0 0.0
Aleiodes sp., Chelonus sp. Probably insularis Cresson, Cotesia sp. probably marginiventris Cresson, Glyptapanteles sp. probably
militaris Walsh, Homolobus sp. probably mellea Cresson, Meteorus sp. probably laphygmae Viereck.
Glyptapanteles sp. was found in Nayarit in two
collections, N2 and N4, with parasitization rates
of 8.3% and 1.9%, respectively, and in one location
in Veracruz (V7) with 1.7% of parasitism rate.
Rohlfs & Mack (1985), and Cave (1993) reported
the occurrence of this parasitoid attacking FAW
larvae in the US and Honduras, collected from
sorghum and maize, respectively. Steffey (2001)
reported G. militaris attacking armyworms and
other caterpillars in Illinois. He speculated that
this braconid and other natural enemies could
suppress armyworm populations and keep them
well below economic levels. Recently, Reis et al.
(2003) suggested that the parasitoid may be well
adapted to the Azorean agricultural systems in
Portugal, characterized by prevalence of the
grass, Lolium perenne L., throughout the year.
The armyworm, Pseudaletia unipuncta (Haworth)
when fed on fresh leaves ofL. perenne is the most
suitable host for the mass rearing of this braconid.
Aleiodes sp. occurred only in one collection in
Nayarit (N2), and Homolobus sp. was found in
Michoacan (M3), and their parasitism was lower
than 1.2%. Ruiz-Cancino (1991) reported species
of Rogas (Syn: Aleiodes) occurring in "La Reserva
de la Biosfera El Cielo" in Tamaulipas, Mexico,
and the family Braconidae is the second more
abundant with 10% of the individuals, these bra-
conids were attacking insect pest of annual,
perennial and ornamental crops. Aleiodes laphyg-
mae was reported by Molina-Ochoa et al. (2001)
with a low parasitism rate (0.3%) on FAW larvae
in Tamaulipas, Mexico. This braconid, A. laphyg-
mae was the most abundant parasitoid attacking
FAW larvae (12.8% parasitism) in South Georgia
(Riggin et al. 1993).
Homolobus sp. probably mellea (Cresson), syn:
Zele mellea (Cresson) was previously found in
small numbers attacking FAW larvae in Hondu-
ras (Cave 1993), Nicaragua (Huis 1981) and the
US (Vickery 1929; Wilson 1933; Riggin et al.
1992), but was not previously reported in Mexico.
Parasitism by this species was low (1.1%), but
finding it contributes to our knowledge on the oc-
currence and diversity of beneficial insects affect-
ing FAW populations in Michoacan.
The ichneumonid parasitoids, 0. flavidus, P
spinator, and C. flavicincta were the most fre-
quently reared species in 18, 17, and 14 of the 64
collections, respectively. Ophion flavidus was re-
covered in more locations in Michoacan, and
Colima (5 and 4 locations, respectively), but the
highest parasitism rate for a single location was
obtained in Colima (C7) with 6.7%. Similar results
were reported by Molina et al. (2001), and Riggin
et al. (1993). Recently, Molina-Ochoa et al., (2003b)
listed the occurrence of 0. flavidus in Argentina,
Brazil, Honduras, Mexico, Nicaragua, and the US.
Ashley et al. (1983) reported that Ophion sp. at-
tacked FAW larvae developing on volunteer corn
and Paragrass at Homestead, Florida. Gross &
Pair (1991) emphasized that the tachinidArchytas
marmoratus (Townsend) and 0. flavidus provide
opportunities for advancing biological strategies
for managing FAW, with the development of eco-
nomical methods for mass-propagation.
P spinator was the second most widely distrib-
uted ichneumonid parasitoid. It was recovered in
17 of the 64 collections, 7 in Colima, 2 in Jalisco, 4
in Michoacan, and Nayarit, but this species was not
recovered from Sinaloa, and Veracruz. The highest
rate of parasitism for a single location was obtained
in Michoacan (M12) with 22.2%. Pristomerus
spinator has been reported in Mexico occurring in
Quintana Roo, Tamaulipas (Carrillo 1980), and Mi-
choacan, Colima, and Jalisco (Molina-Ochoa et al.
Florida Entomologist 87(4)
December 2004
TABLE 3. PERCENTAGE OF Spodoptera frugiperda LARVAE PARASITIZED BY EACH SPECIES OF ICHNEUMONIDAE AND EU-
LOPHIDAE AT EACH LOCATION.
Ichneumonidae Eulophidae
Code* C.f E.v O.f P.s A.sp E.p H.sp
: Pristomerus spina-
C.f= Campoletis flavicincta Ashmead, E.v = Eiphosoma vitticolle Cresson, O.f= Ophion flavidus Brulle, P.s
tor Fabricius,A.sp. = Aprostocetus sp., E.p = Euplectrus plathypenae Howard, H.sp. = Horismenus sp.
Molina-Ochoa et al.: Hymenopteran Parasitoids ofSpodoptera frugiperda Larva 469
TABLE 3. (CONTINUED) PERCENTAGE OF Spodoptera frugiperda LARVAE PARASITIZED BY EACH SPECIES OF ICHNEU-
MONIDAE AND EULOPHIDAE AT EACH LOCATION.
Ichneumonidae Eulophidae
Code* C.f E.v O.f P.s A.sp E.p H.sp
S7 0.0 0.0 0.0 0.0 0.0 1.0 0.0
S8 0.0 0.0 0.0 0.0 0.0 1.1 0.0
V1 0.0 0.0 0.0 0.0 0.0 2.2 0.0
V2 0.0 0.0 0.0 0.0 3.0 6.1 3.0
V3 0.0 0.0 0.0 0.0 0.0 3.1 0.0
V4 0.0 0.0 0.0 0.0 0.0 11.5 0.0
V5 0.0 0.0 0.0 0.0 0.0 4.2 0.0
V6 0.0 0.0 0.0 0.0 0.0 2.2 0.0
V7 0.0 0.0 0.0 0.0 0.0 5.1 0.0
V8 0.0 0.0 0.0 0.0 0.0 4.4 0.0
V9 0.0 0.0 0.0 0.0 0.0 8.9 0.0
V10 0.0 0.0 0.0 0.0 0.0 6.1 0.0
C.f= Campoletis flavicincta Ashmead, E.v = Eiphosoma vitticolle Cresson, O.f= Ophion flavidus Brulle, P.s = Pristomerus spina-
tor Fabricius, A.sp. = Aprostocetus sp., E.p = Euplectrus plathypenae Howard, H.sp. = Horismenus sp.
2001). Two collections from Michoacan during 1998
and 2000 exhibited the highest parasitism rates for
a single location (El Hueso, and El Cirian, Nueva
Italia) with 12.7%, and 22.2%, respectively. The
ichneumonid was previously reported from Brazil,
Honduras, Mexico, Nicaragua, and the US (Molina-
Ochoa et al. 2003b).
Campoletis flavicincta was found in 14 of 64
collections, one in Colima, 5 in Jalisco, 3 in Micho-
acan, and 5 in Nayarit, but it was not recovered in
Sinaloa, and Veracruz. Campoletis flavicincta
had an overall parasitism range from 0 to 3.3%.
The highest parasitism rate for a single location
was obtained in N9. In a previous survey con-
ducted by Molina-Ochoa et al. (2001), C. flau-
icincta accounted for 23% of parasitism in El
Batillero, Michoacan, a location surrounded by
avocado orchards and pine forest near to Apo, Mi-
choacan; however, the FAW larvae from nearby
locations in this survey (Ml and M2) showed low
parasitism rates (3.3%, and 2.2%, respectively) by
this parasitoid. It appears that, C. flavicincta,
prefers or was associated with locations with high
altitude; in this survey, it was found in locations
with altitudes with an average of 1417 meters, as
well as in locations near forests mainly consti-
tuted with pine and oak trees. Molina-Ochoa et
al. (2003b) reported C. flavicincta occurring in
Brazil, Honduras, Mexico, Nicaragua, and the
US. This species was also reported attacking beet
armyworm larvae fed on cotton in Georgia, USA
(Ruberson et al. 1993, 1994).
Eiphosoma vitticole was the ichneumonid with
the most limited distribution in this survey,
found in 6 of the 64 collections. E. vitticole oc-
curred in 2 locations in Colima, 3 locations in Mi-
choacan, and 1 location in Nayarit. The highest
rate of parasitism for a single location was re-
corded in M12 with 5.6%. This species showed low
parasitism rates, and it was not found in Jalisco,
Sinaloa, and Veracruz. It was collected from loca-
tions with an average altitude of 472m, with a
range between 255 and 744m. Pair et al. (1986)
reported the occurrence of E. vitticole in Texas,
and Tamaulipas, Mexico. It also has been re-
ported from Bolivia, Brazil, Colombia, Honduras,
and Nicaragua (Molina-Ochoa et al. 2003b)
Three species of eulophid parasitoids were
found in this survey, Aprostocetus sp., Euplectrus
plathypenae Howard, and Horismenus sp. Euplec-
trus plathypenae was the most widely distributed
eulophid, occurring in 16 of the 64 collections. It
was found in Veracruz in all collections (10), Si-
naloa in 3 collections, 2 in Michoacan, and one in
Jalisco. Molina-Ochoa et al. (2001) reported a par-
asitism rate of 8.3% by E. plathypenae in a single
collection in El Mante, Tamaulipas, similar rates
in several locations in Veracruz, and low rate of
about 1% in Michoacan. We also did not find levels
higher than 1.6% in Michoacan; however, we
found a range of parasitism in Sinaloa between
1% and 4.2%. The highest level of parasitism for a
single location was obtained in the location V4
with 11.5%. Montoya-Burgos (1980) reported nat-
ural parasitism of about 15% by Euplectrus sp.
against L2 FAW developing on corn in Veracruz.
Euplectrus plathypenae is an important and well
distributed parasitoid in the tropical Americas,
and the US (Molina-Ochoa et al. (2003b).
The other eulophids, Aprostocetus sp. and
Horismenus sp., occurred only in the location V2,
with a parasitism rate of 3.0% for both species.
This is the first report of Aprostocetus sp. and
Horismenus sp. as parasitoids of FAW larvae.
Aprostocetus sp. has been reported as a hyperpar-
asitoid of Gelechia senticetella (Stgr.) (Lepidoptera:
Florida Entomologist 87(4)
Gelechiidae) fed on Juniperus excelsa in Bulgaria
(Mirchev et al. 2001).Aprostocetus sp. also was re-
ported as an egg parasitoid of mango leafhoppers
(Fasih & Srivatava 1990). Aprostocetus diplosis
Crawford is a parasitoid of Stenodiplosis sorghi-
cola, a dipterous pest of sorghum in Brazil (Cam-
pos et al. 1998). Horismenus sp. has been reported
to be a parasitoid ofprepupae and pupae of the Cit-
rus leafminer, Phyllocnistis citrella (Lepidoptera:
Gracillariidae) in Mexico (Perales et al. 1996, Bau-
tista-Martinez et al. 1998). Coffelt & Schultz
(1993) mentioned that it is very common to find
species of this genus acting as hyperparasitoids.
Our results demonstrate that hymenopteran
parasitoids of FAW differentially occurred
throughout the six Mexican states surveyed.
However, this may have been influenced by the
size of the FAW larvae collected. The hy-
menopteran parasitoids caused significant mor-
tality of FAW larvae in most of the localities of
this survey. It is important to highlight the occur-
rence and role on the FAW larval mortality
caused by the braconids, C. insularis, and M. la-
phygmae, the ichneumonids, 0. flauidus, P spina-
tor, and C. flavicincta, as well as the eulophid E.
plathypenae. Our findings agree with Ashley
(1986) in that no single parasitoid species exerted
significant mortality throughout a major portion
of the range of FAW. Another important aspect to
note is the need for more taxonomic studies on
two genera, Chelonus and Meteorus, which are
important sources of mortality for FAW larvae.
ACKNOWLEDGMENTS
The authors thank Dr. R. W. Carlson, Dr. E. E. Gris-
sell, Dr. P. M. Marsh, Dr. N. E. Woodley, and Dr. N. W.
Gates (USDA-ARS, Systematic Entomology Laboratory
Beltsville, MD) for the insect identifications, and J. J.
Molina-Cardenas, M. A. Rodriguez-Vega, and F. Arceo-
Palacios (Universidad de Colima, Tecoman, Colima) for
assistance during the collections and preparation of the
specimens. The authors express gratitude to Dr. Carlos
Salazar-Silva, Rector of the Universidad de Colima, and
CONACYT-Mexico for supporting this research, and for
a grant to the senior author, respectively. This paper is
a contribution of the Universidad de Colima-Facultad
de Ciencias Biol6gicas yAgropecuarias, Tecoman, Colima,
M6xico, the USDA-ARS Crop Protection & Manage-
ment Research Laboratory, Tifton, GA 31793, and the
University of Nebraska Agricultural Research Division,
Lincoln, NE 68583, Journal Series No.14482, Depart-
ment of Entomology, University of Nebraska Lincoln.
The authors also thank Dr. John J. Hamm USDA-ARS,
Crop Protection & Management Research Laboratory,
P.O. Box 748, Tifton, GA 31793-0748, USA, and Dr. John
R. Ruberson, Department of Entomology, University of
Georgia Tifton, Tifton, GA 31793, USA, for critical re-
view of the manuscript.
REFERENCES CITED
ADAMCZYK, J. J., JR., J. W. HOLLOWAY, B. R. LEONARD,
AND J. B. GRAVES. 1997. Susceptibility of fall army-
worm collected from different plant hosts to selected
insecticides and transgenic Bt cotton. J. Cotton Sci.
1: 21-28.
ALTIERI, M. A. 1991. Increasing biodiversity to improve
insect pest management in agro-ecosystems. Chap-
ter XIV, pp. 165-182 In D. L. Hawksworth [ed.] The
Biodiversity of Microorganisms and Invertebrates:
Its Role in Sustainable Agriculture. CAB Interna-
tional, Wallingford, UK.
ALTIERI, M. A. 1994. Biodiversity and pest management
in agroecosystems. Haworth Press, New York.
ALVARADO-RODRIGUEZ, B. 1987. Parasites and disease
associated with larvae of beet armyworm
Spodoptera exigua (Lepidoptera: Noctuidae), infest-
ing processing tomatoes in Sinaloa, Mexico. Florida
Entomol. 70: 444-449.
ANDREWS, K. L. 1988. Latin American Research on
Spodoptera frugiperda (Lepidoptera: Noctuidae).
Florida Entomol. 71: 630-653.
ASHLEY, T. R., C. S. BARFIELD, V. H. WADDILL, AND E. R.
MITCHELL. 1983. Parasitization of fall armyworm
larvae on volunteer corn, Bermudagrass, and para-
grass. Florida Entomol. 66: 267-271.
ASHLEY, T. R. 1986. Geographical distributions and par-
asitization levels for parasitoids of the fall armyworm,
Spodoptera frugiperda. Florida Entomol. 69: 516-524.
BAUTISTA-MARTINEZ, N., J. L. CARRILLO-SANCHEZ, H.
BRAVO-MOJICA, AND S. D. KOCH. 1998. Natural par-
asitism ofPhyllocnistis citrella (Lepidoptera: Gracil-
lariidae) at Cuitlahuac, Veracruz, M6xico. Florida
Entomol. 81: 30-37.
BOSSART, J. L., AND D. P. PROWELL. 1998. Genetic esti-
mates of population structure and gene flow: limita-
tions, lessons and new directions. Tree 13:202-206.
BURTON, R. L., AND W. D. PERKINS. 1989. Rearing the
corn earworm and fall armyworm for maize resis-
tance studies, pp. 37-45 In CIMMYT, 1989. Toward
Insect Resistant Maize for Third World: Proceedings
of the International Symposium on Methodologies
for Developing Host Plant Resistance to Maize In-
sects, M6xico, D. F., CIMMYT.
CAMPOS, A. R., F. M. LARA, AND O. R. CAMPOS. 1998. In-
fluencia de gen6tipos de sorgo sobre a mosca Steno-
diplosis sorghicola (Diptera: Cecidomyiidae) e seus
parasit6ides Aprostocetus diplosis Crawford, 1907
(Hymenoptera: Eulophidae). Cult. Agron. 7: 91-100.
CARRILLO, H. 1980. Detenninaci6n del parasitismo nat-
ural de gusano cogollero, Spodoptera frugiperda (J.
E. Smith) en Quintana Roo. Folia Entomol. M6x. 45:
111-112.
CAVE, R. D. 1993. Parasitoides larvales y pupales de
Spodoptera frugiperda (Smith) (Lepidoptera: Noctu-
idae) en Centro Am6rica con una clave para las espe-
cies encontradas en Honduras. Ceiba 34: 33-56.
COFFELT, M. A., AND P. B. SCHULTZ. 1993. Larval para-
sitism of orangestriped oakworm (Lepidoptera: Sat-
urnidae) in the urban shade tree environment. Dept.
Entomology. University of Virginia. Biological Con-
trol 3: 127-134.
FASIH, M., AND R. P. SRIVASTAVA. 1990. Parasites and
predators of insect pest of mango. Int. Pest. Control
32: 39-41.
GROSS, H. R., JR., AND S. D. PAIR. 1986. The fall army-
worm: Status and expectations of biological control
with parasitoids and predators. Florida Entomol. 69:
502-515.
GROSS, H. R., AND S. D. PAIR. 1991. Seasonal distribu-
tion, response to host developmental stage, and
December 2004
Molina-Ochoa et al.: Hymenopteran Parasitoids ofSpodoptera frugiperda Larva 471
screened-cage performance ofArchytas marmoratus
(Diptera: Tachinidae) and Ophion flavidus (Hy-
menoptera: Ichneumonidae) on Spodoptera fru-
giperda (Lepidoptera: Noctuidae) Florida Entomol.
74: 237-245.
HUIS, A. V. 1981. Integrated pest management in the
small farmer's maize crop in Nicaragua. Med.
Landb. Wageningen 81: 221 pp.
JOHNSON, D. E., F. J. SEIDLER, AND T. A. SLOTKIN. 1998.
Early biochemical detection of delayed neurotoxicity
resulting from developmental exposure to chlorpyri-
fos. Brain Research Bulletin 45: 143-147.
LEWIS, W. J., AND D. A. NORDLUND. 1980. Employment
of parasitoids and predators for fall armyworm con-
trol. Florida Entomol. 63: 433-438.
LEWIS, W. J., J. C. VAN LENTEREN, S. C. PHATAK, AND J.
H. TUMLINSON, III. 1997. A total system approach to
sustainable pest management. Proc. Natl. Acad. Sci.
94: 12243-12248.
LEZAMA-GUTIERREZ, R., J. J. HAMM, J. MOLINA-OCHOA,
M. LOPEZ-EDWARDS, A. PESCADOR-RUBIO, M.
GONZALEZ-RAMIREZ, AND E. STYER. 2001. Occur-
rence of entomopathogens of Spodoptera frugiperda
(Lepidoptera: Noctuidae) in the Mexican States of
Michoacan, Colima, Jalisco and Tamaulipas. Florida
Entomol. 84: 23-30.
LEVY, H. C., A. GARCIA-MARUNIAK, AND J. E. MARU-
NIAK. 2002. Strain identification of Spodoptera fru-
giperda (Lepidoptera: Noctuidae) insects and cell
line: PCR-RFLP of cytochrome oxidase C subunit I
gene. Florida Entomol. 85: 186-190.
LOPEZ-EDWARDS, M., J. L. HERNANDEZ-MENDOZA, A.
PESCADOR-RUBIO, J. MOLINA-OCHOA, R. LEZAMA-
GUTIERREZ, J. J. HAMM, AND B. R. WISEMAN. 1999.
Biological differences between five populations of
fall armyworm (Lepidoptera: Noctuidae) collected
from corn in Mexico. Florida Entomol. 82: 254-262.
Lu, Y., AND M. J. ADANG. 1996. Distinguishing fall ar-
myworm (Lepidoptera: Noctuidae) strains using a
diagnostic mitochondrial DNA marker. Florida En-
tomol. 79: 48-55.
LUGINBILL, P. 1928. The fall armyworm. U.S. Dep. Ag-
ric. Tech. Bull. No. 34.
MATTSON, J. L., J. P. J. MAURISSEN, R. J. NOLAN, AND D.
K. A. BRZAK. 2000. Lack of differential sensitivity to
cholinesterase inhibition in fetuses and neonates
compared to dams treated perinatally with chlorpy-
rifos. Toxicological Sciences 53: 438-446.
MEAGHER, R. L. JR., AND M. GALLO-MEAGHER. 2003.
Identifying host strains of fall armyworm (Lepi-
doptera: Noctuidae) in Florida using mitochondrial
markers. Florida Entomol. 86: 450-455.
MIRCHEV, P., T. GEORGIEV, AND G. TSANKOV. 2001.
Studies on the parasitoids of Gelechia senticetella
(Stgr)(Lepidoptera: Gelechiidae) in Bulgaria. An-
zeiger fur Schadlingskunde 74: 94-96.
MOLINA-OCHOA, J. J. J. HAMM, R. LEZAMA-GUTIERREZ,
M. LOPEZ-EDWARDS, M. GONZALEZ-RAMiREZ, AND A.
PESCADOR-RUBIO. 2001. A survey of fall armyworm
(Lepidoptera: Noctuidae) parasitoids in the Mexican
States of Michoacan, Colima, Jalisco, and Tamauli-
pas. Florida Entomol. 84: 31-36.
MOLINA-OCHOA, J., R. LEZAMA-GUTIERREZ, M. RA-
MIREZ-GONZALEZ, M. LOPEZ-EDWARDS, M. A. RODRi-
GUEZ-VEGA, AND F. ARCEO-PALACIOS. 2003a. Patho-
gens and parasitic nematodes associated with popu-
lations of fall armyworm (Lepidoptera: Noctuidae)
larvae in M6xico. Florida Entomol. 86(3): 244-253.
MOLINA-OCHOA, J., J. E. CARPENTER, E. A. HEINRICHS,
AND J. E. FOSTER. 2003b. Parasitoids and parasites
of Spodoptera frugiperda (Lepidoptera: Noctuidae)
in the Americas and Caribbean Basin: an inventory.
Florida Entomol. 86: 254-289.
MONTOYA-BURGOS, J. A. 1980. Observaciones sobre la
importancia de Euplectrus sp., en la integraci6n de
m6todos de control de plagas de maiz. VIII Reuni6n
Nacional de Control Biol6gico. Secretaria de Agricul-
tura y Recursos Hidrdulicos, Sanidad vegetal, 22-25
de abril de 1980. Manzanillo, Colima, M6xico, 7 pp.
NAGOSHI, R. N., AND R. L. MEAGHER. 2003. FR Tan-
dem-reapet sequence in fall armyworm (Lepi-
doptera: Noctuidae) host strains. Ann. Entomol. Soc.
Am. 96: 329-335.
PAIR, S. D., J. R. RAULSTON, A. N. SPARKS, AND P. B.
MARTIN. 1986. Fall armyworm (Lepidoptera: Noctu-
idae) parasitoids: differential spring distribution
and incidence on corn and sorghum in the Southern
United States and Northeastern Mexico. Environ.
Entomol. 15: 342-348.
PASHLEY, D. P. 1988. Current status of fall armyworm
host strains. Florida Entomol. 71: 227-234.
PASHLEY, D. P., T. C. SPARKS, S. S. QUISENBERRY, T.
JAMJANYA, AND P. F. DOWD. 1987. Two fall army-
worm strains feed on corn, rice and bermudagrass.
Louisiana Ag. 30: 8-9.
PERALES, G. A., H. C. ARRENDONDO B., AND E. GARZA G.
1996. Native parasitoids of citrus leafminer Phylloc-
nistis citrella in Colima, Mexico. Southwestern En-
tomol. 21: 349-350.
REIS, J., L. OLIVERA, AND P. GARCIA. 2003. Effects of the
larval diet of Pseudaletia unipuncta (Lepidoptera:
Noctuidae) on the performance of the parasitoid
6. ,.-j.l..-- .. militaris (Hymenoptera: Braconi-
dae). Environ. Entomol. 32: 180-186.
RIGGIN, T. M., B. R. WISEMAN, D. J. ISENHOUR, AND K.
E. ESPELLE. 1992. Incidence of fall armyworm (Lep-
idoptera: Noctuidae) parasitoids on resistant and
susceptible corn genotypes. Environ. Entomol. 21:
888-895.
RIGGIN, T. M., K. E. ESPELIE, B. R. WISEMAN, AND D. J.
ISENHOUR. 1993. Distribution of fall armyworm
(Lepidoptera: Noctuidae) parasitoids on five corn
genotypes in South Georgia. Florida Entomol. 76:
292-302.
ROHLFS, W. M., AND T. P. MACK. 1985. Seasonal para-
sitism rates, host size, and adult emergence pattern
of parasitoids of the fall armyworm, Spodoptera fru-
giperda (J. E. Smith), with emphasis on Ophion fla-
vidus Brulle (Hymenoptera: Ichneumonidae). Ann.
Entomol. Soc. Am. 78: 217-220.
RUBERSON, J. R., G. A. HERZOG, AND W. J. LEWIS. 1993.
Parasitism of the beet armyworm, Spodoptera ex-
igua, in South Georgia cotton. Proc. 1993. Beltwide
Cotton Prod. Conf. 3: 993-997.
RUBERSON, J. R., G. A. HERZOG, W. R. LAMBERT, AND W.
J. LEWIS. 1994. Management of the beet armyworm
(Lepidoptera: Noctuidae) in cotton: role of natural
enemies. Florida Entomol. 77: 440-453.
RUIZ-CANCINO, E. 1991. Habitos parasiticos y alimenti-
cios de las Familias de himen6pteros de La Reserva
de la Biosfera "El Cielo" de Tamaulipas. 8 pp. http://
ecologia.uat.mx/biotam/v2n3/artl.html
SOLOMON, G. M., AND T. SCHETTLER. 2000. Environment
and health: 6. Endocrine disruption and potential hu-
man health implications. Canadian Medical Associa-
tion or its Licensors (CMAL) 163: 1471-1476.
472 Florida Ento
STEFFEY, K. 2001. Parasitoids of armyworm have been
common. http//www.ag.uiuc.edu/cespubs/pest/articles/
200114d.html
VICKERY, R. A. 1929. Studies of the fall armyworm in
the Gulf Coast district of Texas. U.S. Dep. Agric.
Tech. Bull. No. 138.
WILSON, J. W. 1933. The biology of parasites and pred-
ators of Laphygma exigua Huebner reared during
the season of 1932. Florida Entomol. 17: 1-15.
m
ologist 87(4) December 2004
YU, S. J. 1991. Insecticide resistance in the fall army-
worm, Spodoptera frugiperda (J. E. Smith). Pestic.
Biochem. Physiol. 39: 84-91.
Yu, S. J. 1992. Detection and biochemical characteriza-
tion of insecticide resistance in fall armyworm (Lep-
idoptera: Noctuidae). J. Econ. Entomol. 85: 675-682.
Austin et al.: Reticulitermes Genetics
MITOCHONDRIAL DNA VARIATION AND DISTRIBUTION
OF THE SUBTERRANEAN TERMITE GENUS RETICULITERMES
(ISOPTERA: RHINOTERMITIDAE) IN ARKANSAS AND LOUISIANA
JAMES W. AUSTIN', ALLEN L. SZALANSKI1 AND MATTHEW T. MESSENGER2
'Department of Entomology, University of Arkansas, Fayetteville, AR 72701
2Dow AgroSciences LLC, 9330 Zionsville Road, Indianapolis, IN 46268
ABSTRACT
Limited information exists on genetic variation and distribution of Reticulitermes from the
south central United States. Focusing on molecular sequence data from the mitochondrial
DNA 16S gene, this study records the distribution and genetic variation of Reticulitermes
species in Arkansas and updates the current distribution in a neighboring State, Louisiana.
Termite samples were collected from the field, subjected to DNA analysis with Polymerase
Chain-Reaction (PCR), and sequenced. Reticulitermes sp. sequence data were aligned, ge-
netic distances recorded, and their respective haplotypes were evaluated for possible geo-
graphic structure. From 35 Arkansas counties, 59 R. flavipes, 13 R. hageni, and seven R.
virginicus were identified. In Arkansas, 11 mitochondrial haplotypes were observed in R. fla-
vipes, three in R. hageni and three in R. virginicus. Among the 12 Louisiana parishes sam-
pled, 13 R. flavipes, three R. virginicus, and one R. tibialis were identified with six, three, and
one haplotypes for each species, respectively. Genetic variation among the R. flavipes haplo-
types from both States ranged from 0.2 to 0.9%. Reticulitermes flavipes haplotype diversity
observed in Arkansas and Louisiana was lower than observed in Texas and Oklahoma.
Key Words: 16S rRNA gene, DNA sequence, genetic variation, population genetics, Reticuli-
termes, termite.
RESUME
La informaci6n existente sobre la variaci6n gen6tica y la distribuci6n de Reticulitermes en el
sur central del los Estados Unidos es limitada. Enfocandose en los datos de las secuencias
moleculares del gene 16S del ADN mitocondrial, este studio registra la distribuci6n y la va-
riaci6n gen6tica de Reticulitermes spp. en Arkansas y pone al dia la distribuci6n actual en el
estado vecino, Louisiana. Las muestras de termitas recolectadas del campo, fueron sujetas
al andlisis de ADN por la reacci6n en cadena de la polimerasa (RCP), y secuenciadas. Los da-
tos de secuenciaci6n gen6tica para Reticulitermes spp. fueron alineados, las distancias gen6-
ticas registradas, y sus haplotipos respectivos fueron evaluados para su possible estructura
geografica. De los 35 condados de Arkansas, 59 R. flavipes, 13 R. hageni, y siete R. virginicus
fueron identificados. En Arkansas, 11 haplotipos de mitocondria fueron observados en R. fla-
vipes, tres en R. hageni y tres en R. virginicus. Entre las 12 regions de Louisiana muestrea-
das, 13 R. flavipes, tres R. virginicus, y una R. tibialis fueron identificados con seis, tres, y
un haplotipos por cada especie, respectivamente. La variaci6n gen6tica entire los haplotipos
de R. flavipes de ambos estados fue de 0.2 hasta 0.9%. La diversidad de los haplotipos de
Reticulitermes flavipes observada en Arkansas y Louisiana fue menor de la que fue obser-
vada en Texas y Oklahoma.
In Arkansas and Louisiana, subterranean ter-
mites cause millions of dollars of damage annually.
Damage caused by subterranean termite activity
probably exceeds $2.5 billion annually in the
United States (Anonymous 2003) and $22 billion
globally (Su 2002). Several structurally important
species in the genus Reticulitermes are common
throughout the southeastern U.S. (Weesner 1965).
Messenger et al. (2002) completed a comprehen-
sive survey of termites in Louisiana, providing
information for the likely occurrences of Reticuli-
termes spp. in Arkansas. Previously, Snyder (1954)
listed four species in Arkansas: Reticulitermes fla-
vipes (Kollar), R. virginicus Banks, R. hageni
Banks, and R. tibialis Banks. All four species have
been documented in Louisiana, including R. tibia-
lis, which was recently discovered (Messenger
2003). Information on the distribution of Reticuli-
termes spp. in Arkansas has become available
through a current national termite survey, which
confirms that Reticulitermes spp. are common
throughout the State (Messenger 2003). Moreover,
surveys from the neighboring States of Texas and
Oklahoma have included valuable information on
Reticulitermes distribution and genetic composi-
tion (Austin et al. 2004a, b).
Previous genetic studies have primarily fo-
cused on phylogenetic relationships among Reti-
Florida Entomologist 87(4)
culitermes species from the eastern United States
and Western Europe (Jenkins et al. 1998, 2001;
Marini & Mantovani 2002; Uva et al. 2003; Ye et
al. 2003). Recently, Austin et al. (2004a, b) have
conducted the first comprehensive genetic sur-
veys of Reticulitermes sp. for Texas and Oklahoma
using DNA sequencing of a portion of the mito-
chondrial DNA (mtDNA) 16S rRNA gene. We in-
vestigated the extent of genetic variation within
and among Arkansas and Louisiana Reticuliter-
mes relative to Texas and Oklahoma (Austin et al.
2004ab), evaluated these genetic markers for iden-
tifying species, and updated the geographical dis-
tribution of these taxa.
MATERIALS AND METHODS
Termites were collected in Arkansas and Loui-
siana and preserved in 100% ethanol (Table 1).
We solicited the assistance of Pest Management
Professionals (PMPs) throughout both States to
determine which species are most frequently re-
covered from infested structures. We provided col-
lection kits, and PMPs returned samples to our
laboratory for analysis. From various geographic
zones throughout Arkansas and Louisiana (Table
1), 96 samples were used for molecular analysis.
When alates or soldiers were available, Reticu-
litermes sp. were morphologically identified to
species by applying the keys of Krishna &
Weesner (1969); Scheffrahn & Su (1994); Hostet-
tler et al. (1995); and Donovan et al. (2000). In ad-
dition, all samples were identified to species with
mtDNA 16S sequences (Szalanski et al. 2003). All
of the morphological species identifications
agreed with the molecular species identifications.
Three additional taxa (Table 1) were included as
outgroup taxa to corroborate relationships within
the genus for phylogenetic analysis. Voucher spec-
imens preserved in 100% ethanol are maintained
at the Arthropod Museum, Department of Ento-
mology, University of Arkansas, Fayetteville, AR.
DNA was extracted from alcohol-preserved
specimens dried on filter paper according to Liu &
Beckenbach (1992) and Jenkins et al. (1999) from
individual worker termites with the Puregene
DNA isolation kit D-5000A (Gentra, Minneapolis,
MN). Extracted DNA was resuspended in 50 pl of
Tris:EDTA and stored at -20C. Polymerase chain
reaction (PCR) was conducted with the primers
LR-J-13007 (5'-TTACGCTGTTATCCCTAA-3')
(Kambhampati & Smith 1995) and LR-N-13398
(5'-CGCCTGTTTATCAAAAACAT-3') (Simon et
al., 1994). These PCR primers amplify an approx-
imately 428-bp region of the mtDNA 16S rRNA
gene. The PCR reactions were conducted with 1 pl
of the extracted DNA (Szalanski et al. 2000), hav-
ing a profile consisting of 35 cycles of 94C for 45
s, 46C for 45 s and 72C for 60 s. Amplified DNA
from individual termites was purified and concen-
trated with minicolumns according to the manu-
facturer's instructions (Wizard PCRpreps,
Promega). Samples were sent to The University of
Arkansas Medical School DNA Sequencing Facil-
ity (Little Rock, AR) for direct sequencing in both
directions. GenBank accession numbers were
AY603499 to AY603509 for termite DNA se-
quence haplotypes new to this study. Consensus
sequences for each sample were obtained with
Bioedit 5.09 (Hall 1999), and sequences were
aligned by CLUSTAL W (Thompson et al. 1994).
Mitochondrial DNA haplotypes were aligned by
MacClade v4 (Sinauer Associates, Sunderland,
MA). Haplotype distribution between popula-
tions, number of haplotypes, number of unique
haplotypes, haplotype diversity (h), and nucle-
otide diversity (pi) were calculated with DNAsp
v3.51 (Rozas & Rozas 1999).
The distance matrix option of PAUP* 4.0b10
(Swofford 2001) was used to calculate genetic dis-
tances according to the Kimura 2-parameter
model of sequence evolution (Kimura 1980). Mito-
chondrial 16S sequences from the desert subter-
ranean termite Heterotermes aureus (Snyder)
(GenBank AY380299) and Formosan subterra-
nean termite Coptotermes formosanus Shiraki
(GenBank AY558910) were added to the Reticuli-
termes DNA sequences as outgroup taxa. Maxi-
mum parsimony analysis on the alignments were
conducted with PAUP* 4.0b10 (Swofford 2001).
Gaps were treated as missing data. The reliability
of trees was tested with a bootstrap test (Felsen-
stein 1985). Parsimony bootstrap analysis in-
cluded 1,000 resamplings and used the Branch
and Bound algorithm of PAUP*. Because no pre-
vious accounts of the abundance of Reticulitermes
in Arkansas have been published, we compiled all
available data from existing sources and noted
them on our distribution map (Fig. 1).
RESULTS
The DNA sequencing of the 16S rDNA ampli-
con revealed an average size of 428 bp. The aver-
age base frequencies were A = 0.41, C = 0.23, G =
0.13, and T = 0.23. From 35 Arkansas counties 59
R. flavipes, 11 R. hageni, and seven R. virginicus
were identified based on species diagnostic nucle-
otide sites from Szalanski et al. (2003) (Table 1).
In Arkansas, 11 haplotypes were observed in R.
flavipes, three in R. hageni and three in R. virgini-
cus (Table 1, Fig. 1). Among the 12 Louisiana par-
ishes sampled, 13 R. flavipes, three R. virginicus,
and one R. tibalis were identified with six, three,
and one haplotypes for each species, respectively
(Fig. 2, Table 1).
Nine nucleotide sites were variable among the
11 R. flavipes haplotypes (Table 2), and Tajima-
Nei distances (Tajima & Nei 1984) among the
R. flavipes haplotypes ranged from 0.2 to 0.9%
(Table 3). The most common haplotypes were F
and G with 32 and 9 representatives, respectively.
December 2004
Austin et al.: Reticulitermes Genetics
TABLE 1. COLLECTION DATA, AND HAPLOTYPES FOR ARKANSAS AND LOUISIANA Reticulitermes AND OUTGROUP TAXA.
Species City County/Parish State Haplotype n
Amity
Brinkley
Cave City
Clifty
Cushman
Strickler
Fayetteville
El Dorado
Eureka Springs
Ft. Smith
Harrison
Jasper
Marvell
McGehee
Mineral Springs
Nashville
Paragould
Stuttgart
Warm Springs
Hardin
Hoxie
Walnut Ridge
Newport
Hot Springs
Baton Rouge
Houma
Cut Off
Delhi
Port Sulphur
West Monroe
Ashdown
Ft. Smith
Warren
Fayetteville
Pine Bluff
Shreveport
Sulphur
Camp Robinson
Glenwood
Jonesboro
N. Little Rock
Newport
Piggot
Morrilton
Shreveport
Jonesboro
Fayetteville
Harrison
Hoxie
Newport
Walnut Ridge
Fayetteville
Hoxie
Jonesboro
Lake Charles
Lake Dardanelle
R. flavipes
Clark
Monroe
Sharp
Madison
Polk
Independence
Washington
Washington
Union
Carroll
Sebastian
Boone
Newton
Phillips
Desha
Howard
Howard
Greene
Arkansas
Randolph
Jefferson
Lawrence
Lawrence
Jackson
Garland
E. Baton Rouge
Terrebonne
Lafourche
Richland
Plaquemines
Ouachita
Little River
Sebastian
Bradley
Washington
Jefferson
Caddo
Calcasieu
Pulaski
Pike
Craighead
Pulaski
Jackson
Clay
Conway
Caddo
Jackson
Washington
Boone
Lawrence
Jackson
Lawrence
Washington
Lawrence
Craighead
Calcasieu
Pope
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
AR F
LA F
LA F
LA F
LA F
LA F
LA J
AR M
AR M
AR M
AR M
AR M
LA M
LA M
AR G
AR G
AR G
AR G
AR G
AR G
AR G
LA G
LA G
AR Q
AR Q
AR Q
AR Q
AR Q
AR T
AR W
AR P
LA P
AR R
Florida Entomologist 87(4)
TABLE 1. (CONTINUED) COLLECTION DATA, AND HAPLOTYPES FOR ARKANSAS AND LOUISIANA Reticulitermes AND OUT-
Species City County/Parish State Haplotype n
R. hageni
R. uirginicus
R. tibialis
Coptotermes formosanus
Heterotermes aureus
Jonesboro
Strickler
Fayetteville
Searcy
Chauvin
Little Rock
Pocahontas
Sheridan
Sherwood
Ft. Smith
Eureka Springs
Fayetteville
Weddington
Pocahontas
Conway
Clifty
Fayetteville
Marmaduke
Strickler
Hamburg
Fayetteville
Fayetteville
Fayetteville
Morgan Mtn.
Minden
Raceland
Delcambre
Sulphur
Rockwall
Craighead
Washington
Washington
Madison
White
Faulkner
Terrebonne
Pulaski
Randolph
Grant
Pulaski
Sebastian
Carroll
Washington
Washington
Randolph
Faulkner
Polk
Madison
Washington
Greene
Washington
Ashley
Washington
Washington
Washington
Franklin
Webster
Lafourche
Vermilion
Calcasieu
Lamar
AR R
AR R
AR R
AR R
AR R
AR S
LA S
AR V
AR V
AR V
AR V
AR U
AR H1
AR H1
AR H1
AR H1
AR H1
AR H1
AR H1
AR H2
AR H2
AR H3
AR H3
AR V1
AR V2
AR V3
AR V1
LA V1
LA V4
LA V5
LA T8
TX outgroup
AZ outgroup
Within R. hageni, one nucleotide site was variable
between the two observed haplotypes. Haplotype
diversity for R. flavipes from Arkansas was 0.759,
and 0.782 for Louisiana (Table 4). Both States
had high levels of genetic diversity. Tajima's test
resulted in non-significant P values (P < 0.05) in
both States leading to the acceptance of the null-
hypothesis of neutrality for the mtDNA rRNA
16S gene (Table 4).
Bootstrap analysis of the aligned Reticuliter-
mes species and the outgroup taxa resulted in a
consensus tree with several distinct branches
(Fig. 3). These distinct clades included R. flavipes,
R. tibialis, R. hageni, and R. virginicus. No ge-
netic relationship was observed amongR. flavipes
haplotypes.
DISCUSSION
This study represents the first attempt to up-
date the current geographic distribution of Reti-
culitermes spp. and genetically categorize the
genus Reticulitermes in Arkansas. At the same
time, this is the first molecular description of
Reticulitermes spp. from Louisiana, albeit on a
limited scale.
Populations of nearly all species, social or oth-
erwise, exhibit at least some degree of genetic dif-
ferentiation among geographic locales (Ehrlich &
Raven 1969). One of the purposes of the research
presented herein was to estimate the baseline ge-
netic variation which occurs both within and
among Reticulitermes spp. in Arkansas and Loui-
siana. By combining haplotype observations in
the present study with the studies of Austin et al.
(2004a) from Texas and Austin et al. (2004b) from
Oklahoma, we had hoped to observe some type of
spatial continuity which may not have been re-
vealed otherwise. As with animal populations, ad-
ditional genetic structure normally is to be
expected over increasing spatial scales, where
populations can show additional differentiation
due to spatial habitat structure, isolation by dis-
tance, or other factors (Avise 1994). There was no
December 2004
Austin et al.: Reticulitermes Genetics
Fig. 1. Species distribution of Reticulitermes haplo-
types in Arkansas. Samples obtained from the National
Termite Survey, but not subjected to genetic analysis are
marked as "+"R. flavipes, "" R. virginicus, "" R. hageni.
Letters indicating haplotypes are given in Table 1.
apparent consistency of haplotype occurrence for
Reticulitermes spp. in this study based on geogra-
phy. However, we found genetic divergence values
similar to those detected in our previous works
(Austin et al. 2004a, b).
Of the four species of Reticulitermes presented
in this study, R. flavipes was the most common,
followed by R. hageni and R. virginicus. Haplo-
type F was the most common haplotype of R. fla-
vipes observed in Arkansas, and represented 26 of
the 79 (44%) samples from Arkansas. This haplo-
type is also present in both Texas and Oklahoma
but in smaller distributions: 12 and 13% for Okla-
homa and Texas, respectively. In Oklahoma and
Texas, the most abundant R. flavipes haplotypes
are L (23%) and G (28%), respectively (Austin et
al. 2004a, b).
A haplotype or allele is defined by one unique
form of the gene and differs from any other haplo-
Fig. 2. Species distribution of Reticulitermes haplo-
types in Louisiana. Letters indicating haplotypes are
given in Table 1.
type by at least one nucleotide. Haplotype diver-
sity or gene diversity quantifies the number of
haplotypes in relation to their relative frequency
to each other, and is the probability that two se-
quences randomly selected from a population are
different (Nei 1987). Haplotype diversity for R.
flavipes from Arkansas and Louisiana was lower
than Texas or Oklahoma (Austin et al. 2004a, b).
Tajima's D (Tajima 1989) is a standard test for
the neutrality of a gene region. This is a useful
measure because nucleotide diversity (Pi) calcula-
tions are dependant upon the infinite alleles
model that assumes gene neutrality. Values of D
indicate not only if natural selection is influencing
gene frequencies but the type of selection pressure
in operation. Negative values of D can indicate re-
TABLE 2. HAPLOTYPE VARIATION AT 9 NUCLEOTIDE SITES AMONG Reticulitermes flavipes FROM ARKANSAS AND LOUI-
SIANA.
Haplotype 55 97 122 131 162 168 179 270 271
F G A A A G G C T T
G T G A C
J G A
M A
P A C
Q T
R A
S A A
T A C C
U C
V A
Florida Entomologist 87(4)
TABLE 3. GENETIC DIVERGENCE AMONG RETICULITERMES FLAVIPES HAPLOTYPES (HAP) FROM ARKANSAS AND LOUISI-
ANA.
Hap G T P U V F Q R S M J
G
T 0.009 -
P 0.007 0.002 -
U 0.009 0.005 0.007 -
V 0.009 0.009 0.007 0.005 -
F 0.007 0.007 0.005 0.002 0.002 -
Q 0.009 0.009 0.007 0.005 0.005 0.002 -
R 0.007 0.007 0.005 0.007 0.007 0.005 0.007 -
S 0.007 0.009 0.007 0.009 0.005 0.007 0.009 0.002 -
M 0.005 0.005 0.002 0.005 0.002 0.005 0.002 0.005 -
J 0.002 0.007 0.005 0.007 0.007 0.005 0.007 0.005 0.007 0.002
cent expansions in population size while positive
values indicate recent population bottlenecks.
Tajima's D value was negative for both Arkansas
and Louisiana, indicating that R. flavipes in these
States may be expanding in population size. A
similar result was observed from R. flavipes from
Oklahoma, while Texas R. flavipes have a positive
D value, indicating a recent population bottle-
neck. The null hypothesis of this test is that the
gene region of interest is neutral. If the D value is
significant (P < 0.05) then the null hypothesis of
neutrality may be rejected. Tajima's test resulted
in non-significant p values (P > 0.05) for Arkansas
and Louisiana. This was also observed in Texas
and Oklahoma (Austin et al. 2004a, b) leading to
the acceptance of the null-hypothesis of neutrality
for the mtDNA rRNA 16S gene.
The phylogenetic relationships of Reticuliter-
mes evaluated in this study are consistent with
those from neighboring states (Austin et al.
2004a, b) and in other areas (Austin et al. 2002;
Jenkins 1998, 1999, 2001). Extensive collecting in
Louisiana has produced only one sample ofR. tib-
ialis from Sulphur, Calcasieu Parish, which bor-
ders Texas. Although Snyder (1954) lists R.
tibialis as occurring in Arkansas, we have not yet
recovered this species to date.
We speculate that because R. tibialis was gen-
erally not recovered in our collecting efforts, an
eastern transition zone (from east to west) for the
distribution of this species may exist in the prox-
imity of Calcasieu Parish, Louisiana. This species
is known to prefer more arid climates which can
be more readily found in Texas and Oklahoma
where R. tibialis has been more frequently recov-
ered. More intensive collecting efforts should be
performed to validate this hypothesis.
The lack of a geographical haplotype continu-
ity of Reticulitermes suggests that (1) we require
samples from larger geographic zones, (2) we
need to increase the number of samples se-
quenced, or (3) the observed lack of spatial conti-
nuity from Reticulitermes may be attributed to
anthropogenic origins. To evaluate the latter ele-
ment, evaluation of haplotype frequency from un-
disturbed habitats (e.g., protected forests) should
be compared with our current data, which largely
reflects samples obtained from urban landscapes.
Selected sites in undisturbed locations are being
evaluated for intensive collecting efforts in future
studies where more comprehensive statistical
measures can be applied, and a better overall un-
derstanding of population dynamics may be ad-
dressed than in the current study.
TABLE 4. GENETIC DIVERSITY PARAMETERS FOR EACH POPULATION OF RETICULITERMES FLAVIPES FROM TEXAS, OKLA-
HOMA, ARKANSAS, AND LOUISIANA.
No. No. unique Haplotype Nucleotide Tajima's D value
Population n haplotypes haplotypes diversity (h) diversity (Pi) (P-value)
TXb 69 13 3 0.864 0.0071 +0.6021 ns
OKR 41 10 1 0.811 0.0048 -0.6349 ns
AR 59 11 3 0.759 0.0044 -0.3488 ns
LA 13 6 0 0.782 0.0044 -0.0688 ns
aA non-significant value for this test indicates that the null hypothesis of neutrality in the 16S rRNA gene cannot be rejected.
Non-significance of test when P > 0.10.
'Data from Austin et al. (2004a, b).
December 2004
Austin et al.: Reticulitermes Genetics
Rb~vpeshwG
RffavpnIhapT
R&Vevoeha9 P
Rbwesahp U
Rf vwqsI, IV
RVAvqeslhap F
RbvtpAsh W a
RbAvpnhq R
R flav*osI S
Rb-eA-hM
R vqs-api J
kiff.kshw rTh
R hsg.m hp H2
0- R hfgehagWp H3
R hoei hap H1
Rvu-ug hap V1
R m*c t ap V2
--- Rwwhapy2
Fig. 3. Single most parsimonious tree based on the
16S rRNA gene during a branch and bound search with
PAUP*. Bootstrap values for 1,000 replicates are listed
above the branches supported at >50%.
ACKNOWLEDGEMENT
We thank all who contributed specimens to this
study, particularly the Pest Management Professionals
(PMPs) of both Arkansas and Louisiana. Special thanks
to Jerry Hyde of the Arkansas Pest Management Associ-
ation, Martyn Hafley, Mike Merchant, Gregg Hender-
son, and Paul Baker for contribution of samples. We
thank Martha Davis for critical review of this manu-
script. Research was supported in part by the University
of Arkansas, Arkansas Agricultural Experiment Station.
REFERENCES CITED
AVISE, J. C. 1994. Molecular Markers, Natural History
and Evolution. Chapman & Hall, New York, NY. 511
pp.
ANONYMOUS. 2003. NPMA webpage http://65.217.229.171/
media/art_protectyour_home.asp
AUSTIN, J. W., A. L. SZALANSKI, R. E. GOLD, AND B. T.
FOSTER 2004a. Genetic variation and geographical
distribution of the subterranean termite genus Reticu-
litermes in Texas. Southwestern Entomol. 29: 1-11.
AUSTIN, J. W., A. L. SZALANSKI, AND B. M. KARD. 2004b
Genetic variation and distribution of the subterranean
termite Genus Reticulitermes (Isoptera: Rhinotermiti-
dae) in Oklahoma. Florida Entomol. 87: 152-158.
AUSTIN, J. W., A. L. SZALANSKI, P. UVA, A. BAGNERES
AND A. KENCE. 2002. A comparative genetic analysis
of the subterranean termite genus Reticulitermes
(Isoptera: Rhinotermitidae). Ann. Entomol. Soc.
Amer. 95: 753-760.
DONOVAN, S. E., D. T. JONES, W. A. SANDS, AND P. EGG-
ELTON. 2000. Morphological phylogenetics of ter-
mites (Isoptera). Biol. J. Linnean Soc. 70: 467-513.
EHRLICH, P. R., AND P. H. RAVEN. 1969. Differentiation
of populations. Science 165: 1228-1232.
FELSENSTEIN, J. 1985. Confidence limits on phyloge-
nies: An approach using the bootstrap. Evolution 39:
783-791.
HALL, T. A. 1999. BioEdit: a user-friendly biological se-
quence alignment editor and analysis program for
Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98.
HOSTETTLER, N. C., D. W. HALL, AND R. H. SCHEF-
FRAHN. 1995. Intracolony morphometric variation
and labral shape in Florida Reticulitermes (Isoptera:
Rhinotermitidae) soldiers: Significance for identifi-
cation. Florida Entomol. 78: 119-129.
JENKINS, T. M., C. J. BASTEN, R. DEAN, S. E. MITCHELL,
S. KRESOVICH, AND B. T. FORSCHLER. 1998. Matriar-
chal genetic structure of Reticulitermes (Isoptera:
Rhinotermitidae) populations. Sociobiol. 33: 239-263.
JENKINS, T. M., C. J. BASTEN, S. KRESOVICH, AND B. T.
FORSCHLER 1999. Mitochondrial gene sequence
questions Reticulitermes sp. social structure
(Isoptera: Rhinotermitidae). Sociobiol. 34: 161-172.
JENKINS, T. M., R. E. DEAN, R. VERKERK, AND B. T. FOR-
SCHLER.. 2001. Phylogenetic analysis of two mito-
chondrial genes and one nuclear intron region
illuminate European subterranean termite (Isoptera:
Rhinotermitidae) gene flow, taxonomy, and introduc-
tion dynamics. Molec. Phylog. Evol. 20: 286-293.
KAMBHAMPATI, S., AND P. T. SMITH. 1995. PCR primers
for the amplification of four insect mitochondrial
gene fragments. Ins. Molec. Biol. 4: 233-236.
KIMURA, M. 1980. A simple method for estimating evo-
lutionary rate of base substitutions through compar-
ative study of nucleotide sequences. J. Molec. Evol.
16: 111-120.
KRISHNA, K., AND F. M. WEESNER 1969. Biology of Ter-
mites. Vol. 1. Academic Press, New York, NY.
LIU, H., AND A. T. BECKENBACH. 1992. Evolution of the
mitochondrial chyochrome oxidase II gene among 10
orders of insects. Mol. Phylogenet. Evol. 41: 31-52.
MARINI, M., AND B. MANTOVANI. 2002. Molecular Rela-
tionships among European Samples of Reticuliter-
mes (Isoptera, Rhinotermitidae). Mol. Phylogenet.
Evol. 22: 454-459.
MESSENGER, M. T., SU, NAN-YAO, AND R. H. SCHEF-
FRAHN. 2002. Current distribution of the Formosan
subterranean termite and other termite species
(Isoptera: Rhinotermitidae, Kalotermitidae) in Loui-
siana. Florida Entomol. 85: 580-587.
MESSENGER, M. T. 2003. National termite survey. New
Orleans Mosquito & Termite Control Board webpage.
http://www.termitesurvey.com.
NEI, M. 1987. Molecular Evolutionary Genetics. Colum-
bia University Press, New York.
ROZAS, J., AND R. ROZAS. 1999. DNAsp version 3: an in-
tegrated program for molecular population genetics
and molecular evolution analysis. Bioinfomatics 15:
174-175.
SCHEFFRAHN, R. H., AND N.-Y. SU. 1994. Keys to soldier
and winged adult termites (Isoptera) of Florida.
Florida Entomol. 77: 460-474.
SIMON, C., F. FRATI, A. BECKENBACH, B. CRESPI, H. LIU,
AND P. FLOOK. 1994. Evolution, weighting, and phy-
logenetic utility of mitochondrial gene sequences and
a compilation of conserved polymerase chain reaction
primers. Ann. Entomol. Soc. Amer. 87: 651-701.
SNYDER, T. E. 1954. Order Isoptera. The Termites of the
United States and Canada. National Pest Control
Assn., New York. 64 pp.
SU, N.-Y. 2002. Novel technologies for subterranean ter-
mite control. Sociobiol. 40: 95-101.
Swofford, D. L. 2001. PAUP*: Phylogenetic analysis us-
ing parsimony (*and other methods), ver. 4.0b10.
Sinauer, Sunderland, Massachusetts.
SZALANSKI, A. L., D. S. SIKES, R. BISCHOF, AND M.
FRITZ. 2000. Population genetics and phylogenetics
of the endangered American burying beetle, Nicro-
phorus americanus (Coleoptera: Silphidae). Ann.
Entomol. Soc. Amer. 93: 589-594.
SZALANSKI, A. L., J. W. AUSTIN, AND C. B. OWENS. 2003.
Identification of Reticulitermes spp. (Isoptera: Rhino-
termitidae) from South Central United States by
PCR-RFLP. J. Econ. Entomol. 96: 1514-1519.
TAJIMA, F. 1989. Statistical method for testing the neu-
tral mutation hypothesis by DNA polymorphism.
Genetics. 123: 585-595.
December 2004
TAJIMA, F., AND M. NEI. 1984. Estimation of evolution-
ary distance between nucleotide sequences. Mol.
Biol. Evol. 1: 269-285.
THOMPSON, J. D., D. G. HIGGINS, AND T. J. GIBSON.
1994. CLUSTAL W: improving the sensitivity of pro-
gressive multiples sequence alignments through se-
quence weighting, position-specific gap penalties
and weight matrix choice. Nucleic Acids Res. 22:
4673-4680.
UVA, P., J.-L. CLEMENT, J. W. AUSTIN, V. ZAFFAGNINI,
A. QUITANA, AND A. G. BAGNERES. 2003. Origin of a
new Reticulitermes termite (Isoptera, Rhinotermiti-
dae) inferred from mitochondrial DNA data. Mol.
Phylogen. Evol. 30: 344-353.
WEESNER, F. M. 1965. Termites of the United States, A
Handbook. Nat. Pest Contr. Assn., Elizabeth, NJ. 70 pp.
YE, W., C-Y LEE, R. H. SCHEFFRAHN, J. M. ALEONG, N-
Y SU, G. W. BENNETT, AND M. E. SCHARF. 2004. Phy-
logenetic relationships of nearctic Reticulitermes
species (Isoptera: Rhinotermitidae) with particular
reference to Reticulitermes arenincola Goellner.
Molec. Phylogenetics Evolution 30: 815-822.
Florida Entomologist 87(4)
Shelly et al.: Trapping Male Fruit Flies
SYNERGISTIC AND INHIBITORY INTERACTIONS BETWEEN
METHYL EUGENOL AND CUE LURE INFLUENCE TRAP CATCH
OF MALE FRUIT FLIES, BACTROCERA DORSALIS (HENDEL)
AND B. CUCURBITAE (DIPTERA: TEPHRITIDAE)
TODD E. SHELLY, ELAINE PAHIO AND JAMES EDU
USDA-APHIS, 41-650 Ahiki Street, Waimanalo, HI 96795
ABSTRACT
Males of the oriental fruit fly, Bactrocera dorsalis (Hendel) and the melon fly, B. cucurbitae
(Coquillett), are attracted to methyl eugenol (ME) and cue lure (CL), respectively. These
lures, when mixed with a toxicant, are widely used to detect and suppress populations of
these agricultural pests.The objective of this study was to assess the effectiveness of(1) traps
baited with both ME and CL (mixed or presented separately on adjacent wicks), and (2) traps
baited with a single lure but placed in the same tree as a trap containing the alternate lure
(1 or 3 m apart). Jackson traps were placed in a mixed orchard on Oahu, Hawaii, and the
numbers of released (marked) and wild males were recorded. Traps baited with ME and CL
(mixed or separate) captured significantly fewer B. dorsalis males than traps baited with ME
alone. CL placed 1 m from ME-baited traps in the same tree also reduced the number of B.
dorsalis males captured. Conversely, ME appeared to increase capture of B. cucurbitae
males, and traps baited with the 2 lures (mixed but not separate) captured significantly more
released males than traps baited with CL alone. Also, ME placed 1 m (but not 3 m) from CL-
baited traps increased the trap catch of released B. cucurbitae males. Results are discussed
from the perspectives of management and evolution of Bactrocera species.
Key Words: Bactrocera spp., oriental fruit fly, melon fly, methyl eugenol, cue-lure, Diptera.
RESUME
Los machos de la mosca oriental de la fruta, Bactrocera dorsalis (Hendel) y de la mosca del
mel6n, B. cucurbitae (Coquillett), son atraidos al eugenol metil (EM) y al atrayente "cue lure"
(CL), respectivemente. Estos atrayentes, cuando estan mezclados con una toxina, son usados
en muchas lugares para detectar y suprimir las poblaciones de estas plagas agricolas. El ob-
jectivo de este studio fue para avaluar la eficacia de 1) trampas de cebo con ambos EM y CL
(mezcladas o presentadas separadamente en mechas adyacentes) y 2) trampas de cebos con
un solo atrayente pero puestos en el mismo arb61 de una trampa que tiene un atrayente al-
ternativo (separadas por 1 o 3 metros). Las trampas "Jackson" fueron puestas en un huerto
mezclado en Oahu, Hawaii, y el numero de los machos liberados (marcados) y los machos na-
turales fueron registrados. Las trampas con EM y CL (mezclados o separados) capturaron
significativamente menos machos de B. dorsalis que las trampas con solo EM. Tambien, el
CL puesto 1 m de las trampas con EM en el mismo arb61 reduj6 el numero de los machos de
B. dorsalis capturados. Al contrario, el EM aparece que se aumenta el numero capturado de
los machos de B. cucurbitae, y las trampas con los dos atrayentes (mezclados pero no sepa-
rados) capturaron significativamente mas machos liberados que las trampas con solo CL.
Tambi6n, el EM puesto 1 m (pero no 3 m) de las trampas cebadas con el CL aument6 el nu-
mero capturado de machos de B. cucurbitae liberados. Se discuten los resultados desde el
punto de vista de manejo y evoluci6n de las species de Bactrocera.
The genus Bactrocera contains approximately
440 species distributed primarily in Southeast
Asia, the South Pacific, and Australia (White &
Elson-Harris 1992). Males of many Bactrocera
species are attracted to either methyl eugenol (4-
allyl-1,2-dimethoxybenzene) or cue lure [4-(4-ace-
toxyphenyl)-2-butanone] (Drew & Hooper 1981).
Methyl eugenol (ME hereafter) is a widely distrib-
uted plant natural product and occurs in over 200
plant species representing 32 families (Tan &
Nishida 1996). Cue lure (CL hereafter) has not
been isolated as a natural product but is rapidly
hydolyzed to form raspberry ketone (RK hereaf-
ter), which is found in a variety of plants (Metcalf
1990). In Hawaii, the site of the present study, two
major fruit fly pests are the Oriental fruit fly, B.
dorsalis (Hendel), and the melon fly, B. cucurbitae
(Coquillett), males of which respond to ME and
CL, respectively.
Control programs against Bactrocera species
use male lures to both detect and suppress pest
populations. For example, ME- and CL-baited
traps are monitored year-round in southern Cali-
fornia to detect the presence of incipient infesta-
tions ofB. dorsalis and B. cucurbitae, respectively
(Gilbert & Bingham 2002). In a well-known case
Florida Entomologist 87(4)
of male annihilation, Steiner et al. (1965) eradi-
cated B. dorsalis from Rota Island by distributing
thousands of fiber board blocks soaked with ME
and a toxicant. Although a less powerful attrac-
tant than ME, and hence less effective at male an-
nihilation, CL (plus a toxicant) has similarly been
used to suppress populations of B. cucurbitae
prior to the start of sterile release programs (e.g.,
Miyako Island, Kuba et al. 1996).
Several studies have investigated the effective-
ness of traps baited with a mixture of ME and CL
for possible use in detection and suppression pro-
grams in areas containing both ME- and CL-re-
sponding Bactrocera species. Combining lures in
such a manner could potentially reduce the num-
ber of traps needed in a given area and the associ-
ated manpower required to service them. In
addition, combining lures may reduce the amount
of pesticide used, thus lessening economic and en-
vironmental costs. Previous studies suggest an
asymmetry in the effectiveness of ME + CL mix-
tures for ME- versus CL-responding species. On
one hand, data uniformly show that CL, either
mixed with or placed immediately adjacent to ME,
reduces trap capture of ME-responding species
(Hooper 1978; Vargas et al. 2000). In contrast, data
regarding lure mixtures and the response of CL-re-
sponding species have been inconsistent. Data on
B. cucurbitae from Taiwan (cited by Hooper 1978)
showed that adding ME to CL nearly doubled the
number of males captured compared with traps
baited with CL alone. In contrast, Hooper (1978)
found that the mixture of ME + CL reduced trap
capture of CL-responding species in Queensland,
Australia. However, Hooper (1978) did report an
increase in male numbers for traps in which ME
and CL were applied to separate wicks in the same
trap. More recently, Vargas et al. (2000) found that
lure mixtures containing at least 25% CL by vol-
ume attracted similar numbers of B. cucurbitae
males as traps baited solely with CL.
The objective of the present study was to pro-
vide additional data on the responsiveness of B.
dorsalis and B. cucurbitae males to traps baited
only with ME or CL, respectively, relative to traps
containing the 2 lures mixed together or on sepa-
rate wicks. In addition, we monitored the effect of
ME on CL, and vice versa, on trap captures when
2 traps containing the respective lures were
placed in the canopy of the same tree.
MATERIALS AND METHODS
Study Site
Fieldwork was conducted during January No-
vember, 2003, at the University of Hawaii Agri-
cultural Experiment Station, Waimanalo, Oahu,
in a mixed fruit orchard that contained mango
(Mangifera indica L.), guava (Psidium guajava
L.), orange (Citrus sinensis (L.) Osbeck), lime
(C. aurantiifolia (Christm.) Swingle), and bread-
fruit (Artocarpus altilis (Parkins.) Fosb.) along
with other non-host trees. Daily maximum and
minimum temperatures ranged between 25-33C
and 19-25C, respectively, during this period.
Study Insects
Males of B. dorsalis and B. cucurbitae were
from laboratory colonies maintained by the
USDA-ARS Tropical Fruit, Vegetable, and Orna-
mental Crop Laboratory, Honolulu, since the mid-
1980s and mid-1960s, respectively (D. McInnis,
personal communication). We obtained B. cucurb-
itae as pupae and B. dorsalis as eggs, which were
placed on standard larval medium (Tanaka et al.
1969) in plastic containers over vermiculite for
pupation. In both species, males were collected
within 48 h of eclosion (males from laboratory col-
onies attain sexual maturity at 6-7 days of age,
Vargas et al. 1984), held in screen-covered plastic
buckets (volume 5 liters; 100-150 males per
bucket), and provided water and food (a sugar-
yeast hydrolysate mixture, 3:1 v/v). Flies were
maintained at 23-27C and 60-90% RH under a
natural photoperiod (approximately 12:12 L:D).
One day before release, we marked males by cool-
ing them for several minutes and placing a dot of
enamel paint on the thorax. This procedure had
no obvious adverse effects, and males resumed
normal activities within minutes of handling.
When used in the field experiments, males of both
species were 10-19 days old.
Effect of CL on Trap Catch of B. dorsalis
We conducted 4 experiments to assess the ef-
fect of CL on the capture of B. dorsalis males in
ME-baited traps. Lures (2 ml of either mixed or
pure lures, see below) were applied to cotton
wicks (1.2 cm diameter, 4.0 cm length) with a pi-
pette. Wicks were placed individually in perfo-
rated, plastic baskets, which were suspended
inside Jackson traps above a sticky insert resting
on the trap floor. In each experiment, we placed
traps in 8 trees arranged in a circle (radius 40 m)
about a central tree, which served as the release
point. In any given replicate, 4 of the trees con-
tained a single ME-baited Jackson trap. The re-
maining 4 trees contained the following: in
Experiment 1, a single trap baited with a mixture
of CL and ME (2 ml of each lure were mixed, and
the mixture was apportioned equally between 2
wicks held in 2 separate baskets); in Experiment
2, a single trap baited with CL and ME applied
separately to 2 wicks (and housed in separate bas-
kets); in Experiment 3, 2 traps baited with CL
and ME, respectively, placed 1 m apart; or, in Ex-
periment 4, 2 traps baited with CL and ME, re-
spectively, placed 3 m apart. Traps were placed 2
m above ground in shaded locations within the
December 2004
Shelly et al.: Trapping Male Fruit Flies
tree canopy. For a given replicate, adjacent test
trees contained different treatments (i.e., ME
only, or some combination of ME and CL).
For a given replicate, we set out the traps be-
tween 0900-1000 h and then released 300 males
at the central release point. Flies were released
by placing 2 buckets (150 males per bucket) on the
ground beneath the release tree and gently re-
moving the screen cover from the bucket. The
buckets were not tapped or shaken, and the flies
exited the bucket on their own volition. Traps
were collected 48 h after release, and the numbers
of released (marked) and wild (unmarked) B. dor-
salis males were recorded. In general, successive
releases were separated by an interval of 2-4 days
to allow previously released flies time to disperse
from the study area. The same test trees were
used in all experiments, but the treatment as-
signed to a particular tree was alternated be-
tween successive replicates in a given
experiment. Ten replicates were conducted for
each experiment.
Effect of ME on Trap Catch of B. cucurbitae
We repeated Experiments 1-4 to assess the ef-
fect of ME on capture ofB. curcubitae males in CL-
baited traps. These experiments followed the pro-
cedures described above, except that 300 B. cucur-
bitae males were released, and CL was used to bait
traps on trees that had only a single male attrac-
tant present. Experiments involving release of
B. cucurbitae were replicated 8 times except Ex-
periment 4 for which 6 replicates were performed.
Statistical Analysis
For both species, pairwise comparisons were
made by the 2-tailed t-test. Raw data were used
for both species as the assumptions of normality
and homoscedasticity were met, except for Exper-
iment 1 with B. cucurbitae for which a log,1 trans-
formation of the raw data was performed to meet
these criteria. Calculations were performed with
SigmaStat Statistical Software (Version 2.0).
RESULTS
Data regarding trap captures of B. dorsalis
males are presented in Table 1. Traps containing
both ME and CL captured significantly fewer B.
dorsalis males than traps baited with ME only.
This result was evident for released and wild
males both when the lures were mixed together
(Experiment 1) and when they were presented
separately (i.e., on different wicks) in the same
trap (Experiment 2). In addition, CL-baited traps
placed 1 m from ME-baited traps on the same tree
reduced trap catch of released B. dorsalis males
relative to ME-baited traps occurring singly on
trees (Experiment 3). A similar result was ob-
trained for wild males, but in this case the effect
was only marginally significant. No reduction in
trap catch of B. dorsalis males (released or wild)
was apparent when CL- and ME-baited traps
were separated by 3 m on the same tree (Experi-
ment 4). Wild males of B. cucurbitae were not
found in any trap baited solely with ME in any of
the experiments.
Data from the experiments involving B. cucur-
bitae are presented in Table 2. Traps containing
mixed lures (Experiment 1) captured significantly
more released B. cucurbitae males than traps
baited with CL only, while a similar, but margin-
ally significant, result was found for wild males.
When CL and ME were presented separately (i.e.,
on different wicks) in the same trap (Experiment
2), a marginally significant increase in released
males was recorded, but no effect was noted for
wild males. The mean number of released B. cu-
curbitae males captured in CL-baited traps
placed 1 m from ME-baited traps was signifi-
cantly greater than that recorded for CL-baited
traps placed singly on trees, but no difference was
evident for wild males (Experiment 3). ME-baited
traps placed 3 m from CL-baited traps had no ef-
fect on trap catch of wild or released males (Ex-
periment 4). Wild males of B. dorsalis were not
found in any trap baited solely with CL in any of
the experiments.
DISCUSSION
Management
The present results for B. dorsalis are consis-
tent with those of previous studies (Hooper 1978;
Vargas et al. 2000), which showed that when pre-
sented in the same trap as ME (mixed or sepa-
rate), CL significantly reduced capture of B.
dorsalis males. Our study further showed that
trap catch of B. dorsalis males was reduced
(though the effect was only marginally significant
for wild males) even when CL was placed 1 m (but
not 3 m) from ME-baited traps. With respect to B.
cucurbitae, the present findings agree with the
Taiwan study (cited by Hooper 1978) in which the
2 lures mixed together increased capture ofB. cu-
curbitae males (though the effect was only margin-
ally significant for wild males). Also, when ME was
placed 1 m from CL-baited traps, there was an in-
crease in the number of released (but not wild)
males captured. There were, however, no signifi-
cant effects detected when ME was presented sep-
arately in the same trap or 3 m away from CL-
containing wicks. Thus, in general, the inhibitory
effect of CL on capture of B. dorsalis males ap-
pears greater than the synergistic effect of ME on
capture of B. cucurbitae males. The absence of B.
dorsalis males in traps baited with CL only and B.
cucurbitae males in traps baited with ME confirms
earlier laboratory results (Metcalf et al. 1983).
Florida Entomologist 87(4)
TABLE 1. NUMBER OF RELEASED AND WILD B. DORSALIS MALES CAPTURED IN JACKSON TRAPS BAITED WITH METHYL
EUGENOL ONLY AND PLACED SINGLY ON TREES (DESIGNATED ME) VERSUS JACKSON TRAPS BAITED WITH A
METHYL EUGENOL-CUE LURE MIXTURE (ME+CL, EXPERIMENT 1), METHYL EUGENOL AND CUE LURE PRE-
SENTED SEPARATELY (ME+CL, EXPERIMENT 2), OR METHYL EUGENOL ONLY BUT PLACED ON A TREE WITH
CUE LURE-BAITED JACKSON TRAP EITHER 1 M ME+CL, EXPERIMENT 3) OR 3 M (ME+CL, EXPERIMENT 4)
APART. VALUES REPRESENT MEANS +1 SD; 10 REPLICATES WERE CONDUCTED FOR EACH EXPERIMENT. SIG-
NIFICANCE LEVELS (df= 18 IN ALL TESTS): *p < 0.05, **p < 0.01, ***p < 0.001, "p < 0.10, "p > 0.10.
Experiment Trap type
A. Released males Cue lure placement relative to methyl eugenol ME ME+CL t
1 Same trap, same wick 36.8 (14.2) 8.6 (3.4) 6.1***
2 Same trap, separate wick 36.6 (16.7) 19.3 (12.1) 2.2*
3 Separate trap, 1 m away 39.8 (13.9) 22.2 (13.0) 2.9**
4 Separate trap, 3 m away 31.5 (17.4) 33.7 (21.9) 0.2N
B. Wild males
1 Same trap, same wick 182.0 (76.7) 81.1 (56.6) 3.3**
2 Same trap, separate wick 239.2 (65.8) 143.8 (47.4) 3.7**
3 Separate trap,1 m away 128.2 (54.3) 82.9 (49.9) 1.8m
4 Separate trap, 3 m away 199.5 (69.6) 159.1 (79.7) 1.0N
These data imply that where B. dorsalis (or an-
other ME-responding species) is dominant, the
most effective trapping procedure would involve
placement of separate ME and CL traps in differ-
ent trees. However, where B. cucurbitae (or an-
other CL-responding species) is most abundant,
mixing ME with CL in the same trap may be most
effective. In addition, the price of ME is only
about 20% that of CL (J. Knapp, pers. comm.), and
therefore use of a ME-CL mixture could reduce
costs considerably.
Lure Interactions and Evolution
Plants containing ME or RK are attractive to
Bactrocera males, apparently because they signal
the presence of pheromone precursors. Males ofB.
dorsalis that feed on ME (Nishida et al. 1988) or
flowers containing ME (Nishida et al. 1997) con-
tain ME metabolites in the rectal gland, the site
of pheromone synthesis and storage. Likewise,
males ofB. cucurbitae that feed on CL (Nishida et
al. 1990) or plants containing RK (Nishida et al.
1993) sequester RK in their rectal gland. To the
extent that the incorporation of ME or RK in the
sex pheromone promotes a species-specific re-
sponse, we might expect males ofB. dorsalis and
B. cucurbitae to possess specific receptors only for
ME and RK/CL, respectively, and (as reported by
Drew & Hooper 1981) to respond to either ME or
RK/CL but not both.
The present study shows clearly, however, that
males of B. dorsalis and B. cucurbitae respond to
the "alternate" lure when ME and CL are mixed
or presented in close proximity. Although phero-
mones, and not plant volatiles, are involved, sim-
ilar types of interspecific interactions are well-
documented for moths (Phelan 1992) and bark
beetles (Byers 1995). It seems unlikely that the
same single receptor cells are responding to both
compounds, resulting in competitive blocking on
the surface of the receptor cell membrane. Rather,
as described for other instances of olfactory inter-
action (Mustaparta 1984), the compounds proba-
bly activate different receptor cells and the
concurrent transduction from these different cells
interrupts, in the case ofB. dorsalis, or enhances,
in the case of B cucurbitae, the response at a more
central neurophysiological level.
Why the between-lure interaction is inhibitory
for B. dorsalis but synergistic for B cucurbitae is
unknown but may reflect evolutionary relation-
ships. Based on biochemical changes accompany-
ing the evolution of higher plants, it appears that
both ME and RK are derived from the same,
widely distributed compound (p-hydroxycinnamic
acid), and speciation within the Dacinae subse-
quently led to ME- and RK/CL-responding species
(Metcalf 1979). Because contemporary RK/CL-re-
sponding species also respond to this ancestral
compound, whereas ME-responding species do
not, it further appears that the RK/CL-respond-
ing species are more closely related to the ances-
tral dacines and that ME-responding species are
derived taxa (Metcalf et al. 1983).
If ME-sensitive species arose from an RK-re-
sponding lineage, then there may have been
strong selection for RK inhibition in the incipient
species to avoid hybridization (and unsuccessful
reproductive attempts) with females from basal,
RK-sensitive species. Once again, data from pher-
omonal interactions are illustrative. Specific com-
ponents of the pheromone of Heliothus virescens
inhibit response by H. zea and vice versa (Stadel-
bacher et al. 1983). Cross-inhibition is highly
adaptive in this case, as heterospecific matings
December 2004
Shelly et al.: Trapping Male Fruit Flies
TABLE 2. NUMBER OF RELEASED AND WILD B. curcubitae MALES CAPTURED IN JACKSON TRAPS BAITED WITH CUE LURE
ONLY AND PLACED SINGLY ON TREES (DESIGNATED CL) VERSUS JACKSON TRAPS BAITED WITH A CUE LURE-ME-
THYL EUGENOL MIXTURE (CL + ME, EXPERIMENT 1), CUE LURE AND METHYL EUGENOL PRESENTED SEPA-
RATELY (CL + ME, EXPERIMENT 2), OR CUE LURE ONLY BUT PLACED ON A TREE WITH A METHYL EUGENOL-
BAITED JACKSON TRAP EITHER 1 M (CL + ME, EXPERIMENT 3) OR 3 M (CL + ME, EXPERIMENT 4) APART.
VALUES REPRESENT MEANS +1 SD (LOG, TRANSFORMED DATA WERE USED IN T TEST IN EXPERIMENT 1).
EIGHT REPLICATES WERE CONDUCTED FOR EXPERIMENTS 1-3, AND 6 REPLICATES WERE CONDUCTED FOR EX-
PERIMENT 4. SIGNIFICANCE LEVELS (df= 14 IN EXPERIMENTS 1-3 AND 10 IN EXPERIMENT 4): *p < 0.05, **p
< 0.01; "p < 0.10; -p > 0.10.
Experiment Trap type
A. Released males Cue lure placement relative to methyl eugenol ME ME+CL t
1 Same trap, same wick 16.8 (5.2) 33.3 (11.2) 3.1*
2 Same trap, separate wick 14.9 (8.1) 22.9 (9.2) 1.8M
3 Separate trap, 1 m away 10.2 (3.2) 20.3 (8.2) 3.2**
4 Separate trap, 3 m away 15.0 (5.9) 17.5 (6.1) 0.7N
B. Wild animals
1 Same trap, same wick 29.6 (17.2) 42.7 (21.0) 1.9M
2 Same trap, separate wick 40.5 (13.8) 32.7 (13.2) 1.4N
3 Separate trap 1 m away 55.8 (23.4) 54.0 (21.1) 0.3N
4 Separate trap, 3 m away 57.3 (48.4) 46.2 (24.9) 0.5N
are unsuccessful due to incompatible genitalia
(Stadelbacher et al. 1983). The synergism noted
between ME and CL apparently reflects the ab-
sence of any evolutionary cost to B. cucurbitae of
responding to this chemical blend, which, in turn,
may reflect the infrequency with which ME and
RK co-occur in the same plant species. Without
natural exposure to an ME-RK blend, males of
B. cucurbitae have not evolved avoidance of this
stimulus. Comparing the relative strength of in-
hibitory and synergistic effects supports the reli-
ance on evolutionary costs to explain the observed
interactions, namely that the inhibitory effect of
CL on ME is stronger (reflecting the high cost of
inviable matings) than is the synergistic effect of
ME on CL (reflecting the absence of benefits).
ACKNOWLEDGMENTS
We thank Roger Coralis for permission to work at the
Waimanalo facility and for weather data, Michael McK-
enney for the field-collected B. dorsalis used to start our
colony, Mindy Teruya for assistance in rearing and
marking the flies, and E. Jang and J. Millar for input.
REFERENCES CITED
BYERS, J. A. 1995. Host-tree chemistry affecting coloni-
zation in bark beetles, pp. 154-213 In R. T. Carde and
W. J. Bell [eds.], Chemical Ecology of Insects 2.
Chapman and Hall, New York.
DREW, R. A. I., AND G. H. S. HOOPER 1981. The re-
sponses of fruit fly species (Diptera: Tephritidae) in
Australia to various attractants. J. Aust. Entomol.
Soc. 20: 201-205.
GILBERT, A. J., AND R. R. BINGHAM (Eds.). 2002. Insect
Trapping Guide. 9th Edition. State of California, De-
partment of Food and Agriculture, Sacramento.
HOOPER, G. H. S. 1978. Effect of combining methyl eu-
genol and cuelure on the capture of male tephritid
fruit flies. J. Aust. Entomol. Soc. 17: 189-190.
KUBA, H., T. KOHAMA, H. KAKINOHANA, M. YAMAGISHI,
K. KINJO, Y. SOKEI, T. NAKASONE, AND Y. NAKA-
MOTO. 1996. The successful eradication programs of
the melon fly in Okinawa, pp. 543-550 In B. A.
McPheron and G. J. Steck [eds.], Fruit Fly Pests: A
World Assessment of their Biology and Manage-
ment. St. Lucie Press, Delray Beach, FL.
METCALF, R. L. 1979. Plant, chemicals, and insects:
some aspects of coevolution. Bull Entomol. Soc. Am.
25: 30-35.
METCALF, R. L. 1990. Chemical ecology of the Dacinae
fruit flies (Diptera: Tephritidae). Ann. Entomol. Soc.
Am. 83: 1017-1030
METCALF, R. L., W. C. MITCHELL, AND E. R. METCALF.
1983. Olfactory receptors in the melon fly Dacus cur-
cubitae and the oriental fruit fly Dacus dorsalis.
Proc. Nat. Acad. Sci. (USA) 80: 3143-3147.
MUSTAPARTA, H. 1984. Olfaction, pp. 37-70 In W. J. Bell
and R. T Carde [eds.], Chemical Ecology of Insects.
Chapman and Hall, New York.
NISHIDA, R., K. H. TAN, M. SERIT, N. H. LAJIS, S.
SUKARI, S. TAKAHASHI, AND H. FUKAMI. 1988. Accu-
mulation of phenyl propanoids in the rectal glands of
males of the oriental fruit fly, Dacus dorsalis. Expe-
rientia 44: 534-536.
NISHIDA, R., K. H. TAN, S. TAKAHASHI, AND H. FUKAMI.
1990. Volatile components of male rectal glands of
the melon fly, Dacus cucurbitae Coquillett (Diptera:
Tephritidae). Appl. Entomol. Zool. 25: 105-112.
NISHIDA, R., O. IWAHASHI, AND K. H. TAN. 1993. Accu-
mulation of Dendrobium superbum (Orchidaceae)
fragrance in the rectal glands by males of the melon
fly, Dacus cucurbitae. J. Chem. Ecol. 19: 713-722.
NISHIDA, R., T. E. SHELLY, AND K. Y. KANESHIRO. 1997.
Acquisition of female-attracting fragrance by males
of the oriental fruit fly from a Hawaiian lei flower,
Fagraea beteriana. J. Chem. Ecol. 23: 2275-2285.
486 Florida Entc
PHELAN, P. L. 1992. Evolution of sex pheromones and
the role of asymmetric tracking, pp. 265-314 In B. D.
Roitberg and M. B. Isman [eds.], Insect Chemical
Ecology: An Evolutionary Approach. Chapman and
Hall, New York.
STADELBACHER, E. A., M. W. BARRY, A. K. RAINA, AND J.
R. PLIMMER 1983. Fatal interspecific mating of two
Heliothis species induced by synthetic sex phero-
mone. Experientia 39: 1174-1176.
STEINER, L. F., W. C. MITCHELL, E. J. HARRIS, T. T. KO-
ZUMA, AND M. S. FUJIMOTO. 1965. Oriental fruit fly
eradication by male annihilation. J. Econ. Entomol.
58: 961-964.
TAN, K. H., AND R. NISHIDA. 1996. Sex pheromone and
mating competition after methyl eugenol consump-
tion in the Bactrocera dorsalis complex, pp. 147-153
In B. A. McPheron and G. J. Steck [eds.], Fruit Fly
Pests: A World Assessment of their Biology and Man-
agement. St. Lucie Press, Delray Beach, FL.
)m
ologist 87(4) December 2004
TANAKA, N. L. F. STEINER, K. OHINATA, AND R. OKA-
MOTO. 1969. Low-cost larval rearing medium for
mass-production of Oriental and Mediterranean
fruit flies. J. Econ. Entomol. 62: 967-968.
VARGAS, R. I., J. D. STARK, M. H. KIDO, H. M. KETTER,
AND L. C. WHITEHAND. 2000. Methyl eugenol and
cue-lure traps for suppression of male oriental fruit
flies and melon flies (Diptera: Tephritidae) in Ha-
waii: effects of lure mixtures and weathering. J.
Econ. Entomol. 93: 81-87.
VARGAS, R. I., D. MIYASHITA, AND T. NISHIDA. 1984. Life
history and demographic parameters of three labo-
ratory-reared tephritids (Diptera: Tephritidae). J.
Econ. Entomol. 77: 651-656.
WHITE, I. M., AND M. M. ELSON-HARRIS. 1992. Fruit
Flies of Economic Significance: Their Identification
and Bionomics. CAB International, Wallingford, UK.
Teal et al.: Sucrose in Adult Diet of Caribbean Fruit Fly
EFFECTS OF SUCROSE IN ADULT DIET ON MORTALITY OF
MALES OF ANASTREPHA SUSPENSE (DIPTERA: TEPHRITIDAE)
PETER E. A. TEAL, JENNY M. GAVILANEZ-SLONE AND BARBARA D. DUEBEN
Center for Medical, Agricultural and Veterinary Entomology, USDA-ARS,
1700 SW 23 Dr., PO Box 14565, Gainesville, Fl., 32604, USA
ABSTRACT
Survival of adult male Caribbean fruit flies, Anastrepha suspense (Loew) fed sucrose and
protein in the form of hydrolyzed brewers yeast was studied under greenhouse conditions.
Flies fed either a 3:1 mixture of sucrose and protein (optimal) or just sugar from the day of
adult eclosion showed no appreciable mortality during the 14-day test period. However, flies
fed just protein, or those that were not provided with sugar or protein showed rapid rates of
mortality, with 50% mortality occurring at 1.87 and 1.53 days, respectively, and 95% mortal-
ity occurring at 2.8 and 2.5 days. Switching flies from the optimal diet to either the protein-
only diet or nothing at 7 or 11 days after emergence resulted in values of 50% and 95% mor-
tality, respectively, that were similar to those for flies reared from eclosion on either just pro-
tein or nothing. No significant mortality occurred among males maintained on the optimal
or sugar-only diets or when flies were shifted from the optimal diet to only sugar at either
day 7 or 11 after emergence. These data demonstrate that the flies have an absolute require-
ment for carbohydrate in the adult diet. Additionally, the results indicate that the flies are
incapable of converting of amino acids from protein hydrolysate into precursors useful for
generating metabolic energy in sufficient amounts to sustain life.
Key Words: Caribbean Fruit Fly, dietary sucrose, carbohydrate, protein.
RESUME
La sobrevivencia de los machos adults de la mosca del Caribe alimentados con sucrosa y
protein en la forma de levadura hidrolizada de cerveza fue estudiada bajo condiciones de in-
vernadero. Las moscas alimentadas con una mezcla de 3:1 sucrosa y protein (6ptima) o solo
azucar desde el dia de la eclosi6n del adulto no mostraron mortalidad apreciable durante el
period de pruebas de 14 dias. Sin embargo, las moscas alimentadas solo con protein, o las
moscas no proveidas con azucar o protein mostraron una tasa rdpida de mortalidad, con
50% mortalidad ocurriendo a los 1.87 y 1.53 dias, respectivamente, y 95% mortalidad ocu-
rriendo a los 2.8 y 2.5 dias. Cambiando las moscas de una dieta 6ptima a cualquier de las die-
tas de solo protein o nada a los 7 o 11 dias despu6s de la emergencia result en valores de
mortalidad de 50% y 95%, respectivamente, que fueron similares a los datos para las moscas
criadas desde la eclosi6n en la dieta de sola protein o nada. No mortalidad significativa ocu-
rri6 entire los machos mantenidos en las dietas 6ptima o de solo azucar o cuando la dieta de
las moscas fue cambiada de la dieta 6ptima a la dieta de solo azucar a los 7 o 11 dias despu6s
de la emergencia. Estos datos demuestran que las moscas tienen un requisite absolute para
el carbohidratado en la dieta del adulto. Ademas, los resultados indican que las moscas no
fueron capaces de convertir los aminoicidos de proteina-hidrolizada a los precursores tiles
para general la energia metabolica en cantidades suficientes para mantener la vida.
The Caribbean fruit fly, Anastrepha suspense
(Loew), became established in south Florida in
the early 1960s and spread rapidly through most
of southern and central Florida. The rapid spread
of the pest was due to its host range which in-
cludes at least 80 different fruit and vegetable
hosts commonly found in Florida (Swanson &
Baranowski 1972). Although many of the host
plants are not of economic importance, the ability
to infest over-ripe grapefruit and oranges has re-
sulted in quarantine restrictions being placed on
shipment of not only citrus but also other fruits
including tomato, bell pepper, lychee, mango, avo-
cado, guava and carambola (Greany & Riherd
1993). In order to overcome shipment restrictions
the state of Florida developed and implemented
the Caribbean Fruit Fly Pest Management Sys-
tem (CFFPMS) that has resulted in state certifi-
cation of "Fly Free Zones" (Greany & Riherd
1993). Although the CFFPMS has been effective
and Florida fruit is being shipped around the
world, eradication of the fly would significantly
improve the agricultural economics of citrus and
vegetable production in Florida.
One of the most effective and environmentally
sound methods for eradication of pest insects is
the sterile insect technique (SIT) pioneered by
Knipling (1955). It was first implemented effec-
tively to eradicate the screwworm fly from Florida
(Baumhover et al. 1955; Knipling 1959). SIT has
Florida Entomologist 87(4)
been demonstrated to be an effective tool for sup-
pression and eradication of a number of species of
tephritid flies, particularly when coupled with
other control techniques (Kakinohana et al. 1997;
Steiner et al. 1965, 1970; Wong et al. 1992). In-
deed, SIT commonly is used for both direct and
prophylactic control of Mediterranean (Ceratitis
capitata (Wiedemann)) and Mexican (Anastrepha
ludens Loew) fruit flies in the continental United
States and it has been tested for control of the
Caribbean fruit fly (Burditt et al. 1975; Holler &
Harris 1993). Control is achieved in SIT by mass
release of sterile males which mate with wild fe-
males. Females that mate with sterile males do
not produce viable eggs, and, over time, this re-
sults in population decline and possibly eradica-
tion. The key to optimizing efficacy of SIT is to
produce sterile males that compete as well as, or
out compete, wild males in mating opportunities
with wild females. Although SIT is an effective
population management tool, it is expensive in
terms of both money and time. Thus, cost benefits
must be clearly defined in order to balance the
needs for effective control with funding con-
straints. One of the more significant costs associ-
ated with SIT protocols for the Caribbean fruit fly
is the need to hold mass reared adult flies for as
many as 7 days prior to release because the mass-
produced strains require time to become sexually
mature (Teal et al. 2000, and references therein).
Minimizing the costs associated with adult hold-
ing while still optimizing reproductive perfor-
mance of the flies is clearly a key element in
improving efficacy of SIT management protocols.
One way to do this is to minimize the expense as-
sociated with feeding adult flies prior to release
(see Martinez et al. 1987). Although it is well
known that these flies require food as adults and
that reproduction and effective sexual signaling
are positively impacted by consumption of both
carbohydrate and protein (Epsky & Heath 1993;
Landolt & Davis-Hernandez 1993; Landolt & Siv-
inski 1992) little is known about the absolute ne-
cessities for either protein or sugar in the adult
diet. We were interested in determining if adult
males of the Caribbean fruit fly required sugars,
protein or both for survival. The following reports
the results of studies demonstrating that carbo-
hydrate is absolutely essential for adult survival.
MATERIALS AND METHODS
Insect Cultures
Caribbean fruit flies were obtained as pupae
from a culture maintained at the Division of Plant
Industry, Florida Department of Agriculture and
Consumer Services, Gainesville, Florida. Pupae
were housed in screen cages in a greenhouse
maintained at 23-27C and 50-70% RH, and nat-
ural light conditions (Teal et al. 2000). Newly
closed adults were removed daily, separated by
sex and placed into cages immediately after emer-
gence. Females were not used and males were
held in a greenhouse that contained no adult fe-
male flies and maintained under the above condi-
tions. All flies were provided with water
dispensed from a cotton wick pushed through the
top of a covered plastic cup containing 25 ml of
water. Water cups were changed weekly.
Diets vs Survival
For initial studies we assessed the effects of
providing adult males with only water or water
plus diets containing either sugar plus protein,
only sugar or only protein. The sugar plus protein
diet was composed of a cake containing a 3:1 mix-
ture of sucrose and brewers yeast hydrolysate
(ICN Biochemicals, #103304). This diet was con-
sidered to be the "optimal diet" and was used as
the standard diet to which all other diets were
compared. The sugar-only diet was a sugar cube
and the protein-only diet was a cake composed of
yeast hydrolysate. Groups of 25 males were caged
in 30-cm3 screen cages on the day of eclosion and
provided with one of the above diets. Survivorship
was recorded every 24 h. We also conducted an ex-
periment in which we maintained the males on
the optimal diet and then switched to only water,
or sugar-, or protein-only diets on either 7 or 11
days after eclosion. We compared the survival of
flies that were switched to new diets with those
maintained on the optimal diet. A one way analy-
sis of variance was used to test the effect of diet on
survival, and separate analyses were conducted
for flies that were 0, 7, or 11 d old at the time they
were switched to the test diets. Fisher's least sig-
nificant difference test (P = 0.05) was used for sep-
aration of means.
RESULTS AND DISCUSSION:
Results of studies in which we provided males
with only water or water plus diets containing ei-
ther sugar plus protein (optimum diet), or only
sugar, or only protein from the day of adult eclo-
sion showed that sugar was absolutely essential
for adult survival. Flies fed water only or water
plus protein all died within 96 h but essentially
all flies provided with water plus either only
sugar or sugar plus protein lived for the duration
of the experiment (Tables 1 to 3). In fact, in other
experiments in which we fed flies sugar and water
from the day of eclosion we found no significant
mortality for as long as three weeks (n = 6 groups,
25 males; mean mortality at 21 days = 6.7% +
3.67). Transferring flies to protein plus water, or
water only, after feeding for 7 days on the optimal
diet resulted in significant mortality (ca. 50%) on
day 9, and more than 85% males were dead by the
end of the 10th day (Tables 1 and 2). However,
December 2004
Teal et al.: Sucrose in Adult Diet of Caribbean Fruit Fly
TABLE 1. MEAN SURVIVORSHIP OF FLIES ON EACH DAY AFTER BEING FED ONLY PROTEIN. FLIES WERE EITHER PRO-
VIDED WITH ONLY PROTEIN FROM THE DAY OF ECLOSION OR SWITCHED FROM THE OPTIMAL SUGAR PLUS PRO-
TEIN DIET TO ONLY PROTEIN ON EITHER THE SEVENTH OR ELEVENTH DAY AFTER ADULT ECLOSION. ALL FLIES
WERE PROVIDED WITH WATER.
Mean survivorship
Days after feeding of flies fed only protein Mean survivorship of flies Mean survivorship of flies
only protein from day of eclosion* switched to protein on day 7* switched to protein on day 11"
0 100 A 100 A 100 A
1 91.5 A 94.8 A 92.4 A
2 33.0 B 39.7 B 8.0 C
3 8.5 C 8.7 C 2.4 C
4 1.5 C 1.0 C 0.0 C
*Means in the same column or row followed by the same letter are not significantly different by Fisher's least significant differ-
ence test (P = 0.05) applied after ANOVA indicated differences among the means.
flies switched to sugar plus water or maintained
on the optimal diet had no significant mortality
through day 15 (Table 3). Similarly transferring
flies from the optimal diet on the 11th day to pro-
tein plus water or only water resulted in greater
than 90% mortality on day 13 and essentially all
were dead by day 14 (Tables land 2). Mortality
was negligible for males maintained on the opti-
mal diet or when switched to sugar plus water
through day 15 (Table 3). When we compared sur-
vival with respect to time after switching diets we
found that survival at each day was no different
for flies maintained on protein only from eclosion
or if they were transferred from the optimal to
protein on either day 7 or 11 (ANOVA followed by
Fisher's LSD, P = 0.05) (Table 1). However, when
flies were transferred to only water on day 7 fewer
flies died on the second day after transfer than if
they were transferred to water on day 11 (ANOVA
followed by Fisher's LSD, P = 0.05) (Table 2). Sur-
vivorship at other ages was not different. There
was no significant mortality when flies were
maintained on either the optimal diet or when
switched to sugar plus water at 11 days.
A number of studies have demonstrated that
the addition of protein to a carbohydrate source is
important for optimizing reproductive success in
adult Anastrepha species. For example, protein
dietary supplements have been shown to have
positive effects on ovarian maturation and fecun-
dity in females ofA. serpentina, A. ludens, and A.
obliqua (Jacome et al. 1999; Aluja et al. 2001a;
Mangan 2003) and improves sexual performance
of males of at least four species including A. sus-
pensa (Epsky & Heath 1993)A. serpentina,A. stri-
ata, and A. obliqua. (Aljua et al. 2001b). Thus,
dietary protein appears to be an important com-
ponent for reproductive success in allAnastrepha
spp. However, apart from reports that sugars are
necessary dietary requirements (Bateman 1972),
important for long term survival (Jacome et al.
1999), and that starvation, by removal of both
protein and carbohydrate sources from the adult
diet, results in rapid morality, there is little infor-
mation on the dietary need of sugar alone or if
protein can take the place of sugar as an adult nu-
trient. Results of our study indicate clearly that
carbohydrates are absolute dietary requirements,
not only among adult males that have not reached
sexual maturity, but also among sexually mature
adult males. The results demonstrate that adult
male Caribbean flies carry few resources forward
from pupation to the adult stage that can be
tapped for energy utilization and store only lim-
TABLE 2. MEAN SURVIVORSHIP OF FLIES ON EACH DAY AFTER ALL FOOD WAS WITHHELD. FLIES WERE EITHER STARVED
FROM THE DAY OF ECLOSION OR SWITCHED FROM THE OPTIMAL SUGAR PLUS PROTEIN DIET TO NO FOOD ON
EITHER THE SEVENTH OR ELEVENTH DAY AFTER ADULT ECLOSION. ALL FLIES WERE PROVIDED WITH WATER.
Days after Mean survivorship of flies Mean survivorship of flies Mean survivorship of flies
removing food starved from day of eclosion* switched to no food on day 7* switched to no food on day 11"
0 100 A 100 A 100 A
1 81B 90.7 A, B 85.6 A, B
2 21.5 D 54.6 C 8.0 D, E
3 4.0 E 5.8 D, E 1.6 E
4 0.0 E 0.5 E 0.0 E
*Means in the same column or row followed by the same letter are not significantly different by Fisher's least significant differ-
ence test (P = 0.05) applied after ANOVA indicated differences among the means.
Florida Entomologist 87(4)
TABLE 3. MEAN SURVIVORSHIP OF FLIES ON EACH DAY AFTER BEING FED ONLY SUGAR. FLIES WERE EITHER PROVIDED
WITH ONLY SUGAR FROM THE DAY OF ECLOSION OR SWITCHED FROM THE OPTIMAL SUGAR + PROTEIN DIET TO
JUST SUGAR ON EITHER THE SEVENTH OR ELEVENTH DAY AFTER ADULT ECLOSION. ALL FLIES WERE PROVIDED
WITH WATER.
Daysafter feeding Mean survivorship of flies Mean survivorship of flies Mean survivorship of flies
sugar only starved from day of eclosion* switched to no food on day 7* switched to no food on day 11"
0 100 100 100
1 100 100 100
2 100 99.8 100
3 100 98.6 98
4 99.3 96.6 94.7
*Means in the same column or row are not significantly different in an ANOVA or Fisher's least significant difference test (P =
0.05). Mean survivorship of flies fed protein plus sugar for the first 4 days was 98.67%; for 11 days it was 99.6%, and for 15 days it
was 99.6%.
ited energy reserves as adults despite being pro-
vided with far more food, in the form of
hydrolyzed yeast protein, than is required. This
suggests strongly that flies maintain only limited
supplies of glycogen in muscle and fat body tissue
and have limited capacity to convert lipids from
the fat body into energy. Perhaps more surprising
is the apparent inability of flies to convert dietary
protein resources to energy in the absence of car-
bohydrate as is evidenced by the fact that flies
provided with only protein or switched from the
complete diet to protein alone at 7 or 11 days ex-
hibited the same rates of survival as males pro-
vided with water only. This is intriguing given
that the protein fed to males was rich in amino ac-
ids such as proline, alanine, aspartate, and
glutamate which are readily converted by some
insects to pyruvate or oxaloacetate or a-ketoglu-
tatrate for use in the Krebs cycle. Indeed, com-
plete catabolism of one mole of proline can yield
14 moles of ATP and in some Diptera, such as the
Tsetse fly (Glossina morsitans, Westwood), pro-
line is the major source for ATP generation
(Bursell 1981). This suggests strongly that tran-
samination of normally important amino acids is
not a primary method of substrate production for
the Krebs cycle of these flies.
The absolute requirement for sugar and inabil-
ity of the flies to utilize protein for metabolic en-
ergy may reflect a physiological adaptation to
environmental conditions. In the tropics, where
these flies evolved, sugars from fruit and nectar
are available at all times of the year but protein is
a relatively limited resource (Bateman 1972; Hen-
drichs et al. 1993). Indeed, protein dietary supple-
ments have been shown to have positive effects on
ovarian maturation and fecundity in A. serpen-
tina,A. ludens, andA. obliqua (Jacome et al. 1999;
Aluja et al. 2001a; Mangan 2003) indicating that
dietary protein is a critical component for egg pro-
duction and, consequently, reproductive success
in all Anastrepha species. Thus, the flies have
probably developed a physiological strategy in
which they utilize the most available food source,
sugars, to insure survival, and take advantage of
limited protein resources when available to
achieve sexual maturity (Bateman 1972).
ACKNOWLEDGMENTS
Use of a trade, firm or corporation name in this pub-
lication is for information and convenience of the reader.
Such use does not constitute official endorsement or ap-
proval by the United States Department of Agriculture
or the Agriculture Research Service of any product or
service to the exclusion of others that may be suitable.
REFERENCES CITED
ALUJA, M., F. DIAZ-FLEISCHER, D. R. PAPAJ, G. LA-
GUNES, AND J. SIVINSKI. 2001a. Effects of age, diet,
female density and host resource on egg load in
Anastrepha leudens and Anastrepha obliqua
(Diptera: Tephritae). J. Insect Physiol. 47: 975-988.
ALUJA, M., I. JACOME, AND R. MACIAS-ORDONEZ. 2001b.
Effect of adult nutrition on male sexual performance
in four neotropical fruit fly species of the genus
Anastrepha (Diptera: Tephritidae). J. Insect Behav-
ior 14: 759-775.
BATEMAN, M. A. 1972. The ecology of fruit flies. Annu.
Rev. Entomol. 17: 493-518.
BAUMHOVER, A. H., A. J. GRAHAM, B. A. BITTER, D. F.
HOPKINS, W. D. NEW, F. H. DUDLEY, AND R. C.
BUSHLAND. 1955. Screwworm control through release
of sterilized flies. J. Econ. Entomol. 48: 462-466.
BURDITT, A. K., D. LOPEZ, L. F. STEINER, D. L. WINDE-
GUTH, R. M. BARANOWSKI, AND M. ANWAR 1975. Ap-
plication of sterilization techniques to Anastrepha
suspense (Loew) in Florida, USA, pp. 93-101 In Pro-
ceedings of Panel on Sterile Principle for Insect Con-
trol. International Atomic Energy Agency. Vienna,
Austria.
BURSELL, E. 1981. The role of proline in energy metabo-
lism, pp. 135-154 In R. G. H. Downer [ed.], Energy
Metabolism in Insects. Plenum Press, New York.
EPSKY, N. D., AND R. R. HEATH. 1993. Food availability
and pheromone production by males of Anastrepha
suspense (Diptera: Tephritidae). Environ. Entomol.
22: 942-947.
December 2004
Teal et al.: Sucrose in Adult Diet of Caribbean Fruit Fly
GREANY, P. D., AND C. RIHERD. 1993. Caribbean fruit fly
status, economic importance and control (Diptera:
Tephritidae). Florida Entomol. 76: 209-211.
HOLLER, T. C., AND D. L. HARRIS. 1993. Efficacy of ster-
ile release of Anastrepha suspense adults against
wild populations, pp. 329-333 In M. Aluja and P.
Liedo [eds.], Fruit Flies: Biology and Management.
Springer-Verlag, New York.
JACOME, I., M. ALUJA, AND P. LIEDO. 1999. Impact of
adult diet on demographic and population parame-
ters in the tropical fruit fly Anastrepha serpentina
(Diptera: Tephritidae). Bull Entomol. Res. 89: 165-
175.
KNIPLING, E. F. 1955. Possibilities of insect control or
eradication through the use of sexually sterile
males. J. Econ. Entomol. 48: 459-462.
KNIPLING, E. F. 1959. Screwworm eradication: Concepts
and research leading to the sterile male method.
Smithsonian Institution Publication 4365. Smithso-
nian Institution. 409-418.
LANDOLT, P. J., AND K. M. DAVIS-HERNANDEZ. 1993.
Temporal patterns of feeding by Caribbean fruit flies
(Diptera: Tephritidae) on sucrose and hydrolyzed
yeast. Ann. Entomol. Soc. Am. 86: 749-755.
LANDOLT, P. J., AND J. SIVINSKI. 1992. Effects of time of
day, adult food, and host fruit on incidence of calling
by male Caribbean fruit flies (Diptera: Tephritidae).
Environ. Entomol. 21: 382-387.
MANGAN, R. L. 2003. Adult diet and male-female con-
tact effects on female reproductive potential in Mex-
ican fruit fly (Anastrepha ludens (Loew)) (Diptera
Tephritidae). J. Econ. Entomol. 96: 341-347.
MARTINEZ, A. J., T. C. HOLLER, AND J. N. WORLEY. 1987.
A fructose and yeast hydrolysate diet for the irradi-
ated Mexican fruit fly adult, Anastrepha ludens
(Loew) and its effect on male longevity. Southwest-
ern Entomol. 12: 317-320.
STEINER, L. F., W. G. HART, E. J. HARRIS, R. T. CUN-
NINGHAM, K. OHINATA, AND D. C. KAMAKAHI. 1970.
Eradication of the oriental fruit fly from the Mariana
Islands by the methods of male annihilation and
sterile insect release. J. Econ. Entomol. 63: 131-135.
STEINER, L. F., W. C. MITCHELL, E. J. HARRIS, T. T. Ko-
ZUMA, AND M. S. FUJIMOTO. 1965. Melon fly eradica-
tion by overflooding with sterile males. J. Econ.
Entomol. 58: 519-522.
SWANSON, R. W., AND R. M. BARANOWSKI. 1972. Host
range and infestation by the Caribbean fruit fly,
Anastrepha suspense (Diptera: Tephritidae) in south
Florida. Proc. Florida State Hort. Soc. 85: 271-274.
TEAL, P. E. A., Y. GOMEZ-SIMUTA, AND A. T. PROVEAUX.
2000. Mating experience and juvenile hormone en-
hance sexual signaling and mating in male Carib-
bean fruit flies. Proc. National Acad. Sci. USA. 97:
3708-3712.
WONG, T., RAMADAN, J. HERR, AND D. MCINNIS. 1992.
Suppression of Mediterranean fruit fly (Diptera: Te-
phritidae) populations with concurrent parasitoid
and sterile fly releases in Kula, Maui, Hawaii. J.
Econ Entomol. 85: 1671-1681.
Florida Entomologist 87(4)
December 2004
A SIMPLE AND EFFECTIVE CYLINDRICAL STICKY
TRAP FOR FRUIT FLIES (DIPTERA: TEPHRITIDAE)
DAVID C. ROBACKER AND MAURA E. RODRIGUEZ
Crop Quality and Fruit Insects Research, Agricultural Research Service
U.S. Department of Agriculture, 2413 E. Highway 83, Bldg. 200, Weslaco, TX 78596
ABSTRACT
A sticky trap for fruit flies was developed that is 2.5x more effective than yellow panel traps
of equal surface area for capture of Mexican fruit flies (Anastrepha ludens (Loew)). The trap
consists of a slightly conical yellow cardboard cylinder coated on the outside surface with
trapping adhesive. In trapping efficacy, these stand-alone cylinders were equivalent to plas-
tic Liquibaitor trap tops with similar cylinders fitted over the trap top with the sticky sur-
face facing outward. Liquibaitor trap tops with cylinders mounted on the inside with their
sticky surfaces facing inward were ineffective, and Liquibaitor tops with cylinders both in-
side and outside were not more effective than those with the sticky surface only on the out-
side. Besides the increased attractiveness of the stand-alone cylinders with the sticky
surface outside, advantages of this design are that lures can be suspended from the trap
hanger inside the cylinder where they do not contact the sticky surface, sticky cylinders can
be changed in the field without disturbing lures that are suspended from the hangers, and
traps can be stacked like Dixie cups for storage and transport.
Key Words: Mexican fruit fly, Anastrepha ludens, trap design, integrated pest management.
RESUME
Una trampa pegajosa para moscas de las frutas que es 2.5x mas efectiva que las trampas de
panels amarillas de igual area de superficia para capturar la mosca mexicana de la fruta
(Anastrepha ludens Loew) fue desarrollada. La trampa consiste de un cilindro de cart6n un
poco c6nico de color amarillo con la superficie exterior cubierta con un pegamento para atra-
par las moscas. En cuanto la eficacia de las trampas, estos cilindros que se sostienen solos
fueron equivalentes a las trampas de tapa "Liquibaitor" con cilindros similares puestos sobre
la trampa con la superficie pegajosa hacia afuera. Las trampas de tapa "Liquibator" monta-
das con la superficie pegajosa hacia adentro fueron inefectivas y las tapas de "Liquibator"
con cilindros con la superficie exterior y la interior pegajosa no fueron mas efectivas que las
trampas con solo la superficie exterior pegajosa. Aparte de que estos cilindros tienen la su-
perficie exterior pegajosa y se pueden sostener solos y atrapan mas moscas, este diseho tiene
la ventaja de que los senuelos pueden ser suspendidos de un gancho puesto dentro del cilin-
dico donde no tiene contact con la superficie pegajosa, se puede cambiar los cilindros pega-
josos en el campo sin disturbar los seAuelos que estan suspendidos de los ganchos y ademis
puede guardar y transportar las trampas una encima la otra como vasos de la marca "Dixie".
Detection with traps is the first line of defense
against exotic fruit flies and a critical element in
programs to control resident species (Robacker &
Landolt 2002). Two principal types of traps are in
general usage: those that induce flies to land and
become trapped on a sticky surface, and those
that lure flies into an enclosed space where they
drown in a liquid reservoir or contact a killing
agent. Which type works better depends on the fly
species and type of lure. Each type has found nu-
merous niches in fruit fly programs around the
world. Because of the need for earlier and more
reliable detection to improve fruit fly control, de-
velopment of better traps of both types is ongoing
in many labs and agencies concerned with fruit
fly management.
Synthetic lures for Mediterranean fruit fly
(Ceratitis capitata Wiedemann) and various
Anastrepha such as the Mexican fruit fly (A.
ludens (Loew)) have been invented during the
last decade (Biolure, Suterra, Inc., Bend, OR;
Anastrepha Fruit Fly Lure, IPM Tech, Portland,
OR). Whereas traditional baits for detection of
Anastrepha were liquid suspensions that re-
quired McPhail-type traps, these new lures can
be readily used with either enclosed traps or
sticky traps. One of these lures (Biolure) has been
used successfully in Multilure traps (Florence
Agri Investment, Inc., Miami, FL) (Thomas et al.
2001) and other McPhail-type traps with liquid
reservoirs including Liquibaitor traps (Interna-
tional Pheromone, South Wirral, UK) (Epsky et
al. 1999; Katsoyannos et al. 1999; Papadopoulos
et al. 2001). Although agencies charged with fruit
fly trapping may actually prefer to use a dry trap,
a change to dry traps is unlikely unless they are
at least as attractive as existing McPhail-type
wet traps. At this time, no commercially available
Robacker & Rodriguez: Cylindrical Fruit Fly Trap
dry traps can match trapping efficacy of McPhail-
type traps.
Our goal is to develop a more effective sticky
trap. In this work, we investigated effect of trap
shape. It is well known that trap shape affects at-
tractiveness of sticky traps. Although numerous
shapes have been tested, little has been published
on shapes other than panels and spheres (Katsoy-
annos 1989; Epsky & Heath 1998). Nakagawa et
al. (1978) showed that cylinders were among the
least attractive shapes to Mediterranean fruit
flies. However, Heath et al. (1997) described a
highly effective cylindrical sticky trap that was
constructed with paper coated with an extremely
tacky dry adhesive (Atlantic Paste and Glue Co.,
Inc., Brooklyn, NY). The relative effects of the cy-
lindrical shape and the highly sticky surface on
performance of the trap were not evaluated.
Because the cylindrical trap developed by
Heath et al. (1997) was so effective, we wanted to
re-investigate effectiveness of cylindrical traps.
In this work we evaluate the cylindrical shape
with a standard sticking agent rather than the
dry adhesive used by Heath et al. (1997). In one
experiment we investigated the effect of having
the sticky surface on the inside vs. outside of cyl-
inders mounted on the top of Liquibaitor trap
tops. In the second experiment, we evaluated
stand-alone cylinders vs. cylinders mounted on
Liquibaitor trap tops. Traps were compared with
a commercially available sticky trap for catching
irradiated, laboratory-colony Mexican fruit flies
released into a citrus orchard.
MATERIALS AND METHODS
Experimental Traps
Four cylindrical trap types were constructed of
yellow cardboard obtained from IPM Tech (Port-
land, OR) coated with Stickem Special (Seabright
Laboratories, Emeryville, CA). For three of these
types, the cardboard was formed into cylinders
that fit snugly either inside or on the outside of
the plastic top of a Liquibaitor trap (often re-
ferred to as an International Pheromone McPhail
trap). The three trap types constructed this way
consisted of Liquibaitor trap tops with sticky cyl-
inders inside, outside, or both inside and outside.
The fourth trap was a stand-alone (without a
Liquibaitor trap top) cylinder with the sticky
coating on the outside, of the same dimensions as
used to fit over the outside of the Liquibaitor trap
top (Fig. 1). The trap is slightly conical with a top
diameter of 13.5 cm, a bottom diameter of 16 cm,
and a height of 13.5 cm. The total sticky surface
area (618 cm2) was approximately equal to that
(644 cm2) of a Pherocon AM trap (Trece, Inc., Sali-
nas, CA). A wire with a loop in the center was fas-
tened across the top diameter of the cylinder. The
loop in the wire served for hanging the trap.
Fig. 1. Stand-alone sticky cylindrical trap for fruit
flies constructed with yellow cardboard coated with
Stickem Special (top diameter, 13.5 cm; bottom diame-
ter, 16 cm; height, 13.5 cm). The lure and trap (at the
loop in the wire) can be suspended separately from the
trap hanger.
Insects
Mexican fruit flies (Anastrepha ludens) were
used to evaluate the effectiveness of the trap.
Flies were from a culture that originated from
yellow chapote fruit, Casimiroa greggii (Ruta-
ceae), collected in Nuevo Leon, Mexico, in 1987.
Flies were irradiated, due to quarantine laws,
with 70-92 Grays (Cobalt 60) 1 to 2 days before
adult eclosion. Mixed-sex groups of 180-200 flies
were kept in 473-ml cardboard cartons with
screen tops until used in tests. Laboratory condi-
tions for holding flies were 22 + 2C, 50 + 20% rel-
ative humidity and photophase from 0630 to 1930
hours provided by fluorescent lights. Flies were
fed sugar and water until they were released in
test plots 3 to 8 days after eclosion.
Field Evaluations
The purpose of these experiments was to test
the efficacy of the experimental traps against the
Pherocon AM (no bait) trap. Pherocon AM traps
are rectangular (14 x 23 cm) yellow cardboard
panels coated with an adhesive similar to Stic-
kum Special. Experiments were conducted in a
mixed citrus orchard located near the laboratory
in Weslaco, Texas. The orchard contained several
varieties of orange, lemon, and tangerine trees of
various ages. One row of Dancy tangerine (C. re-
ticulata) was chosen for tests since it contained
relatively large (2-3 m height) fruit-bearing trees.
IPM Tech Anastrepha Fruit Fly Lures were sus-
pended inside of Liquibaitor trap tops or from the
hanger of the sticky cylinder traps at the loop in
Florida Entomologist 87(4)
the wire. Pherocon AM (no bait) traps with IPM
Tech Anastrepha Fruit Fly Lures attached to the
trap hangers were used as the control. Traps were
located one to a tree, north of center, at 1-2 m
height. Trapped flies were counted and all of the
traps were replaced each week. Lures were used
for the duration of each of the two field experi-
ments. Each week approximately 3000 flies were
released onto trees in a row adjacent to the test
row so as to create a uniform distribution of flies
near the test trees. The first experiment was a test
of the three trap designs with sticky cylinders on
Liquibaitor trap tops compared with Pherocon
traps. Three linear blocks of four consecutive trees
were used in the row, with one buffer tree between
blocks. Each of the three blocks contained one
each of the four trap types. Trap types were ran-
domized within each block the first time traps
were put into the orchard, and then moved se-
quentially within each block when traps were ser-
viced once per week. Eighteen replicates of each
trap type were tested (3 blocks x 6 service weeks).
The second experiment tested sticky cylinders
without Liquibaitor trap tops (Fig. 1) against
sticky cylinders on the outside of Liquibaitor trap
tops, and Pherocon traps. Four linear blocks of
three consecutive trees were used in the row, with
a buffer tree between blocks. Each of the four
blocks contained one each of the three trap types.
Procedure was the same as in the previous exper-
iment. Twenty-eight replicates of each trap type
were tested (4 blocks x 7 service weeks).
Statistical Analyses
The experimental design for both experiments
was a randomized complete block. Replications
over time (service weeks) were treated like repli-
cations over space (blocks of trees) for the purpose
of statistical analyses. Data were subjected to
analysis of variance with SuperANOVA (Abacus
Concepts 1989).
RESULTS AND DISCUSSION
Results of the experiment testing sticky cylin-
ders on Liquibaitor trap tops are shown in Table
1. The analysis of variance was highly significant
(F = 11.4; df = 3,68; P < 0.0001). Liquibaitor trap
tops with sticky cylinders on the inside captured
fewer flies than Pherocon traps. Traps with sticky
cylinders on the outside did not differ in attrac-
tiveness from those with cylinders both inside
and outside. Both of these designs were more at-
tractive than Pherocon traps. Trap types did not
differ regarding percentage of females captured.
Fly captures on stand-alone sticky cylinder
traps did not differ from those on sticky cylinders
fitted on the outside of Liquibaitor trap tops (F =
8.9; df = 2,81; P < 0.001) (Table 2). Both designs
were more attractive than Pherocon traps. Trap
types did not differ regarding percentage of fe-
males captured.
Cylindrical sticky traps with the sticky surface
on the outside, either stand-alone or mounted on
Liquibaitor trap tops, captured about 2.5xmore
Mexican fruit flies than Pherocon panel traps of
approximately the same sticky surface area.
These results indicate that cylinders are more at-
tractive than panels.
As discussed in the introduction, Heath et al.
(1997) described a highly effective cylindrical
sticky trap made with an extremely tacky dry-ad-
hesive paper. This trap captured twice as many
Mexican fruit flies and Mediterranean fruit flies
as glass McPhail traps with the same lures. Rela-
tive importance of cylindrical shape and the trap-
ping adhesive were not evaluated, however, it
now seems likely that the great effectiveness was
at least partly due to the shape.
The cylinder traps with their sticky surfaces
inside Liquibaitor trap tops were designed to
function like a dry version of a McPhail trap. In
both types of traps, flies must enter from below as
they approach the attractive volatiles coming
from either the liquid reservoir or the lure sus-
pended inside the trap top. The poor performance
of these traps was unexpected based on the his-
torical effectiveness of McPhail traps. However,
Heath et al. (1997) also reported poor captures of
Mexican fruit flies with an open-bottom cylindri-
cal dry trap that required flies to enter from the
bottom or through small holes in the side.
Despite great promise, the cylindrical sticky
trap described by Heath et al. (1997) was never
produced commercially, possibly because small
TABLE 1. MEXICAN FRUIT FLY CAPTURES PER WEEK ON STICKY CYLINDERS ATTACHED INSIDE OR OUTSIDE OF LIQUIBAI-
TOR TRAP TOPS COMPARED WITH PHEROCON TRAPS.1'2
Test trap Males Females Total
Pherocon 1.3 0.3 a 2.2 0.4 b 3.5 0.5 b
Sticky cylinder inside 0.3 + 0.2 a 0.2 + 0.2 a 0.5 + 0.2 a
Sticky cylinder outside 4.0 0.8 b 5.4 1.1 c 9.4 1.7 c
Sticky cylinders inside and outside 3.2 0.5 b 4.2 0.6 c 7.4 0.9 c
1All traps were baited with an IPM Tech Anastrepha Fruit Fly Lure.
Means (SE) in the same column followed by the same letter are not significantly different by Fisher's protected LSD test (P < 0.05).
December 2004
Robacker & Rodriguez: Cylindrical Fruit Fly Trap
TABLE 2. MEXICAN FRUIT FLY CAPTURES PER WEEK ON TWO TYPES OF STICKY CYLINDER TRAPS COMPARED WITH
PHEROCON TRAPS.1'2
Test trap Males Females Total
Pherocon 1.7 0.3 a 2.4 0.4 a 4.1 0.6 a
Sticky cylinder on Liquibaitor trap top 5.1 0.8 b 4.7 0.5 b 9.8 1.1 b
Sticky cylinder 5.1 0.7 b 5.0 0.6 b 10.0 1.2 b
All traps were baited with an IPM Tech Anastrepha Fruit Fly Lure.
'Means in the same column followed by the same letter are not significantly different by Fisher's protected LSD test (P < 0.05).
birds and lizards were sometimes trapped due to
the extreme stickiness (Heath et al. 1997). Also,
the sticky surface of these traps adheres flies so
well that fly damage on removal renders identifi-
cation difficult (T. Holler, USDA-APHIS, pers.
comm). Also, experiments with panel traps made
with the same paper (Robacker & Heath 2001) in-
dicated that rain damages both the adhesive and
the paper, making the traps ineffective.
The stand-alone cylindrical sticky trap devel-
oped in this work has numerous features that en-
hance its effectiveness and ease of use. First, the
looped wire spanning the diameter of the top of the
trap provides a point for attachment of the hanger
and suspension of a lure in the center. If the lure is
suspended from the trap hanger, then the dispos-
able sticky trap body can be easily replaced with-
out disengaging the lure. Further, the lure does
not become sticky because it never comes in con-
tact with the sticky surface of the trap. This is im-
portant because commercial synthetic lures are
manufactured so as to last several months. Also,
because the trap is slightly conical, the sticky sur-
face can be covered with wax paper and traps can
be stacked like Dixie cups for shipping and trans-
port to the field. This feature gives this trap an ad-
vantage over spheres and non-conical cylinders.
With regard to the adhesive, neither the cardboard
nor the Stickem Special trapping adhesive are
damaged by rain and trapping of birds or other
small animals has not been observed. Most impor-
tantly, the cylindrical shape makes it much more
attractive than yellow panel traps, greatly improv-
ing detection of Mexican fruit flies and perhaps
other species of Tephritidae.
ACKNOWLEDGMENTS
We thank Darek Czokajlo (IPM Tech, Portland, OR)
for the yellow cardboard used for trap construction and
USDA-APHIS (Mission, Texas) for irradiation of flies.
We thank A. J. Martinez (USDA-APHIS, Mission, TX),
H. N. Nigg (University of Florida, Lake Alfred, FL), and
T. Holler, USDA-APHIS, Gainesville) for critical
reviews of the manuscript. Mention of a proprietary
product does not constitute an endorsement or recom-
mendation for its use by the USDA.
REFERENCES CITED
ABACUS CONCEPTS. 1989. SuperANOVA. Abacus Con-
cepts, Inc., Berkeley, CA.
EPSKY, N. D., AND R. R. HEATH. 1998. Exploiting the in-
teractions of chemical and visual cues in behavioral
control measures for pest tephritid fruit flies. Fla.
Entomol. 81: 273-282.
EPSKY, N. D., J. HENDRICHS, B. I. KATSOYANNOS, L. A.
VASQUEZ, J. P. Ros, A. ZUMREOGLU, R. PEREIRA, A.
BAKRI, S. I. SEEWOORUTHUN, AND R. R. HEATH.
1999. Field evaluation of female-targeted trapping
systems for Ceratitis capitata (Diptera: Tephritidae)
in seven countries. J. Econ. Entomol. 92: 156-164.
HEATH, R. R., N. D. EPSKY, B. D. DUEBEN, J. RIZZO, AND
F. JERONIMO. 1997. Adding methyl-substituted am-
monia derivatives to a food-based synthetic attrac-
tant on capture of the Mediterranean and Mexican
fruit flies (Diptera: Tephritidae). J. Econ. Entomol.
90: 1584-1589.
KATSOYANNOS, B. I. 1989. Response to shape, size and
color, pp. 307-324 In A. S. Robinson and G. Hooper
[eds.], Fruit Flies: Their Biology, Natural Enemies
and Control. Vol. 3A. Elsevier, Amsterdam.
KATSOYANNOS, B. I., R. R. HEATH, N. T. PAPADOPOULOS,
N. D. EPSKY, AND J. HENDRICHS. 1999. Field evalua-
tion of Mediterranean fruit fly (Diptera: Tephritidae)
female selective attractants for use in monitoring
programs. J. Econ. Entomol. 92: 583-589.
NAKAGAWA, S., R. J. PROKOPY, T. T. Y. WONG, J. R. ZIEG-
LER, S. M. MITCHELL, T. URAGO, AND E. J. HARRIS.
1978. Visual orientation of Ceratitis capitata flies to
fruit models. Entomol. Exp. Appl. 24: 193-198.
PAPADOPOULOS, N. T., B. I. KATSOYANNOS, N. A. KOU-
LOUSSIS, J. HENDRICHS, J. R. CAREY, AND R. R.
HEATH. 2001. Early detection and population moni-
toring of Ceratitis capitata (Diptera: Tephritidae) in
a mixed-fruit orchard in northern Greece. J. Econ.
Entomol. 94: 971-978.
ROBACKER, D. C., AND R. R. HEATH. 2001. Easy-to-han-
dle sticky trap for fruit flies (Diptera: Tephritidae).
Fla. Entomol. 84: 302-304.
ROBACKER, D. C., AND P. J. LANDOLT. 2002. Importance
and use of attractants, pp. 169-205 In G. J. Hallman
and C. P. Schwalbe [eds.], Invasive Arthropods in Ag-
riculture: Problems and Solutions. Science Publish-
ers, Enfield, NH.
THOMAS, D. B., T. C. HOLLER, R. R. HEATH, E. J. SALI-
NAS, AND A. L. MOSES. 2001. Trap-lure combinations
for surveillance of Anastrepha fruit flies (Diptera:
Tephritidae). Florida Entomol. 84: 344-351.
Florida Entomologist 87(4)
December 2004
SEASONAL AND NOCTURNAL FLIGHT ACTIVITY OF SPODOPTERA
FRUGIPERDA MALES (LEPIDOPTERA: NOCTUIDAE) MONITORED
BY PHEROMONE TRAPS IN THE COAST OF CHIAPAS, MEXICO
JULIO C. ROJAS, ARMANDO VIRGEN AND EDI A. MALO
Departamento de Entomologia Tropical, El Colegio de la Frontera Sur
Apdo. Postal 36, Tapachula, Chiapas, M6xico, CP 30700
ABSTRACT
We determined seasonal and nocturnal flight activity of Spodoptera frugiperda (J. E. Smith)
males with traps baited with pheromone in the coast of Chiapas, Mexico.A total of 3015, 3065,
and 838 males were captured in 2000, 2001, and 2002, respectively. Pheromone trap catches
decreased approximately 72% during 2002 with respect to 2000 and 2001. One of five experi-
mental sites caught 90% of the total captured. The pattern of trap captures was quite variable
among years and sites. In general, the flight activity of S. frugiperda males was seasonal, with
two distinctive peaks in trap captures during the year. Males were caught during all hours of
scotophase, however, most males were captured during the first 7 h. Highest peak capture was
between 1900-2000 h. Trap captures were positively correlated with wind speed and temper-
ature, and negatively correlated with relative humidity. Significantly more males were cap-
tured at wind speeds of 100-200 and >200 m/min than at wind speeds of 0-100 m/min.
Key Words: Spodoptera frugiperda, pheromones, monitoring, seasonal activity, nocturnal
activity, Mexico.
RESUME
La actividad de vuelo estacional asi como la actividad nocturna de machos de Spodoptera fru-
giperda (J. E. Smith) fue determinada usando trampas cebadas con feromona en la costa de
Chiapas, M6xico. Un total de 3015, 3065 y 838 machos fueron capturados en 2000, 2001 y
2002, respectivamente. La capture de las trampas cebadas con feromona, disminuy6 aprox-
imadamente en un 72% en 2002 con respect al 2000 y 2001. El perfil de capture de machos
de S. frugiperda fue muy similar en 2000 y 2001, pero diferente al del 2002. En los dos prim-
eros aios se obtuvieron grandes captures de machos de S. frugiperda en los meses de enero
a marzo. Por el contrario en 2002, las mayores captures se obtuvieron en agosto. Del total de
machos capturados, mas del 90% fueron capturados en un sitio experimental y el 10% res-
tante en las otras localidades estudiadas. En cuanto al studio de la actividad nocturna de los
machos de S. frugiperda con trampas cebadas con feromona, se encontr6 que la capture se
inici6 desde las primeras horas y se mantuvo durante toda la noche. Sin embargo, la mayoria
de los machos fueron capturados durante las primeras siete horas de la noche, alcanzando el
mayor pico de capture entire las 1900-2000 h. Las captures obtenidas con las trampas corre-
lacionan positivamente con la velocidad del viento y la temperature, y de manera negative
con la humedad relative. Se capturaron de manera significativa mas machos cuando la ve-
locidad del viento fue de 100-200 y mayor de 200 m/min, que cuando fue de 0-100 m/min.
Translation provided by the authors.
The fall armyworm (FAW), Spodoptera fru-
giperda (J. E. Smith) is indigenous to the tropical
regions of the western hemisphere from Argen-
tina to the United States of America. This species
is considered a generalist feeder, feeding on a very
wide host range of plants in several families, with
preference for grasses. In Mexico, FAW is one of
the most important pests of corn and sorghum,
although occasionally it attacks other crops. It fre-
quently is controlled through the use of insecti-
cides. The misuse of insecticides has led to
resistance of S. frugiperda to several insecticides
in Mexico (Pacheco-Covarrubias 1993) and to pos-
sible harmful effects on human health and the en-
vironment (Tinoco & Halperin 1998). Alternatives
for managing this pest are currently being ex-
plored in Mexico, including the use of agents of bi-
ological control, cultural techniques, host plant
resistance and pheromones (Malo et al. 2001; Cis-
neros et al. 2002; Mendez et al. 2002; Farias-Riv-
era et al. 2003; Molina-Ochoa et al. 2003).
Pheromones can be used for direct control of
some insect pests (mass trapping, kill and lure,
and mating disruption) and for monitoring pest
populations. Pheromone-monitoring traps can be
used for detection of a particular species, timing
of control measures, and economic risk assess-
ment of that pest (Wall 1990). In the case ofS. fru-
giperda, pheromone has been used mainly to
monitor male activity, in determining migration,
Rojas et al.: Seasonal and Nocturnal Activity of S. frugiperda Males
seasonal dynamics, and spatial distribution (Tin-
gle & Mitchell 1979;Waddill et al. 1982; Starrat &
McLeod 1982; Pair et al. 1986; Adams et al. 1989;
Mitchell et al. 1989). The possibility of using
pheromone as a direct way of control has been less
explored (Mitchell et al. 1974a).
FAW is a key pest of corn in Chiapas, Mexico,
but information on moth seasonal dynamics and
flight patterns is lacking for this region. The ob-
jective of this study was to provide information on
seasonal and nocturnal flight activity of FAW
males in the coast of Chiapas with traps baited
with sex pheromone. The information obtained
could be integrated to the control measures to
manage S. frugiperda infestations in an effective
manner.
MATERIALS AND METHODS
Seasonal Flight Activity
We selected five sites in three municipalities at
the coast of Chiapas, Mexico (Fig. 1). This region
experiences a humid temperate climate with
heavy rain in the summer, with average annual
rainfall of 2,063 mm, with a rainy season nor-
mally occurring from late May through Novem-
ber. The average annual temperature is 26C, and
April and May are the hottest months. Most of the
land is flat, but there are rolling hills in the
Northeast region. The most common soil types
are luvisol, nitosol andosol, and planosol. A great
diversity of plants are cultivated, including an-
nual (e.g., corn, sorghum and soybean) and peren-
nial crops (e.g., mango, coffee, and cacao). Many
hectares are dedicated to cattle ranching, with
less space devoted to pigs and poultry. Two of the
sites selected were located in the Tapachula mu-
nicipality. In the first site, "Los Toros" (1448'N,
9219'W, 40 masl), 45 ha of soybean are cultivated
once a year (June to October). Mango orchards
and native trees surround this area. In this site,
traps were placed 15 m away from a living fence of
Jatropha curcas L.
The second site is known as "El Manzano"
(1444'N, 9219'W, 20 masl), where about 700 ha
are devoted to cultivation of two crops annually.
Sorghum or corn is cultivated from January to
May, and is watered by a sprinkler irrigation sys-
tem. Soybean is cultivated during the rainy sea-
son from July to October. In this site, traps were
placed 200 m away from a cashew orchard. Two
additional sites were selected in the municipality
of Suchiate; the first known as "20 de Noviembre"
(1442'N, 9216'W, 20 masl), where 108 ha are
dedicated to cultivate star grass (Cynodon nle-
muensis Vanderyst) for feeding cattle. Mango or-
chards and native trees surround the area. In this
site, traps were placed 200 m away from a patch
of native trees. In the second site called "Ciudad
Hidalgo" (1440'N, 9210'W, 25 masl), 13 ha are
devoted to cultivation of corn once a year from
June to September. Banana plantations, mango
orchards and native trees surround the area. In
this site, traps were placed 20 m away from a
patch of native trees. The fifth site known as
"Chincuyo" (1451'N, 9211'W, 155 masl) was lo-
cated at the municipality ofTuxtla Chico, where 2
ha of corn are cultivated during the rainy season
from June to October. Cacao and lemon orchards
surround the site. Traps were placed 20 m away
from a lemon orchard.
At each site, two Scentry Heliothis traps were
placed 1.5 m above the ground on wooden stakes,
spaced more than 30 m apart. Each trap was
baited with a S. frugiperda pheromone bubble cup
lure (Chemtica, Costa Rica) that was replaced
monthly. Moths were collected every 10 d,
counted, and numbers recorded by trap and local-
ity. Trapping was conducted from January 2000 to
December 2002, resulting in a total of 103 obser-
vation dates.
Nocturnal Flight Activity
This experiment was conducted at El Manzano
locality in Tapachula. The experimental field was
planted with sorghum ('V-M') variety sown at a
density of 50,000 plants/ha with row spacing of
0.75 m. Four Scentry Heliothis traps were placed
at intervals of 100 m in a straight line inside the
field, starting with the first trap 100 m from the
crop edge. Traps baited with a S. frugiperda pher-
omone formulated as a bubble cup (Chemtica,
Costa Rica) were hung approximately 1.5 m above
the ground on wooden stakes. When the experi-
ment began, sorghum plants were 25 cm high.
Traps were emptied each h, beginning at 1800 h
(30 min before sunset) until 6000 h (10 min before
dawn). Males caught in each h were killed with
ethyl acetate and counted. Before each observa-
tional night, traps were cleaned and rotated along
the line to remove possible trap bias on male cap-
ture. In addition, temperature, relative humidity,
wind direction, and wind speed each 5 min were
recorded 2 m above ground level at the edge of the
field. The experiment was conducted during 5
nights, from 10-14 February, 2003.
Statistical Analysis
Analysis of variance (ANOVA) was used to de-
termine if the number of males caught varied
through the night. Data were transformed by In
(X + 1) before ANOVA, to correct for heterogeneity
of variances. A possible correlation between the
number of males trapped and the meteorological
parameters recorded was examined by Spearman
rank correlation analysis. Data from all traps
were used for the correlation analysis because
preliminary analysis showed that trap position
did not influence the numbers of males captured.
Rojas et al.: Seasonal and Nocturnal Activity of S. frugiperda Males
92*30'
9230'
92'20'
92*10'
9220'
92"10'
NJ
V.
l"t
/-.
'
. \.j. %% ......,
N '^- *\ '**.;'" -.^ .,.
HOtAUU k..
N, .4'
Fig. 1. Map showing where pheromone traps were placed in Soconusco region, Chiapas, Mexico during 2000-
2002. The sites sampled are marked with black stars.
14*50'
14'40'-
4 1:
CHI JCU 0
,LO TOROS
EL F.lA ZAIIO
*
;;.' DE NOuIEEibPE
CIUDAD HIDALGO
*
PA 1HC OCEAN
'i
N 7
N
II !I
"14"40'
Rojas et al.: Seasonal and Nocturnal Activity of S. frugiperda Males
Also, wind speed was categorized as light (0-100
m/min), moderate (100-200 m/min), or strong
(>200 m/min), and an ANOVA performed among
the traps captures across wind-speed categories.
When statistical significance was found after
ANOVA, treatment means were separated by the
Tukey test procedure. The level of probability con-
sidered significant in all analysis was P < 0.05.
RESULTS
Seasonal Flight Activity
A total of 6770 S. frugiperda males was cap-
tured during the three years of trapping. A total
of 3015, 3065, and 838 males was captured in
2000, 2001, and 2002, respectively. Pheromone
trap catches decreased approximately 72% dur-
ing 2002 compared to 2000 and 2001. The El Man-
zano site caught 93.4% of all males captured
among the sites. The pattern of trap captures was
quite variable among years and sites (Fig. 2). For
example, the pattern of trap captures in El Man-
zano was quite similar during 2000 and 2001, but
slightly different in 2002. In the first two years,
the flight activity of S. frugiperda males was sea-
sonal, with two distinctive peaks in trap catches.
The first peak occurred from January to March
during the dry season, and the second peak be-
tween June and September, during the rainy sea-
son. During 2002, there were also two peaks, the
first between January and February, and the sec-
ond one occurred in August. In the other sites the
flight activity of S. frugiperda was also bimodal,
the first and higher peak occurred between June
and July, whereas a smaller peak was observed
between September and November at the end of
the rainy season. No moths were caught from
January to April during the three years in the ex-
perimental sites Chincuyo, 20 de Noviembre and
Los Toros, except that a small number of males
was caught at Los Toros in 2001 (Fig. 2).
Nocturnal Flight Activity
The trap position within the field did not affect
the number of males captured during the experi-
ment (F = 0.7; df= 3, 16; P = 0.57). The nocturnal
activity of S. frugiperda males began at sunset
and they were active throughout the night, al-
though the number of males trapped varied sig-
nificantly through time (F = 4.5; df = 11, 228; P <
0.0001). Most males were captured during the
first 7 h, with the highest catches between 1900
and 2000 h (Fig. 3). Trap captures were positively
correlated with wind speed (r = 0.29,P = 0.03) and
temperature (r = 0.32, P = 0.016), and negatively
correlated with relative humidity (r = -0.32, P =
0.016). More males were captured at wind speeds
of 100-200 and >200 m/min than at wind speed of
0-100 m/min (F = 4.9; df= 2, 50; P = 0.01) (Fig. 4).
DISCUSSION
Our results show great variation in the num-
ber of S. frugiperda males caught among the dif-
ferent experimental sites, with the "El Manzano"
site capturing most of the insects. Capture of
adults at this site was bimodal, the greatest num-
ber captured from January to March, in the dry
season, with a second smaller peak between June
to September, during the rainy season. In general,
our results from this site are in agreement with
those reported by Raulston et al. (1986), who in-
vestigated the population trends of S. frugiperda
along the Mexican Gulf Coast, the Isthmus of Te-
huantepec, and the Yucatan Peninsula with
Harstack pheromone traps. They found that low
numbers of males were caught during the mid-
portions of the year, while peak captures occurred
either early or late in the year. In contrast, the
highest captures occurred during the rainy season
and a few males were captured in the dry season
at the other trap sites. These results agree with
those of Mitchell et al. (1991), who reported that
in the tropics, S. frugiperda populations have a
tendency to vary with seasonal changes in rain-
fall, with the lowest populations recorded during
the dry seasons. In French Guiana, the highest
populations of adults and larvae of S. frugiperda
occur during the rainy season and the lowest in
the dry seasons (Silvain & Hing 1985). Our study
was conducted in a relatively small area and fac-
tors such as temperature and rainfall are ex-
pected to be quite similar and therefore these
hardly could explain the difference in traps cap-
tures among experimental sites. Gutierrez-Mar-
tinez et al. (1989) reported that temperature and
rainfall did not affect the captures ofS. frugiperda
males in the central area of Chiapas State, Mex-
ico. On the contrary, the wind speed and direction
changes from place to place, and this variation
could explain why traps placed at El Manzano
captured more males than other sites. Also, "El
Manzano" is not surrounded by native or culti-
vated trees (e.g., mangoes) as was the case with
the other experimental sites, which may consti-
tute a barrier for both insects and pheromone dis-
persion. It has been documented that the distinct
plume of pheromone is constructed from the pher-
omone source and a large proportion of males en-
tering it can reach the source when the wind is
strong (Lewis & Macaulay 1976). Mitchell et al.
(1991) reported that favorable wind currents con-
tributed to the distribution of S. frugiperda into
and from different areas in the USA. However, the
most important factor affecting trap captures
seems to be the availability of host plants. For ex-
ample, average host area in El Manzano was 6.8,
16.4, 56.7, and 368.5 times higher than in 20 de
Noviembre, Los Toros, Ciudad Hidalgo and Chin-
cuyo, respectively. Also, at El Manzano, a greater
number of S. frugiperda males/trap/day was
Florida Entomologist 87(4)
December 2004
Los Toros
/\
El Manzano
500
400-
300
200
100
0
20 de Noviembre
0
E
J_
LL 0)
0
U, Q*
7D_
E FCU
1^
UL 0)
iU
(0
10
40 -Chincuyo
D| 30
20
u ) 10-
0
Fig. 2. Mean number of S. frugiperda males caught with pheromone traps at five sites in Soconusco region, Chi-
apas, Mexico during 2000-2002. Lines extending from each dot are standard error of the mean.
20 Cd. Hidalgo
15
10 -
5
0
::, ,\ A ^ ,,,,,,, ,,,,,,.,,A ,,,,
/+t ~h,
Rojas et al.: Seasonal and Nocturnal Activity of S. frugiperda Males
50
30
ab
S20
20 bed bd d T ab c
S a bcd
10 bcd
odd
187 19.8 20.9 21.10 2211 2312 241 1,2 23 34 45 56
Hour of the night
Fig. 3. Nocturnal flight ofS. frugiperda males (mean
+ SE) caught with pheromone traps during five succes-
sive nights from 10-14 February, 2003. Arrows indicate
sunset and sunrise. Values (mean SE) followed by the
same letter indicate no significant differences at the 5%
level according to the Tukey test.
caught when crops (sorghum or corn and soybean)
were present than when crops were absent (Mean
+ SE = 5.8 0.7 and 3.4 0.7, respectively) (t = 2.1,
df = 93, P = 0.04). Gutierrez-Martinez et al. (1989)
showed that the first S. frugiperda males are cap-
tured right after plant emergence, with the high-
est frequency from the 10ththrough the 41st day
of emergence, when the plant is more susceptible.
Pair et al. (1986) reported that the availability
and amounts of susceptible stages of corn planted
in more northerly areas may be the most impor-
tant factors determining the magnitude of S. fru-
giperda populations each year throughout the
southeastern states of the USA. In other moth
species, it also has been shown that host availabil-
ity influences trap captures (Slosser et al. 1987;
Parajulee et al. 1998).
We found that S. frugiperda populations were
lower in 2002 than in 2000 and 2001 despite the
fact that cropping patterns and total crop area did
0-100
100-200
Wind speed (m/min)
200
>200
Fig. 4. Effect of wind speed on S. frugiperda capture
with pheromone traps. Values (mean SE) followed by
the same letter indicate no significant differences at the
5% level according to the Tukey test.
not vary during the three-year study. Thus, host
availability cannot explain the decrease of FAW
populations in 2002. One possible explanation is
that climatic factors affected the populations.
Pair et al. (1986) reported that S. frugiperda pop-
ulations were lower in 1984 than in 1983 or 1985
in the southeastern states of USA. The authors
mentioned that one possible cause of the diminu-
tion of populations in 1984 was the colder temper-
ature, 0C or lower recorded in winter 1983 and
spring 1984 in the study area. It is known that
0C kill S. frugiperda life stages and their host
plants (Luginbill 1928). In our case, the minimal
temperatures recorded in winter were about
19C. Thus, low temperature does not seem to ex-
plain the decrease of populations in 2002. Rain-
fall is another possible factor to explain the lower
capture rates of S. frugiperda males in 2002. Van
Huis (1981) reported that heavy and light rain
kill significant numbers of early instars of FAW.
Data from a meteorological station located 10 km
away from El Manzano show that it rained in De-
cember 2001, but not in December 1999 and 2000.
Thus, it is possible that this rain killed S. fru-
giperda larvae, which reduced adult populations.
We found that S. frugiperda males were
caught throughout the night but trap captures
among night intervals was variable. Most males
were captured during the first 7 h, reaching the
highest capture peak between 1900 and 2000 h.
That males responded to pheromone traps during
the whole night could be due to the fact that we
used synthetic lures which continuously emit
pheromone and response could be independent
from female calling periodicity. Mitchell et al.
(1974b), who used S. frugiperda virgin females as
lures, found that males responded to traps from
0.5 h before sunset to 1.5 h before sunrise, which
is in agreement with our results.
Raina & Menn (1987) indicated that synchro-
nous timing of calling behavior of females and
flight activity of males in Lepidoptera could pro-
vide a high probability of mate finding with a min-
imum expense of energy. We do not have
information about the diel periodicity of calling
behavior of S. frugiperda females in the field, but
under laboratory conditions, 2-d-old virgin fe-
males started to call 90 min after lights were
turned off and individuals called throughout the
night. Females stopped calling as soon as lights
were turned on. Older females began to call 15
min after lights were turned off and calling
peaked 3 h after lights were off (Reyes-Galvez
1999). In another study, mating pairs of S. fru-
giperda were observed throughout the night, al-
though the percentage of moths in copulation
peaked 3-4 h into the dark cycle (Simmons &
Marti 1992). Temporal differences between peak
calling, mating, and peak male flight may be at-
tributed to differences in temperatures and photo-
period regimes. In other moth species, the flight
b
time of males is well synchronized with female
calling (Carde 1974; Sasaki & Riddiford 1984;
Cibrian-Tovar & Mitchell 1991). However, in some
cases, the calling behavior of females and the ac-
tivity cycle of males seems not to be well coordi-
nated because the peak of male flight does not
coincide with the peak of female calling (Sanders
1971; Carde 1974). For instance, females of Chori-
stoneura fumiferana (Clemens) started calling 4.5
h before sunset and the peak of females calling
was reached 2.5 h later. On the other hand, males
were caught at all times during the 24 h period
and the peak catch occurred 1.5 h after sunset.
These results suggest that factors other than the
release of pheromone have strong influence on the
activity of C. fumiferana males (Sanders 1971).
ACKNOWLEDGMENTS
We thank Leopoldo Cruz-L6pez, Antonio Santieste-
ban, and Alejandro del Mazo for assistance during part
of the nocturnal flight periodicity experiment, and Jav-
ier Valle-Mora for help in statistical analysis of data.
Thanks also to Higinio Lopez Urbina for the elaboration
of the map, and to Alfredo Urbina Martinez (Los Toros),
Eloy Cardenas Ozuna (El Manzano), Fortino Galvez To-
ledo (20 de Noviembre), Benigno Reyes Morales (Ciudad
Hidalgo), and Santana Becerra Vazquez (Chincuyo) for
allowing us to use their land for the experiments. This
study received financial support from SIBEJ (project
980501024).
REFERENCES CITED
ADAMS, R. G., K. D. MURRAY, AND L. M. LOS. 1989. Ef-
fectiveness and selectivity of sex pheromone lures
and traps for monitoring fall armyworm (Lepidop-
tera: Noctuidae) adults in Connecticut sweet corn. J.
Econ. Entomol. 82: 285-290.
CARDE, R. T 1974. Diel periodicities of female calling
and male pheromone attraction in Homomelina au-
rantiaca (Lepidoptera: Arctiidae). Can. Entomol.
106: 933-934.
CIBRIAN-TOVAR, J., AND E. R. MITCHELL. 1991. Court-
ship behavior of Heliothis subflexa (Gn.) (Lepidop-
tera: Noctuidae) and associated backcross insects
obtained from hybridization with H. virescens. Envi-
ron. Entomol. 20: 419-426.
CISNEROS, J., J. A. PEREZ, D. I. PENAGOS, J. RUIZ V., D.
GOULSON, P. CABALLERO, R. D. CAVE, AND T. WILL-
IAMS. 2002. Formulation of a nucleopolyhedrovirus
with boric acid for control of Spodoptera frugiperda
(Lepidoptera: Noctuidae) in maize. Biol. Control 23:
87-95.
FARIAS-RIVERA, L. A., J. L. HERNANDEZ-MENDOZA, J.
MOLINA-OCHOA, AND A. PESCADOR-RUBIO. 2003. Ef-
fect of leaf extracts of teozinte, Zea diploperennis L.,
and a Mexican maize variety, criollo 'Uruapefio', on
the growth and survival of the fall armyworm (Lepi-
doptera: Noctuidae). Florida Entomol. 239-243.
GUTIERREZ-MARTINEZ, A., C. LLANDERAL-CAZARES, G.
ZARATE DE LARA, AND R. ALATORRE. 1989. Fluctu-
aci6n poblacional de Spodoptera frugiperda (J. E.
Smith) (Lepidoptera: Noctuidae). Agrociencia 76:
167-174.
December 2004
LEWIS, T., AND E. D. M. MACAULAY. 1976. Design and el-
evation of sex-attractant traps for pea moth, Cydia
nigricana (Steph.) and the effect of plume shape on
catches. Ecol. Entomol. 1: 175-187.
LUGINBILL, P. 1928. The Fall Armyworm. USDA Tech.
Bull. No. 34.
MALO, EDI A., L. CRUZ-LOPEZ, J. VALLE-MORA, A. VIR-
GEN, J. A. SANCHEZ, AND J. C. ROJAS. 2001. Evalua-
tion of commercial pheromone lures and traps for
monitoring male fall armyworm (Lepidoptera: Noc-
tuidae) in the coastal region of Chiapas, Mexico.
Florida Entomol. 84: 659-664.
MENDEZ, W. A., J. VALLE, J. E. IBARRA, J. CISNEROS, D. I.
PENAGOS, AND T. WILLIAMS. 2002. Spinosad and nu-
cleopolyhedrovirus mixtures for control of
Spodoptera frugiperda (Lepidoptera: Noctuidae) in
maize. Biol. Control 25: 195-206.
MITCHELL, R. R., W. W. COPELAND, A. N. SPARKS, AND
A. A. SEKUL. 1974a. Fall armyworm: Disruption of
pheromone communication with synthetic acetates.
Environ. Entomol. 3: 778-780.
MITCHELL, E. R., W. W. COPELAND, AND A. N. SPARKS.
1974b. Fall armyworm: Nocturnal activity of adult
males as indexed by attraction to virgin females. J.
Ga. Entomol. Soc. 9: 145-146.
MITCHELL, E. R., H. R. AGEE, AND R. R. HEATH. 1989.
Influence of pheromone trap color and design on the
capture of male velvetbean caterpillar and fall army-
worm moths (Lepidoptera: Noctuidae). J. Chem.
Ecol. 15: 1775-1784.
MITCHELL, E. R., J. N. MCNEIL, J. K. WESTBROOK, J. F.
SILVAIN, B. LALANNE-CASSOU, R. B. CHALFANT, S. D.
PAIR, V. H. WADDILL, A. SOTOMAYOR-RIOS, AND F. I.
PROSHOLD. 1991. Seasonal periodicity of fall army-
worm, (Lepidoptera: Noctuidae) in the Caribbean
basin and northward to Canada. J. Entomol. Sci. 26:
39-50.
MOLINA-OCHOA, J., R. LEZAMA-GUTIERREZ, M. GONZALEZ-
RAMIREZ, M. LOPEZ-EDWARDS, M. A. RODRIGUEZ-
VEGA, AND F. ARCEO-PALACIOS. 2003. Pathogens and
parasitic nematodes associated with populations of
fall armyworm (Lepidoptera: Noctuidae). Florida
Entomol. 244-254.
PACHECO-COVARRUBIAS, J. J. 1993. Monitoring insecti-
cide resistance in Spodoptera frugiperda popula-
tions from the Yaqui Valley, Son., Mexico. Resistant
Pest Management Newsletter 5: 3-4.
PAIR, S. D., J. R. RAULSTON, A. N. SPARKS, J. K. WEST-
BROOK, AND G. K. DOUCE. 1986. Fall armyworm dis-
tribution and population dynamics in the
southeastern states. Florida Entomol. 69: 468-487.
PARAJULEE, M. N., J. E. SLOSSER, AND E. P. BORING III.
1998. Seasonal activity ofHelicoverpa zea and Helio-
this virescens (Lepidoptera: Noctuidae) detected by
pheromone traps in Rolling Plains of Texas. Environ.
Entomol. 27: 1203-1219.
RAINA, A. K., AND J. J. MENN. 1987. Endocrine regula-
tion of pheromone production in Lepidoptera, pp.
159-174 In G. D. Prestwich and G. J. Blomquist
[eds.], Pheromone Biochemistry. Academic Press,
New York.
RAULSTON, J. R., S. D. PAIR, A. N. SPARKS, J. G. LOERA,
F. A. PEDRAZA, A. PALAMON, A. ORTEGA, J. RUIZ
SANCHEZ, P. C. MARQUEZ, H. A. RULES, J. M.
PEREZ, R. R. RODRIGUEZ, H. R. CARRILLO, R. R. AR-
CHUNDIA, AND F. R. HERRERA. 1986. Fall armyworm
distribution and population dynamics in the Texas-
Mexico Gulf Coast Area. Fla. Entomol. 69: 455-468.
Florida Entomologist 87(4)
Rojas et al.: Seasonal and Nocturnal Activity of S. frugiperda Males
REYES-GALVEZ, G. 1999. Analisis quimico y electrofisi-
ol6gico de la feromona sexual de Spodoptera fru-
giperda (Lepidoptera: Noctuidae) del Soconusco.
Unpublished Bachelor Science thesis, Universidad
Aut6noma de Chiapas, M6xico.
SANDERS, C. J. 1971. Daily activity patterns and sex
pheromone specificity as sexual isolating mecha-
nism in two species of Choristoneura (Lepidoptera:
Tortricidae). Can. Entomol. 103: 498-502.
SASAKI, M., AND L. M. RIDDIFORD. 1984. Regulation of
reproductive behavior and egg maturation in the to-
bacco budworm, Manduca sexta. Physiol. Entomol.
9: 315-327.
SILVAIN, J. F., AND J. TI-A-HING. 1985. Prediction of lar-
val infestation in pasture grasses by Spodoptera fru-
giperda (Lepidoptera: Noctuidae) from estimates of
adult abundance. Florida Entomol. 68: 686-691.
SIMMONS, A. M., AND O. G. MARTI JR. 1992. Mating by
the fall armyworm (Lepidoptera: Noctuidae): Fre-
quency, duration, and effect of temperature. Envi-
ron. Entomol. 21: 371-375.
SLOSSER, J. E., J. A. WITZ, G. J. PUTERKA, J. R. PRICE,
AND A. W. HARTSTACK. 1987. Seasonal changes in
bollworm (Lepidoptera: Noctuidae) moth catches in
pheromone traps in a large area. Environ. Entomol.
16: 1296-1301.
STARRAT, A. N., AND D. G. R. MCLEOD. 1982. Monitoring
fall armyworm, Spodoptera frugiperda (Lepidop-
tera: Noctuidae), moth population in southwestern
Ontario with sex pheromone traps. Can. Entomol.
114: 545-549.
TINGLE, F. C., AND E. R. MITCHELL. 1979. Spodoptera
frugiperda: Factors affecting pheromone trap catches
in corn and peanuts. Environ. Entomol. 8: 989-992.
TINOCO, R., AND D. HALPERIN. 1998. Poverty, production
and health: Inhibition of erythrocyte cholinesterase
through occupational exposure to organophosphate
insecticides in Chiapas, Mexico. Arch. Environ.
Health 53: 29-35.
VAN HUIS, A. 1981. Integrated pest management in the
small farmer's maize crop in Nicaragua. Meded.
Landbou. Wageningen 81-6, 221 pp.
WADDILL, V. H., E. R. MITCHELL, W. H. DENTON, S. L.
POE, AND J. SCHUSTER 1982. Seasonal abundance of
the fall armyworm and velvetbean caterpillar (Lepi-
doptera: Noctuidae) at four locations in Florida.
Florida Entomol. 65: 350-354.
WALL, C. 1990. Principles of monitoring, pp. 9-23 In
R. M. Ridway, R. M. Silverstein, and M. N. Inscoe
(eds.), Behaviour-modifying Chemicals for Insect
Management. Marcel Dekker, Inc., New York.
Florida Entomologist 87(4)
December 2004
OVIPOSITION PREFERENCE OF HOMALODISCA COAGULATA
FOR TWO CITRUS LIMON CULTIVARS AND INFLUENCE OF
HOST PLANT ON PARASITISM BY GONATOCERUSASHMEADI
AND G. TRIGUTTATUS (HYMENOPTERA: MYMARIDAE)
NICOLA A. IRVIN AND MARK S. HODDLE
Department of Entomology, University of California, Riverside, CA 92521, U.S.A.
ABSTRACT
Oviposition preference of Homalodisca coagulata (Say) and two of its mymarid egg parasi-
toids, Gonatocerus ashmeadi Girault and G. triguttatus Girault, for two Citrus limon L. cul-
tivars ('Eureka' and 'Lisbon') was investigated. In laboratory oviposition choice tests, the
number of leaves containing H. coagulata egg masses, the number of H. coagulata egg
masses, and the total number of H. coagulata eggs were significantly higher at 187.2%,
204.2%, and 181.7%, respectively, on 'Eureka' versus 'Lisbon'. In the field, there was no sig-
nificant difference in the number of H. coagulata motiles counted in five-minute searches of
foliage on 'Eureka' and'Lisbon' trees, and numbers of leaves with old (emerged) and new (un-
emerged) H. coagulata egg masses were equivalent between field-planted cultivars. In the
laboratory, parasitism of H. coagulata egg masses by G. ashmeadi and G. triguttatus was
18.6% and 23.2% higher, respectively, for eggs laid on 'Eureka' leaves compared to 'Lisbon',
but these differences were not significant. Leaf surface morphology and thickness of leaf cell
layers of both lemon cultivars were compared with scanning electron microscopy (SEM).
SEM demonstrated that total leaf thickness and the thickness of the palisade layer was
19.2% and 38.6% higher, respectively, in'Eureka' leaves compared to'Lisbon', and that H. co-
agulata egg placement was between the lower epidermis and spongy parenchyma layer for
both cultivars. Furthermore, 'Lisbon' leaves had a smooth underside, whereas 'Eureka'
leaves had many small ridges. The thickness and rough surface of 'Eureka' leaves may be
beneficial for H. coagulata oviposition. However, additional research is required to further
investigate whether leaf characteristics or xylem chemistry are responsible for H. coagulata
oviposition choice. For mass rearing programs with lemons as host plants, it is recommended
that the'Eureka' cultivar be used in preference to 'Lisbon' because H. coagulata prefers this
cultivar for oviposition and parasitoid foraging is not adversely affected.
Key Words: Homalodisca coagulata, Hemiptera, Cicadellidae, Hymenoptera, Mymaridae,
oviposition preference, host plant influence, Citrus limon.
RESUME
La preferencia para la oviposici6n de Homalodisca coagulata (Say) y dos de sus parasitoides
de huevos mymaridos, Gonatocerus ashmeadi Girault y G. triguttatus Girault, en dos varie-
dades de Citrus limon L. ('Eureka' y 'Lisbon') fue investigada. En pruebas del laboratorio
donde pueden escoger el lugar de la oviposici6n, el numero de las hojas que tuvieron masas
de huevos de H. coagulata, el numero de masas de huevos de H. coagulata, y el numero total
de huevos de H. coagulata fueron significativamente mas altos a 187.2%, 204.2%, y 181.7%,
respectivamente, en la variedad 'Eureka' versus 'Lisbon'. En el campo, no hubo una diferen-
cia significativa en el numero de H. coagulata movies contados en una busqueda de 5 minu-
tos del follaje de las variedades de 'Eureka' y 'Lisbon' y el numero de hojas con masas de
huevos de H. coagulata viejas (emergidas) y nuevas (no emergidas) fueron equivalentes en-
tre las variedades sembradas en el campo. En el laboratorio, el parasitismo de las masas de
huevos de H. coagulata por G. ashmeadi y G. triguttatus fue 18.6% y 23.2% mas alta, respec-
tivamente, para los huevos puestos en hojas de 'Eureka' comparado con 'Lisbon', pero estas
diferencias no fueron significativas. La morfologia de la superficie de la hoja y el grueso de
las tapas de celulas de la hoja de ambas variedades de lim6n fueron comparados con un mi-
croscopio electronic de barrido (SEM). El SEM demostr6 que el grueso total de la hoja y el
grueso de la capa empalizada fue 19.2% y 38.6% mas alto, respectivamente, en hojas de
'Eureka' comparado a las hojas de'Lisbon' y que la postura de los huevos de H. coagulata fue
entire la epidermis menor y la capa del par6nquima esponjosa para ambas variedades. Ade-
mas, las hojas de 'Lisbon' tienen el env6s liso, mientras que el env6s en las hojas de'Eureka'
tienen un gran numero de pequeias estrias. El grueso y la superficie aspera de las hojas de
'Eureka' puede ser ben6fico para la oviposici6n de H. coagulata. Sin embargo, se require de
investigaci6n adicional para saber si las caracteristicas de la hoja o la quimica del xilema son
responsables para la selecci6n del sitio de oviposici6n de H. coagulata. Para los programs
Irvin & Hoddle: Oviposition Preference of Homalodisca coagulata
de cria en masa usando limones como plants hospederas, se recomienda que se use la va-
riedad'Eureka' en preferencia a 'Lisbon' por que H. coagulata prefiere esta variedad para la
oviposici6n y la actividad forrajera del parasitoide no es afectada adversamente.
The glassy-winged sharpshooter, Homalodisca
coagulata (Say) (Hemiptera: Cicadellidae), is na-
tive to southeastern U.S.A. and northeastern
Mexico, and following its establishment in Cali-
fornia in the 1980s, it has become a significant
threat to agricultural and ornamental industries
due to its ability to spread the plant pathogenic
bacterium, Xylella fastidiosa Wells et al. (Hop-
kins & Adlerz 1988; Purcell & Saunders 1999;
Purcell & Feil 2001; Hopkins & Purcell 2002;
CDFA 2003; Hoddle et al. 2003). A major classical
biological control program has been launched in
California against H. coagulata, including the
mass rearing and release of two mymarid egg par-
asitoids, Gonatocerus ashmeadi Girault and G.
triguttatus Girault. Citrus limon L. (lemon) has
been chosen as a host plant for H. coagulata rear-
ing and related experiments because it is one of
the preferred hosts for H. coagulata in southern
California and research on this pest is predomi-
nately conducted in agro-ecosystems where citrus
dominates (Perring et al. 2001). However, several
lemon cultivars exist, and two, 'Eureka' and 'Lis-
bon', are the most common commercial cultivars.
It has been well documented that leafhoppers
demonstrate significant oviposition preferences
between plant species and between cultivars of
the same species (McClure 1980; Stiling 1980;
Singh & Agarwal 1988; Catindig et al. 1996;
Sharma & Singh 2002). However, when this work
was conducted it was not known which cultivar of
C. limon H. coagulata preferred for oviposition
under artificial rearing conditions. Consequently,
the research undertaken here sought to deter-
mine the oviposition preference of H. coagulata
for two C. limon cultivars, 'Eureka' and 'Lisbon',
and whether G. ashmeadi and G. triguttatus ex-
hibited preferences for H. coagulata eggs laid on
'Eureka' or 'Lisbon' leaves. To assist with inter-
pretation of preference data collected from the
laboratory, we conducted field surveys ofH. coag-
ulata life stages on 'Eureka' and 'Lisbon' lemons
and compared leaf surface characteristics and
measured leaf cell layers of both cultivars with
scanning electron microscopy (SEM).
MATERIALS AND METHODS
Homalodisca coagulata Oviposition Preference in the
Laboratory
'Lisbon' and 'Eureka' trees approximately two
years of age and grafted to Macrophylla sp. root-
stock were obtained from C & M Nurseries,
Nipomo, CA. Five trees of each cultivar were
pruned to 60 cm in height (mean no. leaves per tree
= 43 + 5.7 for'Eureka' and 37 8.4 for'Lisbon'). The
trees were potted into 4-litre containers and fertil-
ized every two weeks with Miracle-Gro (20 ml/3.5
liters of water, Scotts Miracle-Gro Products Inc.,
Marysville, OH). Each of five cages (30 x 60 x 35
cm) received one tree of each cultivar in a random-
ized design. Cages with trees were held in a green-
house at 26 2C and 30-40% RH under natural
light. Forty female and ten male H. coagulata were
collected from the field and placed into cages hold-
ing experimental plants. Every two days, approxi-
mately 25 field collected female and five male H.
coagulata were added to each cage. The number of
leaves with H. coagulata eggs, the number of egg
masses, and the number of eggs laid on each culti-
var were recorded daily. This study was conducted
over the period May 10-24, 2001. Each combination
of date and replicates was treated as a block, and a
paired comparison t-test at the 0.05 level of signif-
icance was applied to the data in SAS (SAS 1990).
Homalodisca coagulata Oviposition Preference in the
Field
A five-minute search for H. coagulata eggs on
foliage 1-2 m above the ground was conducted on
six 'Eureka' and six 'Lisbon' trees planted as part
of a completely randomized block design variety
trial at the University of California, Riverside Ag-
ricultural Operations Area. Surveyed trees were
17 years of age and cultivar scions were grafted to
Macrophylla sp. rootstock. Old (emerged) and new
(unemerged) H. coagulata egg masses found dur-
ing time searches were harvested from trees and
placed in labeled plastic bags. Additionally, a one-
minute search for adult H. coagulata and nymphs
was conducted on foliage within a 1-2 m band
above ground around the tree for each replicate.
For collected leaves with emerged eggs, the
number ofH. coagulata egg masses, H. coagulata
eggs per egg mass, emerged H. coagulata nymphs,
solitary parasitoid emergence holes, unemerged
nymphs and parasitoids, and 'unemerged un-
knowns' (those that could not be identified) were
recorded for each lemon cultivar. Leaves with un-
emerged egg masses were held at 26 2C and
30-40% RH under a L14:10D photoperiod in 130-
ml plastic vials (40-dram Plastic Vial, Thornton
Plastics, Salt Lake City, UT) filled with deionized
water and 3 ml of antiseptic [(Listerine Antiseptic
Mouthwash, Pfizer Inc., New York, NY) (to pre-
vent bacterial rot)] for two weeks to allow H. coag-
ulata nymphs and parasitoids to emerge before
recording data. This sampling and rearing proto-
col was repeated every two weeks over the period
July 12-October 18, 2001. The total number of
Florida Entomologist 87(4)
H. coagulata (adults and nymphs) counted on cul-
tivars and the number of leaves with H. coagulata
egg masses was log transformed prior to analysis.
Data were compared between cultivars by two-
way ANOVA in SAS (SAS 1990).
For emerged egg masses, percentage parasit-
ism [(the number of solitary parasitoid emergence
holes + the number of unemerged parasitoids)/
(number of H. coagulata eggs) x 100], percentage
nymphs [(the number of emerged H. coagulata
nymphs + unemerged nymphs)/(number of eggs) x
100] and percentage unknowns [(unknowns/num-
ber of eggs) x 100] were calculated. The number of
H. coagulata eggs was square-root transformed,
and egg, H. coagulata nymph, parasitism, and un-
knowns data were compared between cultivars by
two-way ANOVA. For unemerged egg masses, per-
centage parasitism, percentage nymphs and per-
centage unknowns were calculated as above. The
number of H. coagulata eggs was square-root
transformed, and all data were compared between
cultivars by Friedman's X2test. All statistical tests
were performed in SAS and means presented here
are back-transformed.
Gonatocerus ashmeadi and G. triguttatus Oviposition
Preference
Parasitoid colonies were maintained at the
University of California, Riverside at 26 2C
and 30-40% RH under a L14:10D photoperiod in
cages (50 x 40 x 40 cm) on H. coagulata eggs laid
on 'Eureka' leaves. Colonies were provisioned
with honey-water solution (3:1 Natural uncooked
honey, Wild Mountain Brand, Oakland CA). Har-
vested H. coagulata egg masses on excised leaves
were removed from colonies and checked daily for
parasitoid emergence to assure uniform age for
choice experiments. One 'Eureka' and 'Lisbon' leaf
containing approximately 15 H. coagulata eggs
(~24-48 h of age) per leaf were placed through
holes in a lid of a 130-ml plastic vial filled with
deionized water and 3 ml of antiseptic. A second
130-ml plastic vial with ventilation [three 2-cm
holes (one on the bottom, and one on each of two
sides) covered with mesh netting (80 pm Jelliff
Corporation, Southport, CT)] was inverted and at-
tached to the lid of the vial holding the water and
leaves. One newly emerged mated naive female
parasitoid (~24 h old) was placed inside the in-
verted vial that covered the test material and left
for 1 h to forage and oviposit at 26 2C and 30-
40% RH under 1.2 0.2 log lumens/sqm light be-
fore being removed. This set up was replicated 15
times for G. ashmeadi and G. triguttatus. Vials
containing leaves with egg masses exposed to par-
asitoids were held at 26 2C and 30-40% RH
under a L14:10D photoperiod for three weeks to
allow parasitoids to emerge. Percentage parasit-
ism [(the total number of emerged and unemerged
parasitoids/total number of H. coagulata eggs) x
100] was calculated for each 'Eureka' and 'Lisbon'
leaf. Females that were not mated (producing
male only progeny) were excluded from the analy-
sis. Data were transformed by square root and
arcsine transformation and compared between
'Eureka' and'Lisbon' cultivars by paired t- tests at
the 0.05 level of significance in SAS (SAS 1990).
Means presented here are back-transformed.
Investigation of Leaf Characteristics with SEM
One ~2-mm wide leaf section was cut from the
middle of each of 10 fully expanded 'Eureka' and
'Lisbon' leaves (fifth leaf down from the growing
tip of branches from containerized trees) with a
new razor blade and fixed in 2% gluteraldahyde in
0.1 M phosphate buffer for 2 h. Leaf pieces were
washed in buffer and post fixed in 1% osmium
tetroxide in 0.1 M phosphate buffer for 2 h. Leaf
pieces were then washed in distilled water and de-
hydrated in an ethanol series and critical point
dried. Prepared leaf pieces were mounted on SEM
stubs and coated with gold-palladium. Photo-
graphs taken at ~800 x magnification for each of
the 10 samples for each cultivar of the vertical dis-
tribution of leaf cell layer were used to calculate
mean thickness of the waxy cuticle, epidermis,
palisade layer, and spongy parenchyma layer.
Leaf parameters were compared between culti-
vars by two sampled t-test in SAS (SAS 1990). In
addition, SEM was used to examine and photo-
graph the intact leaf surfaces. Finally, 'Eureka'
and 'Lisbon' leaves with H. coagulata egg masses
(sourced from the "H. coagulata oviposition pref-
erence laboratory trial" above) were prepared for
SEM as previously described, and examined to de-
termine where the placement of eggs under the
leaf cuticle on the undersides of leaves occurred.
RESULTS
Homalodisca coagulata Oviposition Preference in the
Laboratory
In the paired choice studies, the number of
leaves containing H. coagulata egg masses, the
number ofH. coagulata egg masses, and the total
number ofH. coagulata eggs were 187.2%, 204.2%
and 181.7% greater, respectively, on'Eureka' trees
compared with 'Lisbon' (leaves: t = 4.95, df= 76, P
< 0.005; egg masses: t = 4.99, df= 76, P < 0.005; to-
tal eggs: t = 4.38, df = 76, P<0.005) (Fig. 1).
Homalodisca coagulata Oviposition Preference in the
Field
There was no significant difference in the total
number of H. coagulata nymphs and adults and
the number of leaves with H. coagulata egg masses
between 'Eureka' and 'Lisbon' cultivars (Table 1).
For emerged egg masses in the field, the number of
December 2004
Irvin & Hoddle: Oviposition Preference of Homalodisca coagulata
s5
..4 U 'Eureka'
E 'LUsbon'
= b
1 a a
Number of waves Number of egg Total number of
masses eggs
Fig. 1. The mean number of leaves containing H. co-
agulata egg masses, the number of H. coagulata egg
masses and the total number of H. coagulata eggs laid
on 'Eureka' and 'Lisbon' lemon trees in the laboratory
[different letters indicate significant (P < 0.05) differ-
ences between cultivars; error bars indicate SEM].
H. coagulata egg masses, total H. coagulata eggs,
percentage parasitism and percentage unknowns
were not significantly different between the two
cultivars (Table 1). However, the percentage
nymphs was significantly higher (5.8% greater) on
'Lisbon' trees compared with 'Eureka' (0%) (Table
1). For egg masses harvested in the field and re-
turned to the laboratory, there were no significant
differences in emergence rates between both culti-
vars for all parameters measured (Table 1).
Gonatocerus ashmeadi and G. triguttatus Oviposition
Preference
In paired oviposition preference tests in the
laboratory, parasitism by G ashmeadi and G trigut-
tatus was respectively 61.7% 8.5 and 35.4%
12.2 on 'Eureka, and on 'Lisbon' parasitism was
43.1% 9.9 and 17.2% + 7.6, respectively, for G.
ashmeadi and G. triguttatus. These differences
were not significant (t = 1.3, n = 15, P = 0.22 and
t = 1.17, n = 15, P = 0.26).
Investigation of Leaf Characteristics with SEM
Mean total leaf thickness was higher (19.2%)
for 'Eureka' leaves (156.8pm 6.2) compared to
'Lisbon' (131.5pm 7.3) (t = 2.63, df= 18, P < 0.05)
(Fig. 2). The mean thickness of the palisade layer
was thicker (t = 2.32, df = 11.1, P < 0.05) in 'Eu-
reka' (38.6%) leaves compared to 'Lisbon' (Fig. 2).
There was no difference in cell layer thickness be-
tween the cultivars for cuticle (t = 0.39, df = 18, P
= 0.70), epidermis (t = 0.60, df= 18, P = 0.56), and
spongy parenchyma layers (t = 1.31, df = 18, P =
0.21) (Fig. 2). The underside of'Lisbon' leaves was
smooth, whereas 'Eureka' had many small ridges
(Fig. 3). Homalodisca coagulata egg placement
was between the lower epidermis and spongy pa-
renchyma layer for both cultivars (Fig. 4).
DISCUSSION
Results from this study showed that in the lab-
oratory, the number of H. coagulata egg masses
laid on 'Eureka' was 204.2% higher than on 'Lis-
bon'. This indicates that when given a choice un-
der artificial rearing conditions, H. coagulata
preferred 'Eureka' over 'Lisbon' for oviposition,
and suggests that this cultivar should be used for
research projects that require large numbers of
H. coagulata eggs laid on lemon leaves that can
be harvested from young potted plants.
TABLE 1. MEAN NUMBERS OF HOMALODISCA COAGULATA EGG MASSES, AND PERCENTAGE NYMPH AND PERCENTAGE
PARASITOID EMERGENCE FROM H. COAGULATA EGGS COLLECTED FROM FIELD PLANTED 'EUREKA' AND 'LIS-
BON' LEMON TREES.
Mean variable/tree/sampling event 'Eureka' SEM 'Lisbon' SEM Significance
Number of H. coagulata nymphs and adults 30.9 1.6 33.0 2.2 F = 0.96, df = 18, P = 0.33
Number of old and new leaves selected 21.0 1.0 24.0 1.1 F = 3.86, df= 1,8 P = 0.06
Emergence from egg masses in the field
Number of old masses 31.1 2.0 35.1 2.1 F = 1.87, df= 1,2 P = 0.18
Number of old eggs 196.5 13.9 212.8 12.9 F = 1.04, df= 1,12 P = 0.31
Percentage parasitism 42.8 2.4 37.8 2.1 F = 2.34, df= 1,12 P = 0.13
Percentage nymphs 26.5 2.0 32.4 1.7 F = 5.53, df= 1,12 P < 0.05
Percentage unknowns 16.8 1.2 19.5 1.3 F = 2.39, df= 1,12 P = 0.13
Emergence from egg masses in the lab
Number of new masses 1.3 0.2 1.2 0.2 X2 = 1.29, df = 1,P = 0.26
Number of new eggs 8.0 1.5 7.6 1.5 X2 = 0.18, df= 1,P = 0.67
Percentage parasitism 53.9 7.6 48.9 6.8 X2 = 1.29, df= 1,P = 0.26
Percentage nymphs 5.8 3.8 0.0 0.0 C2 = 3.00, df = 1, P = 0.08
Percentage unknowns 37.2 7.6 43.0 7.1 X2 = 0.14, df= 1,P = 0.71
Florida Entomologist 87(4)
180
160
140
120
S100
80
660
20 E
no0 -
Leaf cell layer
SEureka
SLisbon
Fig. 2. Mean thickness (um SEM) of the cuticle, epi-
dermis, palisade and spongy parenchyma leaf cell lay-
ers within 'Eureka' and 'Lisbon' lemon cultivars
[different letters indicate significant (P < 0.05) differ-
ences between cultivars].
The reasons for this demonstrated H. coagu-
lata oviposition preference are unknown. Scan-
ning electron microscopy (SEM) showed that the
underside of 'Lisbon' leaves was smooth, whereas
'Eureka' had many small ridges on the cuticle
that may provide female H. coagulata with en-
hanced tarsal grip during oviposition. However,
the depth of these ridges was not quantified. Re-
sults also showed that the thickness of the pali-
sade layer was significantly higher (38.6%) for
'Eureka' leaves compared to 'Lisbon'. A thicker
leaf structure may be more favorable for H. coag-
ulata oviposition.
Female H. coagulata preferentially oviposit on
lower leaf surfaces, and SEM showed that egg
placement occurs between the lower epidermis and
spongy parenchyma layer for both lemon cultivars.
A thick palisade layer, which is positioned at the
top of the leaf, may afford some protection ofH. co-
agulata eggs from adverse conditions. For example,
solar radiation falls onto the upper leaf surfaces so
a thicker palisade layer may insulate eggs from ex-
cessive sun exposure. However, the experimental
design used in the current study did not standard-
ize 'Eureka' and 'Lisbon' trees to possess equivalent
numbers of leaves, or surface areas, which may
have influenced oviposition preference by H. coag-
ulata because of varying light intensity. Other fac-
tors that influence leafhopper oviposition
preference, such as plant chemistry, leaf vein char-
acteristics, number of branches, and trichome den-
sities (Singh & Agarwal 1988; Lit & Bernardo
1990; Denno & Roderick 1991; Andersen et al.
1992; Sharma & Sharma 1997; Sharma et al. 1999;
Sharma & Singh 2002) were not investigated in
this study and may warrant further research if it
becomes necessary to explain in detail factors in-
fluencing H. coagulata oviposition preferences.
Field results showed that there were no signif-
icant differences in the total number ofH. coagu-
/
Fig. 3. SEM photographs of the lower leaf surface
of 'Lisbon' and 'Eureka' Citrus limon cultivars: (A)
'Lisbon' with smooth leaf surface; (B) 'Eureka' leaf
surface with many small ridges on the cuticle that
may enhance H. coagulata tarsal grip during egg lay-
ing; (C) Close up of small ridged on 'Eureka' leaves
that were not present on 'Lisbon'.
lata nymphs and adults, and the number of leaves
with H. coagulata egg masses between 'Eureka'
and 'Lisbon' cultivars. Consequently the oviposi-
December 2004
Irvin & Hoddle: Oviposition Preference of Homalodisca coagulata
-'I!
uI
Fig. 4. SEM photographs showing placement of H.
coagulata egg mass within 'Eureka' and 'Lisbon' lemon
leaves: (A) 'Eureka' leaf-egg sliced latitudinally, shows
placement of egg between epidermis and spongy paren-
chyma; (B) 'Eureka' leaf-egg sliced longitudinally,
shows placement directly beneath epidermis; (C) 'Lis-
bon' leaf-egg sliced latitudinally, shows placement of
egg between epidermis and spongy parenchyma (note:
egg protein filled leaf cells).
tion preference for 'Eureka' demonstrated in the
laboratory was not confirmed in the field. In fact,
field results showed that significantly more (5.9%)
H. coagulata nymphs successfully emerged from
'Lisbon' leaves compared with 'Eureka' (leaf thick-
ness of field-planted trees was not investigated in
this study). The lack of congruence between labo-
ratory and field studies may be due to differences
in environmental factors and tree age between
field-planted (~17 years of age) and small contain-
erized trees (~2 years of age). Environmental as-
pects, such as sunlight, water, soil and nutrients,
may affect leaf characteristics and influence host
oviposition and utilization (DeReffye et al. 1995),
while xylem chemistry varies drastically with
plant age and phenology (Andersen et al. 1992,
1995a,b). It has been demonstrated that H. coag-
ulata is very sensitive to changes in nutritional
quality (Brodbeck et al. 1990, 1999), and in Cali-
fornia Toscano et al. (2003) showed that H. coagu-
lata numbers were over 6-fold higher on young
citrus trees compared to older trees. It is also pos-
sible that given the very high densities of ovipos-
iting females in citrus at the time this field survey
was conducted less preferred lemon cultivars
were being used due to shortages of oviposition
sites on the most preferred cultivars.
Parasitism by G. ashmeadi and G. triguttatus of
H. coagulata eggs masses on'Eureka' and'Lisbon',
were not significantly different. The lack of statis-
tical significance suggests that using 'Eureka'
trees for future experiments with these parasitoids
should not detrimentally influence parasitism rate
data or fecundity estimates, and demonstrates
that parasitoids were not conditioned to oviposit
preferentially on 'Eureka' over 'Lisbon' even
though they had been reared on 'Eureka'.
SUMMARY
Based on the results of oviposition preference
studies conducted in the laboratory we conclude
that the lemon cultivar 'Eureka' is preferred over
'Lisbon' for oviposition by field-collected H. coag-
ulata females. This cultivar preference by H. co-
agulata, however, was not observed in the field.
Furthermore, the mymarid parasitoids G. ash-
meadi and G. triguttatus do not exhibit cultivar
preferences indicating that the use of 'Eureka'
will not adversely affect studies on the reproduc-
tive and developmental biology of Gonatocerus
spp. It is recommended that for mass rearing pro-
grams that require the use of small young con-
tainerized lemons to maximize H. coagulata egg
mass or Gonatocerus spp. production the 'Eureka'
cultivar should be used in preference to'Lisbon'.
ACKNOWLEDGMENTS
This work was supported in part by the California
Department of Food and Agriculture (CDFA) Pierce's
Disease-Glassy-Winged Sharpshooter Management Re-
search Program. We thank David Morgan, CDFA, Mt
Rubidoux Field Station, Riverside, California for sup-
plying parasitoids to initiate our colonies, and Liangwei
Wang for assistance with statistical analysis. We thank
Betsy Boyd, Bryan Carey, and Ruth Vega for assistance
in the field and laboratory.
REFERENCES CITED
ANDERSEN, P. C., B. V. BRODBECK, AND R. F. MIZELL.
1992. Feeding by the leafhopper Homalodisca coag-
ulata in relation to xylem fluid chemistry and ten-
sion. J. Insect Physiol. 38(8): 611-622.
ANDERSEN, P. C., B. V. BRODBECK, AND R. F. MIZELL.
1995a. Water stress- and nutrient solution-mediated
changes in water relations and amino acids and sug-
ars in xylem fluid of Prunus salicina and Lagerstro-
emia indica. J. Amer. Hort. Soc. 120: 36-42.
ANDERSEN, P. C., B. V. BRODBECK, AND R. F. MIZELL.
1995b. Diurnal variations in tension, osmolarity and
the composition of N- and C- assimilates in xylem
fluid of Prunus persica, Vitis hybrid and Pyrus com-
munis. J. Amer. Hort. Soc. 120: 600-606.
BRODBECK, B. V., P. C. ANDERSEN, AND R. F. MIZELL.
1999. Effects of total dietary nitrogen and nitrogen
form on the development of xylophagous leafhop-
pers. Arch. Insect Biochem. Physiol. 42: 37-50.
BRODBECK, B. V., R. F. MIZELL, W. J. FRENCH, R. C
ANDERSEN, AND J. H. ALDRICH. 1990. Amino acids as
determinants of host preference for the xylem feed-
ing leafhopper, Homalodisca coagulata (Homoptera:
Cicadellidae). Oecologia 83: 338-345.
CATINDIG, J. L. A., A. T. BARRION, AND J. A. LITSINGER
1996. Plant host range and life history of the orange
leafhopper Cicadulina bipunctata (Melichar) (Hemip-
tera: Cicadellidae). Philippine Entomol. 10(2): 163-174.
CDFA, 2003. Pierce's Disease Program Report to the
Legislature, May, 2003. California Department of
Food and Agriculture.
DENNO, R. F., AND G. K. RODERICK. 1991. Influence of
patch size, vegetation texture, and host plant architec-
ture on the diversity, abundance, and life history
styles of sap-feeding herbivores, pp. 169-196 In S. S.
Bell, E. D. McCoy, and H. R. Mushinsky [eds.], Habi-
tat Structure: The Physical Arrangement of Objects in
Space. Chinsegut Hill Confer. Population and Com-
munity Biology Series. Chapman and Hall, London.
DERAFFYE, P., F. HOULLIER, F. BLAISE, D. BARTHE-
LEMY, J. DAUZAT, AND D. AUCLAIR 1995. A model
stimulating above- and below-ground tree architec-
ture with agroforestry applications. Agroforestry
Systems 30: 175-197.
HODDLE, M. S. S., V. TRIAPITSYN, AND D. J. W. MORGAN.
2003. Distribution and plant association records for
Homalodisca coagulata (Hemiptera: Cicadellidae) in
Florida. Florida Entomol. 86: 89-91.
December 2004
HOPKINS, D. L., AND W. C. ADLERZ. 1988. Natural hosts
ofXylella in Florida. Plant Disease 72(5): 429-431.
HOPKINS, D. L., AND A. H. PURCELL. 2002. Xylella fas-
tidiosa: cause of Pierce's Disease of grapevine and
other emergent diseases. Plant Disease 86(10):
1056-1066.
LIT, M. C., AND E. N. BERNARDO. 1990. Mechanism of re-
sistance of eggplant (Solanum melongera Linn.) to
the cotton leafhopper, Amrasca biguttula (Ishida) II.
Morphological and biochemical factors associated
with resistance. Phillippine J. Crop Sci. 15(2): 79-84.
MCCLURE, M. S. 1980. Role of wild host plants in the
feeding, oviposition, and dispersal of Scaphytopium
acutus (Homoptera: Cicadellidae), a vector of peach
X-disease. Environ. Entomol. 9(2): 265-274.
PERRING, T. M., C. A. FARRAR, AND M. J. BLUA. 2001.
Proximity to citrus influences Pierce's Disease in Te-
mecula Valley vineyards. California Agriculture
55(4): 13-18.
PURCELL, A. H., AND H. FEIL. 2001. Glassy-winged
sharpshooter. Pesticide Outlook 12(5): 199-203.
PURCELL, A. H., AND S. R. SAUNDERS. 1999. Glassy-
winged sharpshooters expected to increase plant dis-
ease. California Agriculture 53(2): 26-27.
SAS INSTITUTE. 1990. SAS/STAT User's Guide: Statis-
tics Version 6. SAS Institute, Cary, North Carolina.
SHARMA, A., R. SINGH, A. SHARMA, R. SINGH, AND S.
DERRIDJ. 1999. Influence of leaf-vein characteristics
on oviposition preference of the cotton leafhopper,
Amrasca devastans. Bulletin OILB SROP 22(10): 19-
29.
SHARMA, A., AND R. SINGH. 2002. Oviposition preference
of cotton leafhopper in relation to leaf-vein morphol-
ogy. J. Appl. Entomol. 126(10): 538-544.
SHARMA, G. H., AND P. D. SHARMA. 1997. Oviposition be-
haviour of cotton leafhopper, Amrasca biguttula big-
uttula (Ishida) vis-a-via morphological characters of
cotton cultivars. Ann. Plant Protection Sciences 5(1):
15-17.
SINGH, R., AND R. A. AGARWAL. 1988. Role of chemical
components of resistance and susceptible genotypes
of cotton and okra in ovipositional preference of cot-
ton leafhopper. Proceedings of the Indian Academy
of Sciences, Animal Sciences 97(6): 545-550.
STILING, P. D. 1980. Host plant specificity, oviposition
behavior and egg parasitism in some leafhoppers of
the genus Eupteryx (Hemiptera: Cicadellidae). Ecol.
Entomol. 5(1): 79-85.
TOSCANO, N., J. BI, F. BYRNE, AND S. CASTLE. 2003.
Plant-glassy-winged sharpshooter interactions: Bio-
chemical mechanisms involved in host plant selec-
tion between young and old orange trees, pp. 234-
236 In Proceedings of the Pierce's Disease Research
Symposium Dec 8-11, 2003, Coronado, CA.
Florida Entomologist 87(4)
Logarzo et al.: Egg Parasitoid of Sharpshooter
BIOLOGY OF ZAGELLA DELICATE
(HYMENOPTERA: TRICHOGRAMMATIDAE), AN EGG PARASITOID
OF THE SHARPSHOOTER TAPAJOSA RUBROMARGINATA
(HEMIPTERA: CLYPEORRHYNCHA: CICADELLIDAE) IN ARGENTINA
GUILLERMO A. LOGARZO 1, EDUARDO G. VIRLA2, SERGUEI V. TRIAPITSYN3 AND WALKER A. JONES4
1USDA-ARS South American Biological Control Laboratory
USDA-ARS SABCL 3130 Buenos Aires Place, Washington D.C. 20521-3130.
2CONICET-PROIMI, Avenida Belgrano y Pasaje Caseros, T4001MVB
San Miguel de Tucuman, Tucuman, Argentina
3Department of Entomology, University of California, Riverside, CA 92521, USA
4Beneficial Insects Research Unit, USDA-ARS, Kika de la Garza Subtropical Agricultural Research Center
2413 E. Highway 83, Weslaco, TX 78596, USA
ABSTRACT
Research on biological control of glassy-winged sharpshooter (GWSS) Homalodisca coagu-
lata (Say) started in the 1990s. This sharpshooter, vector of Pierce's disease bacteria Xylella
fastidiosa Wells, expanded its geographical distribution and it has become a very serious pest
problem in several crops, especially grapes, in southern California. In 2000, a survey of
sharpshooter egg parasitoids was initiated in Argentina. Fourteen species of egg-parasitoids
were collected. We report here on laboratory studies of adult longevity, oviposition preference,
sex ratio, and development time of Zagella delicate De Santis. Field result of the incidence on
its hosts and seasonal occurrence also are provided. Zagella delicate produced one adult per
host egg. The overall results indicated that 72.5% of the sharpshooter eggs exposed to Z. del-
icata were parasitized. Wasps emergence was 43.8%. In host plant searching preference tests,
Z. delicate females parasitized 66.7% of host eggs on sugar cane, 57.0% of eggs on corn and
4.5% on citrus leaves. The development time (from oviposition to adult emergence) averaged
23.5 1.2 days. The average adult longevity was 10.3 5.8 days. Females lived longer than
males (females: 12.2 5.6 days, males: 6.2 3.7 days). The sex ratio in the laboratory was 1:
2.1 (males/females). In a hyperparasitism test, no adults of Z. delicate emerged from eggs
previously exposed to Gonatocerus tuberculifemur. Seasonal sampling carried out in San
Miguel de Tucuman showed that Z. delicate occurred from spring to fall, with maximum
abundance at the beginning of the spring, where 57.2% out of the 1568 sampled eggs were
parasitized. Field and laboratory data suggest that Z. delicate could be a prospective biolog-
ical control agent against other, exotic, proconiine sharpshooters including H. coagulata.
However, the efficiency ofZ. delicate is restricted to habitats dominated by grasses.
Key Words: glassy-winged sharpshooter, Homalodisca coagulata, Trichogrammatidae, egg
parasitoid, biological control, Zagella delicate.
RESUME
Las investigaciones sobre el control biol6gico de Homalodisca coagulata (Cicadellidae: Pro-
coniini) fueron iniciadas en la d6cada del 90. Esta chicharrita, vector de la bacteria Xylella
fastidiosa, expandi6 su area de distribuci6n geografica y aument6 su abundancia trans-
formandose en una seria plaga de various cultivos, especialmente en la vid, en el sudeste de
California. En el ano 2000 fue iniciada una exploraci6n de parasitoides de huevos de chich-
arritas proconiinas en Argentina, donde fueron colectadas 14 species de parasitoides. Aqui
se reportan los resultados de campo y laboratorio de uno de estos parasitoides, Zagella del-
icata De Santis, aportando informaci6n sobre su bionomia (longevidad de adults, preferen-
cia de oviposici6n, proporci6n de sexos y duraci6n del desarrollo), incidencia sobre su
hospedador en el campo, y ocurrencia estacional. Zagella delicate produjo un solo adulto por
huevo. En general, parasite el 72.5% de los huevos y emergiendo avispas del 43.8% de los
huevos parasitados. En las pruebas de preferencia de oviposici6n de plant hospedadora, las
hembras de Z. delicate parasitaron 66.7% de los huevos en cana de azicar, 57.0% de los huevos
en maiz y fueron atacados 4.5% de los huevos depositados en Citrus. La duraci6n del tiempo
de desarrollo (desde huevo a adulto) fue de 23 1.2 dias. La longevidad promedio de los adul-
tos fue de 10.3 5.8 dias. Las hembras vivieron mas que los machos (hembras: 12.2 + 5.6;
machos: 6.2 3.7 dias). La proporci6n de sexos en el laboratorio fue de 1: 2.1 (machos/hem-
bras). En los studios de hyperparasitismo, ningun adulto de Z. delicate emergi6 de huevos
Florida Entomologist 87(4)
previamente expuestos al mymarido Gonatocerus tuberculifemur. El muestreo estacional re-
alizado en San Miguel de Tucuman mostr6 que Z. delicate aparece desde la primavera hasta
el otofio, con maxima abundancia a principios de la primavera, donde el 57.2% de los 1568
huevos muestreados estaban parasitados. Los studios de campo y laboratorio sugieren que
Z. delicate tiene potencialidades como agent de control biol6gico de otras chicharritas pro-
coniinas ex6ticas incluyendo a H. coagulata. Sin embargo, Z. delicate esta circunscripta a
habitats dominados por gramineas.
Translation provided by the authors.
In the early 1990s, the glassy-winged sharp-
shooter Homalodisca coagulata (Say) (Hemi-
ptera: Cicadellidae) established in California,
USA. It has become a very serious problem as an
efficient vector of Pierce's disease (Xylella fastid-
iosa Wells) (Blua et al. 1999). The insect is native
to the southeastern USA and northeastern Mex-
ico. Although parasitization of H. coagulata can
reach 100% in California, it seems that natural
enemies cannot control the pest (Triapitsyn et al.
1998; Triapitsyn & Phillips 2000). Although the
glassy-winged sharpshooter appears to be adapt-
ing to California environment, it is not certain
that native parasitoids will be as effective there
as they are in their co-evolved native range. Thus,
egg parasitoids of closely related hosts were
sought from areas in South America where sub-
climate types and habitats were similar to that in
California.
In order to identify candidates for a neo-classi-
cal biological control program against this pest, a
survey of proconiine sharpshooter egg parasitoids
was initiated in South America in 2000 (Jones
2001). Most of the collection efforts were carried
out in northwestern Argentina with sentinel egg
masses of Tapajosa rubromarginata (Signoret), a
native South American proconiine sharpshooter.
Tapajosa is close related to Homalodisca; both
genera have the posterior meron exposed and are
included in the same group due to their phyloge-
netic proximity (Young 1968). This survey re-
sulted in collections of nine different species of
the genus Gonatocerus Nees (Hymenoptera: My-
maridae), at least one species of Paracentrobia
Howard, two species of Oligosita Walker, and two
species of Zagella Girault (Hymenoptera: Tri-
chogrammatidae).
To date, there are nine described species ofZa-
gella in the New World; three are from the Nearc-
tic region (Triapitsyn 2003), and six from the
Neotropical region (De Santis 1957, 1970, 1997).
The only known hosts of a Zagella species are H.
coagulata and Oncometopia orbona (Fabricius)
(Cicadellidae: Proconiini) for Z. spirit (Girault)
in the USA (Triapitsyn & Phillips 2000; Tri-
apitsyn 2003). Of the two Zagella species col-
lected from T rubromarginata eggs in Argentina,
Z. delicate was the most abundant in terms of fre-
quency of occurrence and population density. The
other, tentatively identified as Z. platensis (De
Santis) by Triapitsyn, was much less abundant.
Zagella delicate De Santis was described from
a single female without information on its host as-
sociation(s) (De Santis 1970). Detailed taxonomic
and biological studies are indispensable for a bio-
logical control program. Essential biological infor-
mation about a Zagella species has not been
available before this study on the biology, geo-
graphic distribution, and hosts. We report the re-
sults of laboratory and field studies on Z. delicate,
providing information on its bionomics (i.e., adult
longevity, oviposition preference, sex ratio, devel-
opment time), the incidence of its host in the field,
and seasonal occurrence of the parasitoid. The
possibilities of using this egg parasitoid as a po-
tential agent for biological control ofH. coagulata
in the USA are discussed.
MATERIALS AND METHODS
The studies on development time, sex ratio,
adult longevity, and oviposition of Z. delicate on
eggs of T rubromarginata laid on different host
plants were carried out at the USDA-ARS South
American Biological Control Laboratory in Hurl-
ingham, Buenos Aires Province, and at PROIMI,
San Miguel de Tucuman, Tucuman Province.
Field studies included (a) a seasonal sampling of
egg masses in Tucuman Province for the entire
2002 growing season to estimate parasitoid inci-
dence and occurrence, and (b) a survey with sen-
tinel eggs in a wide geographical range between
22 and 42' LS in Argentina.
Laboratory Studies
The initial stock ofZ. delicate was obtained by
collecting 68 egg masses of T rubromarginata on
Johnson grass, Sorghum halepense (L.) Person, in
an open field in Tafi Viejo, Tucuman Province, in
January 2001. Additional collections of egg
masses of T rubromarginata on Johnson grass
were made when necessary in a soccer field near
PROIMI in San Miguel de Tucuman. The colony
of Z. delicate was reared in the laboratory of
PROIMI. About 5-7 wasps, both males and fe-
males, were placed in 20-cm high x 2-cm diameter
glass tubes with 1-2 egg masses (6-15 eggs) of
T rubromarginata and left until the wasps died
(approximately in 5-7 days. The eggs used in the
experiment were 24-48 h old and were laid on the
edge of the distal portions of corn leaves. For aer-
December 2004
Logarzo et al.: Egg Parasitoid of Sharpshooter
ation, the glass tube top was either closed by a ny-
lon mesh or the hole was fitted with cotton plugs
that were moistened with water and honey as
needed. After exposure, the egg masses were
checked daily to ensure the freshness of the
leaves until adult wasps emerged. Percentage of
parasitism of the exposed eggs and percentage of
wasps emerged were calculated on the two gener-
ations obtained in the laboratory. The effects of
host plant species on the preference ofZ. delicate
to parasitize eggs of T rubromarginata was tested
on Citrus sp. (11 egg masses, 148 eggs), sugar
cane (3 egg masses, 51 eggs) and corn (5 egg
masses, 100 eggs). The experiments were con-
ducted at room temperature (24.5+ 6.2C, 70-80%
RH, photoperiod 14:10 h. L/D), and the colony was
maintained under the same conditions.
Host eggs that changed to brownish or reddish
after 5-7 d were considered parasitizedd eggs",
and those that developed eyespots of the host's
nymphs were considered "unparasitized". The
number of sharpshooter nymphs that hatched
from the unparasitized eggs was counted daily.
After 25 d, when parasitoid emergence was nearly
complete, each leaf was dissected and the remain-
ing host eggs were counted. By this time, most
Z. delicate adults had either emerged or attained
the pupal stage, so it was easy to distinguish par-
asitized eggs from unparasitized ones. The per-
centage of parasitism was calculated as follows:
% parasitism = (number of parasitized eggs/ num-
ber of host eggs offered) x 100.
Percentage of wasp emergence was calculated
as follows:
% of wasp emergence = (number of wasps
emerged/ number of offered eggs) x 100.
Test of Hyperparasitism
In field-collected egg masses of T rubromargi-
nata, it was common to observe the adults of
Z. delicate and Gonatocerus spp. emerging from
the same egg mass. Zagella delicate emerged
about 10-15 days after Gonatocerus, suggesting it
could be a hyperparasitoid of the latter, or that it
is not an obligate primary parasitoid (Triapitsyn
2003). Considering that Z. delicate might be used
as biocontrol agent, mixtures of primary parasi-
toids and hyperparasitoids often lead to errone-
ous host parasitoid records (Noyes 1994). We
tested whether Z. delicate could develop in the
eggs of T rubromarginata parasitized by Gonato-
cerus. Seventy-two hours after an egg mass of this
host laid in a sugar cane leaf, it was exposed to
Gonatocerus tuberculifemur (Ogloblin). Then, the
same egg mass was exposed to Z. delicate as ex-
plained above. Nine replications were carried out,
with a total of 100 eggs exposed. Two egg masses
(39 eggs) exposed to Z. delicate with no previous
exposition to G. tuberculifemur served as control.
Adult longevity
Adult longetivy of Z. delicate was monitored
twice a day using 84 individuals (57 females and
27 males) from a few h after emerging (within 12
h) until adult death. Observations were con-
ducted on single wasps in individual vials without
host material, but with honey as a food source.
Differences between male and female longevity
and duration of development were analyzed by a t
test at the 0.05 level of significance (Statistica
5.0). The total time required for the development
from egg to adult was based on 84 individuals.
Field Studies
To estimate parasitic activity a collection of
egg masses was conducted in Tafi Viejo, Tucuman
Province between 4 and 7 January 2001, when 68
egg masses (1018 eggs) were collected on Johnson
grass. In addition, 12 samples were collected from
September 2002 to April 2003 except in December
(spring, summer and part of the fall) in San
Miguel de Tucuman. We sampled a total of 6,253
eggs. For parasitoid emergence, each field-col-
lected egg mass was placed in a Petri dish with
the bottom filled with wet tissue paper. Each Petri
dish was covered with food wrap to prevent eggs
and leaves from dehydration and to keep the
emerging wasps from escaping.
The survey with sentinel eggs was conducted
from November 2000 to December 2002 in north-
western Argentina in areas that closely match the
sub-climate types in the grape-growing regions of
California. It also was conducted in the citrus-
growing areas of eastern Argentina because it was
suggested that the first generation ofH. coagulata
on citrus host plants is under poor natural control
in California (Triapitsyn and Phillips 2000).
Therefore sentinel egg masses in Argentina were
produced on potted citrus plants. Field-collected
females of T rubromarginata were placed in
sleeve cages on citrus plants. They were checked
for presence of eggs daily, and when egg masses
were detected, the sharpshooters were removed.
Sleeves were kept on the plants to avoid uncon-
trolled parasitization from wild parasitoids. The
sentinel eggs on the potted plants were placed in
41 selected sites in Tucuman, Jujuy, Misiones,
Catamarca, La Rioja, Entre Rios, Corrientes, San
Juan, Salta, and Mendoza Provinces. Overall
8,000 eggs were exposed to parasitization.
RESULTS
Laboratory Studies
Zagella delicate is a solitary wasp parasitoid
producing only one adult per host egg. The overall
results showed that out of 353 eggs exposed to
Z. delicate females, 72.5 + 28.3% (Mean SD)
were parasitized, and wasps emerged from 43.8
Florida Entomologist 87(4)
TABLE 1. PARASITIC ACTIVITY OF ZAGELLA DELICATE ON EGGS OF TAPAJOSA RUBROMARGINATA LAID ON THREE DIF-
FERENT PLANT SPECIES IN THE LABORATORY.
Egg masses Egg masses Host nymphs
offered Eggs offered attacked Z. delicate emerged emerged
Corn 5 10 5 57 (57.0 14.5%) 29 (29.0 17.9%)
Sugar cane 3 51 3 34 (66.7 25.0%) 11 (21.6 25.4%)
Citrus 12 148 1 7 (4.7 2.0%) 69 (46.6 4.7%)
41.4% of those. In the Fl progeny, 61.7 40.3% of
the exposed eggs were parasitized, and adult
emergence was 26.9 38.6%. In the F2 progeny,
94.5 6.7% of the eggs were parasitized, and
adult emergence was 85.0 6.0%. Nymphs of
T rubromarginata emerged from 13.3 28.7% of
the exposed eggs.
In host plant searching preference test, Z. del-
icata females parasitized 34 host eggs (66.7 +
25.0%) (3 egg masses) on sugar cane, 57 eggs (57.0
+ 14.5%) (5 egg masses) on corn and 7 (4.7 2.2%)
on citrus leaves (Table 1). In laboratory, Z. deli-
cata females were able to find eggs of T rubro-
marginata on citrus leaves. However, the wasps
commonly failed to oviposit after several at-
tempts. Citrus cuticle leaf is thicker than sugar
cane and corn cuticle leaves, and probably cuticle
thickness is a factor that interferes with oviposi-
tion ofZ. delicate.
Developmental time (from oviposition to adult
emergence) ofZ. delicate averaged 23.5 1.2 days
(Fig. 1). Females and males did not show signifi-
cant differences in developmental time (t =
-1.1652, df= 1, P > 0.05) (females: 23.4 + 1.2 days,
males: 23.7 1.1 days).
The average adult longevity was 10.3 5.8 days.
The analysis of adult longevity for females and
males indicated significant differences, with fe-
males living longer than males (t = 5.8717, P < 0.05)
(females: 12.2 5.6 days, males: 6.2 3.7 days).
The overall sex ratio ofZ. delicate in the labo-
ratory was 1: 2.1 (males/females). In Fl progenies
it was 1: 3.2 and in F2 progenies 1: 1.9, respec-
25
& 20
5 15
10
5
0o
m illi..
19 20 21 22 23 24 25 26 27 28 29
Duration of development (days)
Fig. 1. Development time from egg to adult of Za-
gella delicate in the laboratory at room temperature.
tively. In field-collected egg masses, the sex ratio
was female biased, 1: 3.8 (27 males/103 females).
Tests of Hyperparasitism
No adults ofZ. delicate emerged from eggs pre-
viously exposed to G. tuberculifemur. Out of 100
exposed eggs, only 71 adults ofG. tuberculifemur,
and 2 nymphs of T rubromarginata emerged. The
remaining eggs yielded neither host nymphs nor
parasitoids. Twenty seven adults of Z. delicate
emerged from the 39 eggs used as control.
Field Studies
In the sample collected from Johnson grass in
Tafi Viejo, 724 (71.1%) T rubromarginata eggs
were found to be attacked by a complex of egg par-
asitoids composed of several species of Tricho-
grammatidae and Mymaridae. Zagella delicate
emerged from 626 eggs, or 86.5% of all parasitized
eggs. Seasonal sampling carried out in San Miguel
de Tucuman showed that Z. delicate occurred
from September to March, with maximum abun-
dance at the beginning of the season in October,
where 57.2% out of the 1568 sampled eggs were
parasitized (Table 2). From the 41 sites sampled in
Argentina with sentinel eggs on citrus, Z. delicate
adults were obtained in Santa Clara, Jujuy Prov-
ince, and Tafi Viejo, Tucuman Province. In Santa
Clara, 155 (61.7%) out of 251 exposed eggs were
attacked by the parasitoid complex mentioned
above. Sixty of these were parasitized by tri-
chogrammatid species, including 38 (15.1%) by
Z. delicate. In Tafi Viejo, 1,549 (47.0%) parasitoids
emerged from 3,299 eggs exposed, but only 24
(1.5%) adults ofZ. delicate emerged.
Distribution
In addition to Chacras de Coria in Mendoza
Province (De Santis 1970), the locality where the
type specimen was collected, Z. delicate also was
collected in Tafi Viejo, Yerba Buena, and San
Miguel de Tucuman in Tucuman Province as well
as in Santa Clara, Jujuy Province, all in Argentina.
Voucher specimens were deposited in Fundaci6n e
Institute Miguel Lillo, San Miguel de Tucuman
(IMLA), and Entomology Research Museum, Uni-
versity of California at Riverside (UCRC).
December 2004
Logarzo et al.: Egg Parasitoid of Sharpshooter
TABLE 2. OCCURRENCE OF ZAGELLA DELICATE, OTHER TRICHOGRAMMATIDAE, AND GONATOCERUS SPP. (MYMARIDAE)
IN EGGS OF TAPAJOSA RUBROMARGINATA ON JOHNSON GRASS AT SAN MIGUEL DE TUCUMAN, ARGENTINA.
Sampled eggs Z. delicate Other Trichogrammatidae1 Gonatocerus spp.
September 2722 0.82 0.0 4.4
October 1568 57.2 1.5 0.4
November 248 16.5 4.8 2.4
January 370 19.2 27.6 4.3
February 559 21.1 17.7 7.2
March 611 9.3 17.8 21.9
April 175 0.0 3.4 19.4
May3 0 0.0 0.0 0.0
Includes at least 3 different species in 3 genera: Oligosita (1), Paracentrobia (1 or 2), and Zagella (1).
Percent of sampled eggs.
'No eggs were found in the field.
DISCUSSION
Biological characteristics ofZ. delicate, a pri-
mary egg parasitoid of T rubromarginata, have
allowed us to estimate its overall potential as a
biological control agent of H. coagulata. This is
the first study that provides biological informa-
tion on a species belonging to the genus Zagella.
Zagella delicate was successfully reared under
laboratory conditions for two generations before
the colony was discontinued. The low percentage
of emergence (35.4%) of Z. delicate adults in the
laboratory was due to the damage (rotting or
drying of the leaves and the host eggs) during
the long preimaginal period of this parasitoid,
thus complicating colony management. Most of
the unemerged host eggs were parasitized, and
nymphs of T rubromarginata emerged from
13.3% of the exposed eggs. In F2 progenies, the
percentage of adult emergence was higher than
in Fl progenies because humidity control in the
Petri dishes was improved. Females showed a
strong preference for parasitizing host eggs on
monocotyledon plants such as corn, Johnson
grass, and sugar cane, whereas they hardly par-
asitized eggs laid on citrus. The low preference of
Z. delicate to attack eggs laid on citrus plants is
not a desirable characteristic for the use of this
natural enemy in the control of glassy-winged
sharpshooter. Regulation strategy for H. coagu-
lata in California calls for its control in the citrus
orchards. Also, the results of this survey with
sentinel eggs of T rubromarginata on citrus
plants should be considered with caution be-
cause absence of Z. delicate in most of the sam-
ples does not necessarily mean that this species
is absent in the sampled area. For example, in
January 2001 in Tafi Viejo, 61.6% of T rubro-
marginata eggs on Johnson grass in the field
were parasitized, whereas at the same time and
1 km from that site, only 4% out of 99 sentinel
eggs of the same host on citrus leaves were par-
asitized by Z. delicate (E. G. V. unpubl. data).
Sex ratios observed in the field and under lab-
oratory conditions were similar, being female bi-
ased. The sex ratio in another trichrogrammatid
species studied in Argentina, Paracentrobia sp.
(misidentified as P. subflava (Girault) by L. De
Santis), showed a predominance of females over
males (Virla 1999).
Females of Z. delicate were able to parasitize
more than 60% of the eggs in the laboratory, and
about 61.6% in the field, suggesting great potential
of this species as a biological control agent in some
crops. The importance of synchronization of the oc-
currence (in time and space) of a parasitoid with its
target host is often emphasized in biological con-
trol. Zagella delicate occurs in the field from Sep-
tember to April, during the same period as its host,
T rubromarginata, and has two generations for ev-
ery generation of the host. Most importantly, it is
the predominant natural enemy of T rubromargin-
ata early in the season (Fig. 2). Overall, field and
laboratory data suggest that Z. delicate could be a
prospective biological control agent against other,
exotic,proconiine sharpshooters including H. coag-
ulata. However, the efficiency ofZ. delicate appears
to be restricted to habitats dominated by grasses.
- agriat delicar
- -Other 'r:ichognimawlida
..... M'maridac
.st'm {xTI Mlt mrs JAI- irI Wm -t *rFI M .J
Fig. 2. Field occurrence of Zagella delicate and other
egg parasitoids of Tapajosa rubromarginata on Johnson
grass at San Miguel de Tucuman, Tucuman Province.
ACKNOWLEDGMENTS
The authors thank Gabriela Murua (PROIMI) for
providing climatological data, Ver6nica Manrique
(USDA-ARS South American Biological Control Labo-
ratory), Laura Varone (USDA-ARS South American Bi-
ological Control Laboratory), and Erica Luft (PROIMI)
for technical assistance. Patricio Fidalgo (CONICET-
CRILAR, Anillaco, La Rioja, Argentina). The authors
thank two anonymous reviewers for detailed comments
that helped improve the manuscript.
REFERENCES CITED
BLUA, M. J., P. A. PHILLIPS, AND R. A. REDAK. 1999. A
new sharpshooter threatens both crops and orna-
mentals. California Agriculture 53: 22-25.
DE SANTIS, L. 1957. Descripci6n de nuevos g6neros y es-
pecies de calcidoideos argentinos, II (Hymenoptera).
Universidad Nacional de La Plata, Facultad de Cien-
cias Naturales y Museo, Notas del Museo 19: 129-
144.
DE SANTIS, L. 1970. Tricogramatidos neotropicales,
nuevos y conocidos (Hymenoptera). Universidad Na-
cional de La Plata, Facultad de Ciencias Naturales y
Museo, Revista del Museo de La Plata (n.s.) 11: 1-19.
DE SANTIS, L. 1997. Afelinidos y tricogramatidos de la
colecci6n del Dr. Alejandro A. Ogloblin (Insecta-
Hymenoptera). 2a Comunicaci6n. Academia Nacio-
nal de Agronomia y Veterinaria 51: 7-17.
December 2004
JONES, W. A. 2001. Classical biological control of the
glassy-winged sharpshooter, pp. 50-51 In Proceed-
ings of the Pierce's Disease Research Symposium,
December 5-7, 2001, Coronado Island Marriott Re-
sort, San Diego, California. California Department
of Food and Agriculture, Copeland Printing, Sacra-
mento, CA. 141 pp.
NOYES, J. S. 1994. The reliability of published host-par-
asitoids records: a taxonomic's review. Norwegian J.
Agricultural Sciences 16: 59-69.
TRIAPITSYN, S. V. 2003. Taxonomic notes on the genera
and species ofTrichogrammatidae (Hymenoptera)-
egg parasitoids of the proconiine sharpshooters (Ho-
moptera: Cicadellidae: Proconiini) in southeastern
USA. Trans. Am. Entomol. Soc. 129: 245-265.
TRIAPITSYN, S. V., MIZELL, R. F. III, BOSSART, J. L., AND
C. E. CARLTON. 1998. Egg parasitoids of Homalo-
disca coagulata (Homoptera: Cicadellidae). Florida
Entomol. 81: 241-243.
TRIAPITSYN, S. V., AND P. A. PHILLIPS. 2000. First record
of Gonatocerus triguttatus (Hymenoptera: Myma-
ridae) from eggs of Homalodisca coagulata (Ho-
moptera: Cicadellidae) with notes on the distribution
of the host. Florida Entomologist 83: 200-203.
VIRLA, E. G. 1999. Aportes preliminares acerca de la
bionomia de Paracentrobia subflava (Hym.: Tricho-
grammatidae), parasitoide de Hom6pteros Cicade-
loideos Argentinos. Revista de la Sociedad
Entomol6gica Argentina 58: 17-22.
YOUNG, D. A. 1968. Taxonomic study of the Cicadelli-
nae. 1. Proconiini. U.S. Nat. Mus. Bull. 261. 287 pp.
Florida Entomologist 87(4)
Garcia & Corseuil: Parasitoids Associated with Fruit Flies in Brazil
NATIVE HYMENOPTERAN PARASITOIDS ASSOCIATED WITH FRUIT FLIES
(DIPTERA: TEPHRITIDAE) IN SANTA CATARINA STATE, BRAZIL
FLAVIO R. M. GARCIA1 AND ELIO CORSEUIL2
'Unochapec6, CCAA, Laborat6rio de Entomologia, Caixa Postal 747, CEP. 89809-000, Chapec6, SC, Brazil
E-mail: flaviog@unochapeco.edu.br
2PUCRS, Faculdade de Bioci6ncias, PPG em Bioci6ncias, Laborat6rio de Entomologia
Av. Ipiranga, 6681, Pr6dio 12c, CEP. 90619-900, Porto Alegre, RS, Brazil
E-mail: corseuil@pucrs.br
ABSTRACT
A survey oftephritid fruit fly species and their parasitoids was conducted in the western por-
tion of Santa Catarina state, Brazil. A total of 9,197 fruits belonging to 46 plant species in
24 families was collected. The parasitoids recovered were Aganaspis pelleranoi (Brethes),
Lopheucoila anastrephae (Rhower) (Figitidae), Doryctobracon areolatus (Szepligeti), Doryc-
tobracon brasiliensis (Szepligeti), Odontosema anastrephae Borgmeier, Opius bellus Gahan,
Opius sp., Utetes (Bracanastrepha) anastrephae (Viereck) (Braconidae), and Trichopria
anastrephae Lima (Diapriidae). Agasnaspis pelleranoi is the most frequent parasitoid spe-
cies found in the west region of Santa Catarina. We recorded the first occurrence ofL. anas-
trephae in Santa Catarina. Parasitism ranged from 1.2 to 46.9%.
Key Words: Insecta, Anastrepha, Braconidae, Figitidae, Diapriidae, parasitoids, fruit flies.
RESUME
Se realize un studio del reconocimiento de las species tefritidas de las moscas de las frutas
y sus parasitoides en el area occidental del estado de Santa Catarina, Brasil. Un total de
9,197 frutas que pertenecen a 46 species de plants fueron recolectadas. Los parasitoides
recolectados fueron: Aganaspis pelleranoi (Brethes), Lopheucoila anastrephae (Rhower) (Fi-
gitidae), Doryctobracon areolatus (Szepligeti); Doryctobracon brasiliensis (Szepligeti), Odon-
tosema anastrephae Borgmeier, Opius bellus Gahan, Opius sp., Utetes (Bracanastrepha)
anastrephae (Viereck)(Braconidae) y Trichopria anastrephae Lima (Diapriidae).Agasnaspis
pelleranoi es el parasitoide encontrado mas frecuentemente en la region oeste de Santa Ca-
tarina. El primer registro de la ocurrencia de L. anastrephae en Santa Catarina fue obtenido.
El rango de parasitismo fue de 1.2 a 46.9%.
Surveys oftephritid fruit flies and their parasi-
toids are a first step to a better understand of the
ecology of these economically important taxa (Zuc-
chi 2000a). The major natural enemies of the fruit
flies in Brazil belong to the families Braconidae,
Figitidae, and Pteromalidae (Hymenoptera). Pter-
omalidae, mainly Pachycrepoideus vindemiae
Rondani, are sporadically collected and specimens
of Figitidae are collected in small numbers
throughout the country. Typically the most fre-
quent parasitoids collected in Brazil are members
of the Braconidae family (Canal & Zucchi 2000).
In Santa Catarina state, Nora et al. (2000) pre-
viously found, in order of abundance, Dorycto-
bracon areolatus (Szepligeti), Doryctobracon bra-
siliensis (Szepligeti), Opius bellus Gahan, and
Opius tomoplagiae Lima. In addition, they ob-
tained unidentified Diapriidae, Eulophidae, Fig-
itidae, and Pteromalidae. Guimaraes et al. (2000)
added to this list the figitidsAganaspis pelleranoi
(Brethes) and Odontosema anastrephae Borg-
meier, and Leonel, Jr. et al. (1995) and Canal &
Zucchi (2000) obtained braconids Microcrasis lon-
chaea (Lima) and Utetes (Bracanastrepha) anas-
trephae (Viereck).
The present paper describes the parasitoids of
fruit flies from the western portion of Santa Cata-
rina state, an area which is a growing producer of
citrus in the state (Koller et al. 1999) and which
has not been surveyed thoroughly in the past. We
collected 9,197 mature fruit from trees or on the
soil comprising 46 species belonging to 25 fami-
lies in the six towns Anchieta (2653'S and
5333'W), Chapec6 (2706'S and 5316'W), Cunha
Pora (2607'S and 5316'W), Palmitos (2706'S
and 5316'W), Sao Carlos (2707'S and 5300'W),
and Xanxer6 (2687'S and 5240'W), Santa Cata-
rina. Each fruit was weighed and put in a plastic
container with about seven centimeters of steril-
ized sand, and covered with a net. The containers
were kept in the entomology laboratory of the Ag-
ricultural and Environmental Science Center at
the Universidade Comunitaria Regional de
Chapec6 at 25 3C, 70 + 10% and a 12-h photo-
period. After five days, the sand with pupae was
transferred to lab Petri dishes containing filter
TABLE 1. PARASITOID SPECIES (HYMENOPTERA) FROM FRUIT FLIES COLLECTED IN SIX DIFFERENT LOCALITIES FROM THE WEST OF SANTA CATARINA, BRAZIL, DURING 1998-2000.
Hymenoptera parasitoid species and relative frequency (%)
Vegetal specie S' N2 PaT3 Braconidae Figitidae and Diapriidae %P4 TIP5
Fabaceae
Inga sellowiana 5 246
Myrtaceae
Psidium cattleianum
Eugenia involucrata
Psidium guajava
Feijoa sellowiana
Myrcianthes pungens
Campomanesia xanthocarpa
Britoa guazumaefolia
Eugenia pyriformis
Rosaceae
Prunus domestic
Prunus auium
11 635 42 D. areolatus (19.0)
D. brasiliensis (33.3)
Opius sp. (9.7)
U anastrephae (19.0)
3 446 46 D. areolatus (69.6)
Opius bellus (8.7)
Opius sp. (4.3)
U anastrephae (17.4)
17 190 147 D. brasiliensis (2.0)
Opius bellus (1.1)
Opius sp. (2.0)
2 80 58 D. areolatus (48.3)
D. brasiliensis (48.3)
Opius bellus (3.4)
2 52 6 D. brasiliensis (100.0)
4 702 4 D. brasiliensis (100.0)
6 255 48 D. areolatus (16.7)
D. brasiliensis (37.5)
5 264 4
L. anastrephae (100.0)
A. pelleranoi (19.0)
4.1 4.7
1.9 2.1
A. pelleranoi (49.0)
T anastrephae (45.9)
A. pelleranoi (3.5)
0. anastrephae (8.3)
A. pelleranoi (100.0)
3.0 4.0
109 26 D. brasiliensis (23.1)
U anastrephae (76.1)
18 8 D. areolatus (75.0)
U anastrephae (25.0)
'S-sample, 'n-number of fruit, 'PaT-parasitoids total, '%P-Parasitism percentage, 'TIP-total index Parasitism.
TABLE 1. (CONTINUED) PARASITOID SPECIES (HYMENOPTERA) FROM FRUIT FLIES COLLECTED IN SIX DIFFERENT LOCALITIES FROM THE WEST OF SANTA CATARINA, BRAZIL, DURING
1998-2000.
Hymenoptera parasitoid species and relative frequency (%)
Vegetal specie S' N2 PaT3 Braconidae Figitidae and Diapriidae %P4 TIP5
Eriobotryajaponica 9 1166 48 D. areolatus (8.3) 4.3 5.9
D. brasiliensis (41.6)
Opius bellus (4.3)
Opius sp. (4.2)
U anastrephae (41.6)
Pyrus communis 2 62 4 A. pelleranoi (100.0) 14.8 33.3
Prunes persica 16 562 18 D. brasiliensis (22.2) A. pelleranoi (66.7) 1.2 1.6
0. anatrephae (11.1)
Total 89 4787 461
S-sample, nn-number of fruit, 'PaT-parasitoids total, '%P-Parasitism percentage, 'TIP-total index Parasitism.
Florida Entomologist 87(4)
paper dampened with distilled water. Flies and
parasitoids were counted after seven days.
The relationship between a fly species and its
parasitoids was determined only when a single
species of fly was held in an emergence container
(Canal et al. 1994).
The total index of parasitism (TIP) was calcu-
lated as the number of parasitoids emerged x 100/
number of flies emerged + number of parasitoids
emerged. The relative frequency of fly species and
parasitoids (RF) was defined as number of sam-
ples of a given species collected x 100/total num-
ber of collected species according to Matrangolo et
al. (1998), and the parasitism percentage was cal-
culated as %P = total parasitism x 100/total pu-
pae, which was modified from Silveira Neto et al.
(1997).
Species of Anastrepha were identified with
Steyskal's key (1997) and Zucchi's key (2000b),
which includes only Brazil species. The Bra-
conidae were identified according to the key of Ca-
nal & Zucchi (2000). The flies and parasitoids
belonging to other families were sent to Prof Dr.
Manoel Ara6cio Uch6a Fernandes, Biologist Jorge
Anderson Guimaraes, Dr. Allen Norrbom, and
Prof Dr. Roberto Antonio Zucchi for identification.
Of the 46 fruit species collected, 35 were in-
fested by fruit flies, but only 14 of these fruit spe-
cies contained parasitoids (Table 1). A total of 682
samples of parasitoids belonging to nine species
and three families were obtained, as follows: D.
areolatus, D. brasiliensis, 0. bellus, Opius sp., U.
anastrephae (Braconidae), A. pelleranoi, 0. anas-
trephae (Figitidae), and Trichopria anastrephae
Lima (Diapriidae).
Of the 461 hymenopterans associated with a
particular fruit fly,A. pelleranoi was the most com-
mon and represented 25.6% of the total, followed
byD. brasiliensis (21.1%) andD. areolatus (18.6%).
These figures differ from those obtained by Salles
(1996) for the state of Rio Grande do Sul and Le-
onel Jr. (1995, 1996) for the state of Sao Paulo.
However, the relative abundance of parasitoids col-
lected in the valley of Rio do Peixe, state of Santa
Catarina by Nora et al. (2000) are similar to ours.
All the parasitoids developed inA. fraterculus,
except for L. anastrephae, which was associated
only with Neosilba sp. Also a lonchaeid was ob-
tained in the samples (Table 2). D. areolatus and
U. anastrephae also parasitized Neosilba sp. De
Santis (1980) catalogued eleven species of Trichop-
ria for Brazil, and believed that only T anas-
trephae parasitized the genusAnastrepha. Aguiar-
Menezes et al. (2001) previously reportedA. frater-
culus as the host of T anastrephae. L. anastrephae
has been recorded previously only in the southern
part of the country, in the state of Sao Paulo and
central Mato Grosso do Sul (De Santis 1980).
Parasitoids attacked fruit flies in 14 host-plant
species, one Fabaceae, five Rosaceae and eight
Myrtaceae (Table 1). Myrtaceae contained 79.6%
TABLE 2. ASSOCIATION BETWEEN COLLECTED PARASI-
TOIDS AND FRUIT FLIES IN SIX DIFFERENT LO-
CALITIES IN THE WEST OF SANTA CATARINA,
BRAZIL, DURING 1998-2000.
Fruit flies
Parasitoids A. fraterculus Neosilba sp.
Braconidae
D. areolatus X X
D. brasiliensis X
0. bellus X
Opius sp. X
U anastrephae X X
Diapriidae
T anastrephae X
Figitidae
A. pelleranoi X
L. anastrephae X
0. anastrephae X
of the parasitoids and Rosaceae had 21.2%. The
total parasitism index and the parasitism per-
centage were greatest in Eugenia involucrata, fol-
lowed by Prunus avium and Myrcianthes
pungens. These indexes are higher than those ob-
tained by Leonel, Jr. et al (1996), Salles (1996)
and Matrangolo et al. (1998), but are similar to
those of Guimaraes et al. (1999).
The high percent parasitism in E. involucatra
was previously found by Salles (1996) in the Rio
Grande do Sul state, and may be due, as he sug-
gested, to the thin peel and small size of the fruit.
Guimaraes et al. (2000) previously observed affin-
ity ofA. pelleranoi to the Myrtaceae fruit, which
we confirmed. Sivinski (1991), Sivinski et al.
(1997, 2000) and Hickel (2002) found that bra-
conid parasitism was negatively correlated to
fruit pulp thickness, and we showed that weight
data correlated to parasitism.
ACKNOWLEDGMENTS
We thank the following researchers for taxonomic
identification: Professor Dr. Manoel Ara6cio Uch6a
Fernandes-UFMS (Lonchaeidae), Biologist Jorge
Anderson Guimaraes-ESALQ/USP (Figitidae), Dr.
Allen Norrbom-Systematic Entomology Laboratory,
and Professor Dr. Roberto Antonio Zucchi-ESALQ/
USP for some Anastrepha species confirmation. We
thank Professors Lucia S. Verona and Rosiane B. Denar-
din from the Botanic Department of UNOCHAPECO for
identification of the plants.
REFERENCES CITED
AGUIAR-MENEZES, E., E. B. MENEZES, P. S. SILVA, A. C.
BITTAR, AND P. C. CASSINO. Native hymenopteran
parasitoids associated with Anastrepha spp.
December 2004
Garcia & Corseuil: Parasitoids Associated with Fruit Flies in Brazil
(Diptera: Tephritidae) in Seropedica city, Rio de Ja-
neiro Brazil. Florida Entomol. 84: 706-711.
CANAL, N. A., AND R. A. ZUCCHI. 2000. Parasit6ides Bra-
conidae, pp. 119-126 In A. Malavasi and R. A. Zucchi
[eds.], Moscas-das-frutas de Importancia Economica
no Brasil, Conhecimento Basico e Aplicado. Holos,
Ribeirao Preto.
CANAL, N. A., R. A. ZUCCHI, N. M. SILVA, AND F. L. LE-
ONEL, JR. 1994. Reconocimiento de las species de
parasitoides (Hym.: Braconidae) de moscas de las
frutas (Dip.: Tephritidae) en dos municipios del Es-
tado de Amazonas, Brasil. Bol. Mus. Entomol. Uni-
versidad del Valle 2: 1-17
DE SANTIS, L. 1980. Catalogo de los Himenopteros
Brasilenos de la Serie Parasitica Incluyendo Bethy-
loidea. UFPR, Curitiba. 395 pp.
GUIMARAES, J. A., N. B. DIAZ, AND R. A. ZUCCHI. 2000.
Parasit6ides-Figitidae (Eucoilinae), pp. 127-134 In
A. Malavasi and R.A. Zucchi [eds.], Moscas-das-fru-
tas de Importancia Economica no Brasil, Conheci-
mento Basico e Aplicado. Holos, Ribeirao Preto.
GUIMARAES, J. A., R. A. ZUCCHI, N. B. DIAZ, M. F. DE
SOUZA FILHO, AND M. A. UCHOA F. 1999 Esp6cies de
Eucoilinae (Hymenoptera: Cynipoidea: Figitidae)
parasit6ides de larvas frugivoras (Diptera: Tephriti-
dae e Lonchaeidae) no Brasil. An. Soc. Entomol. Bra-
sil 28: 263-273.
HICKEL, E. R. 2002. Espessura da polpa como condicio-
nante do parasitismo de moscas-das-frutas (Diptera:
Tephritidae) por Hymenoptera: Braconidae. Ci6ncia
Rural 32:1005-1009
KOLLER, O. L., E. SOPRANO, L. A. F. VERONA, L. A. CHI-
ARADIA, A. BRANCHER, AND E. N. DE OLIVEIRA. 1999.
Citros, pp. 49-52 In EPAGRI, Recomendacoes de
Cultivares para o Estado de Santa Catarina 1999/
2000. EPAGRI, Florian6polis.
LEONEL JR., F. L., R. A. ZUCCHI, AND N. A. D. CANAL.
1996. Parasitismo de moscas-das-frutas por Bra-
conidae (Hymenoptera) em duas localidades do Estado
de Sao Paulo. An. Soc. Entomol. Brasil 25: 199-206.
LEONEL JR., F.L., R. A. ZUCCHI, AND R. A. WHARTON.
1995. Distribution and tephritid hosts (Diptera) of
braconid parasitoids (Hymenoptera) in Brazil. Intl.
J. Pest Management 41: 208-213
MATRANGOLO, W. J. R., A. S. NASCIMENTO, R. S. CAR-
VALHO, E. D. MELO, AND M. DE JESUS. 1998. Para-
sit6ides de moscas-das-frutas (Diptera: Tephritidae)
associados a frutiferas tropicais. An. Soc. Entomol.
Brasil 27: 593-603.
NORA, I., E. R. HICKEL, AND H. F. PRANDO. 2000. Mos-
cas-das-frutas nos estados brasileiros: Santa Catar-
ina, pp. 271-275 In A. Malavasi and R. A. Zucchi
[eds.], Moscas-das-frutas de Importancia Economica
no Brasil, Conhecimento Basico e Aplicado. Holos,
Ribeirao Preto.
SALLES, L. A. B. 1996. Parasitismo de Anastrepha
fraterculus (Wied.) (Diptera: Tephritidae) por Hy-
menoptera, na regiao de Pelotas, RS. Pesq. Agrop.
Brasileira 31: 769-774.
SILVEIRA NETO, S., NAKANO, O., BARBIN, D., AND N. A.
VILLA NOVA. 1976. Manual de Ecologia dos Insetos.
Agronomica Ceres, Piracicaba, 419 pp.
SIVINSKI, J. 1991. The influence of host fruit morphol-
ogy on parasitism rates in the Caribbean fruit fly
(Anastrepha suspense (Loew)). Entomophaga 36:
447-455
SIVINSKI, J. M., M. ALUJA, AND M. LOPEZ. 1997. Spatial
and temporal distribution of Mexican Anastrepha
species (Diptera: Tephritidae), within the canopies
of fruit trees. Ann. Entomol. Soc. Amer. 90: 604-618.
SIVINSKI, J. M., J. PINERO, AND M. ALUJA. 2000. The dis-
tributions of parasitoids (Hymenoptera) of Anas-
trepha fruit flies (Diptera: Tephritidae) along an
altitudinal gradient in Veracruz, Mexico. Biological
control 18: 258-269.
STEYSKAL, G. C. 1977. Pictorial Key to Species of the
Genus Anastrepha (Diptera: Tephritidae). The Ento-
mol. Soc. of Washington, Washington, D.C. 35 pp.
ZUCCHI, R. A. 2000a. Esp6cies de Anastrepha, sinon-
imias, plants hospedeiras e parasit6ides, pp. 41-48
In A. Malavasi and R.A. Zucchi [eds.], Moscas-das-
frutas de Importancia Economica no Brasil, Conhec-
imento Basico e Aplicado. Holos, Ribeirao Preto.
ZUCCHI, R. A. 2000b. Taxonomia, pp. 13-24 In A. Mala-
vasi and R.A. Zucchi [eds.], Moscas-das-frutas de
Importancia Economica no Brasil, Conhecimento
Basico e Aplicado. Holos, Ribeirao Preto.
Florida Entomologist 87(4)
December 2004
MONITORING AND MANAGEMENT OF RED
IMPORTED FIRE ANTS IN A TROPICAL FISH FARM
DAVID H. OIl, CRAIG A. WATSON2 AND DAVID F. WILLIAMS'
1USDA-ARS Center for Medical, Agricultural, and Veterinary Entomology
1600 SW 23rd Drive, Gainesville, FL 32608
2Tropical Aquaculture Laboratory, University of Florida, 1408 24th Street S.E., Ruskin, FL 33570
ABSTRACT
Tropical fish farms provide a prime habitat for the red imported fire ant, Solenopsis invicta
Buren, which is an invasive, stinging ant that has spread throughout the southern United
States. Stings can be a serious health hazard to hypersensitive individuals, and the presence
of large populations can interfere with operational activities. The most efficient method of
controlling fire ants is the application of ant bait. However, most fish farmers are reluctant
to use baits or other chemical methods of control because of the unknown risk to fish. Baited
stations in combination with maps generated by geographical information system software
were used to estimate locations of fire ant nests. Maps of estimated fire ant locations encom-
passed from 10 to 100% of actual fire ant nests surveyed when there was a minimum of a sin-
gle fire ant within a station. Low percentages of overlap between mapped areas and fire ant
nests were associated with low nest densities and when higher ant count thresholds were
used to indicate positive stations. Ant bait containing the insect growth regulator methop-
rene was broadcast between ponds, with some unavoidable bait entry into ponds. Fire ant
nest densities declined 57% within 4 months. In contrast, nest densities increased 86% in
untreated areas. During the summer, fire ant populations declined an average of 68% and in-
creased 110% for treated and untreated areas, respectively. Fire ant nest densities and pop-
ulations began to increase by December in both treated and untreated areas. No obvious fish
mortality related to the ant baiting was noted during the study.
Key Words: Solenopsis invicta, aquaculture, GIS application, pest detection, pest control,
methoprene, bait, detection.
RESUME
Las fincas de peces tr6picales proven un perfect habitat para la hormiga de fuego roja im-
portada, Solenopsis invicta Buren, la cual es una especie invasora que pica, la cual se ha es-
pacido por todo el sur de los Estados Unidos. Las picaduras pueden ser de gran peligro para
la salud de individuos hipersensibles, y la presencia de altas poblaciones pueden interferir
con actividades de trabajo. El m6todo mas eficaz para controlar la hormiga de fuego roja im-
portada es la aplicaci6n de cebos para hormigas. Sin embargo, la mayoria de los productores
de peces son renuentes para usar los cebos u otros m6todos quimicos de control por el riesgo
desconocido para los peces. Las estaciones de cebo en combinaci6n con mapas generados con
programs de computadores de sistemas de informaci6n geograficos fueron usados para esti-
mar la localizaci6n de los nidos de la hormiga de fuego roja importada. Los mapas de las lo-
calidades de la hormiga de fuego roja importada estimadas abarcaron 10 a 100% de los nidos
existentes en el muestreo cuando fue un minimo de una sola hormiga de fuego roja dentro de
una estaci6n. Los porcentajes bajos de traslapes entire las areas en los mapas y los nidos de
la hormiga de fuego roja fueron asociados con las densidades bajas de nidos y cuando se usa-
ron umbrales de conteo altos de las hormigas para indicar estaciones positivas. El cebo para
hormigas que cotiene el regulador de crecimiento de los insects metoprina fue aplicado entire
las lagunas, con alguna entrada inevitable de cebo en la laguna. Las densidades de los nidos
de la hormiga de fuego roja bajaron 57% dentro de 4 meses. En contrast, las densidades de
los nidos aumentaron 86% en areas no tratadas. Durante el verano, las poblaciones de la hor-
miga de fuego roja bajaron un promedio de 68% en areas tratadas y aumentaron 110% en
areas no tratadas. La densidad y la poblaci6n de la hormiga de fuego roja empezaron a au-
mentar por diciembre en ambas areas tratadas y no tratadas. No se observe mortalidad al-
guna sobre los peces relacionada con el cebo para las hormigas durante este studio.
The production of tropical fish accounted for (Florida Agricultural Statistics Service 2002).
about 43% of the value of Florida aquaculture Most farms are located in areas with sandy loam
farm gate sales in 2001 from 160 active growers soils with a relatively high water table, allowing
Oi et al.: Fire Ants on Tropical Fish Farms
for relatively inexpensive pond construction.
Ponds are small (0.04 ha), and a typical 0.4 ha (1
acre) of "farm" will have up to eight ponds. Be-
cause of the soil type, open water, and high water
table, these farms provide an ideal habitat for the
red imported fire ant, Solenopsis invicta Buren.
This ant is an invasive, exotic species from South
America, which has spread throughout the south-
ern United States since its inadvertent introduc-
tion in the 1930s (Callcott & Collins 1996). It has
a painful sting and deaths have been reported in
hypersensitive individuals (deShazo et al. 1990,
1999). Fish farmers typically work barefoot and
in short pants along the pond banks, and because
a fire ant colony can contain over 200,000 individ-
uals (Markin et al. 1973; Tschinkel 1988; Macom
& Porter 1996), the probability of being stung is
very high. In addition, there are anecdotal reports
of mass fish mortality due to the ingestion of fire
ants, but studies have failed to confirm this phe-
nomenon (Ferguson 1962; Crance 1965). On most
tropical fish farms, fire ants are not controlled be-
cause of the fear that insecticides will detrimen-
tally impact fish being raised in ponds (C. A.
Watson, pers. obs.).
Effective control options for fire ants in areas
where there is a high potential for people to be
stung generally involve the use of chemical insec-
ticides. The application of fire ant baits is one of
the most efficient methods of control (Lofgren &
Weidhass 1972). Baits utilize the natural behav-
ior of ants to forage for food and then share it
with colony members. In this manner, insecti-
cides incorporated into baits are spread through-
out the colony. Alternative methods of control are
to treat individual colonies with an insecticide
that must contact the majority of the members of
the colony including the queen, or use of non-
chemical alternatives such as steam or hot water
(Drees et al. 1998; Tschinkel & Howard 1980).
However, treating individual colonies is labori-
ous and time consuming in contrast to broadcast-
ing fire ant bait (Barr et al. 1999). Inherent to the
integrated pest management approach of con-
trolling insect pests is determining the location
and population of the pest (Pedigo 1996). In this
study our objectives were to assess the feasibility
of(1) using baited monitoring stations to indicate
locations of fire ant nests, and (2) broadcasting a
fire ant bait to control fire ants on a tropical fish
farming facility.
MATERIALS AND METHODS
The study site was located at the University of
Florida Tropical Aquaculture Laboratory in
Ruskin, FL, which contains 51 ponds on approxi-
mately 2.63 ha (6.5 acres). Ponds contained
swordtails, Xiphophorus helleri L. The banks and
paths between ponds, as well as adjacent areas,
were visually surveyed for red imported fire ant
nests. Each nest was partially opened with a
shovel to estimate the amount of adult ants and
to search for immature worker caste ants, or
brood. The presence of brood indicates that the
colony is healthy. In contrast, a colony with no
brood or a preponderance of immature reproduc-
tive caste ants indicates that the colony is abnor-
mal and declining in vigor. To quantify fire ant
populations, a standardized rating system, devel-
oped by the U.S. Department of Agriculture, pro-
vided a population index that incorporated the
estimates of the number of adult ants and the
presence or absence of worker caste brood in each
nest (Lofgren & Williams 1982). The population
indices of all nests in an area were then summed
to provide a population index for the area. In ad-
dition, all nests containing live ants were counted
to provide a density of active nests per area.
Baited monitoring stations also were utilized
to determine if they could indicate areas contain-
ing active fire ant nests. Stations consisted of two
plastic petri dishes (50 mm diameter x 9 mm
depth, Gelman Sciences, Ann Arbor, MI). One
dish was lined with a cotton cosmetic pad that
was moistened with approximately 2 ml of su-
crose-based ant attractant (Vail et al. 1999). The
other dish was lined with filter paper, and held a
toothpick dipped in peanut butter, which served
as an oil-based attractant.
Baited monitoring stations were placed at the
corners of each pond, and an additional station
was placed in the middle of each long side for rect-
angular ponds or on all sides for square ponds.
This resulted in stations being placed at intervals
of 8.2 to 9.1 m (27-30 ft) along the upper banks of
each pond. Distances between ponds were 3.0 to
6.1 m (10-20 ft). In an area where ponds were not
present (control area on the west end), stations
were placed at intervals of 7.6 m (25 ft) (Fig. 1).
We utilized a total of 230 stations on each sam-
pling date. Stations were set when air tempera-
tures were conducive to ant foraging, i.e., 24.4 to
32.80C (76-910F). Fire ants inside stations were
counted 30 to 45 min after placement.
Ponds, active nest locations, and monitoring
stations were mapped with a geographical posi-
tioning system (Trimble Pathfinder Pro-XL,
Sunnyvale, CA 94088) which was capable of less
than 1 m accuracy. For each sampling date, an in-
verse distance weighted (IDW) interpolator pro-
vided in ArcView GIS software (version 3.2) was
used to generate maps that displayed estimated
areas of fire ant populations based on fire ant
counts at each monitoring station. The IDW inter-
polator produced maps that were based on the as-
sumption that there would be fewer fire ants in
areas that were further away from positive sta-
tions. To determine if the maps could be used to
indicate the presence of fire ant nests, we calcu-
lated the percentages of actual nest locations that
were encompassed by these maps.
Florida Entomologist 87(4)
Fig. 1. Study site on March 13, 2000 with monitoring stations which contained fire ants (filled squares) and sta-
tions without fire ants (open squares). The lightly shaded area is the map generated from interpolating the ant
counts from stations with at least a single fire ant. Actual fire ant nest locations are designated by circles containing
a cross. Darkly shaded rectangular areas are the ponds.
The shapes of maps, or estimated areas with
fire ants, are affected by the criteria used to define
a positive station. Estimated areas with fire ants
were mapped with three thresholds for positive
stations: 1, 25, and 50 fire ants per station. We hy-
pothesized that by increasing the threshold used
to designate a positive station, the percentages of
actual nest locations that were encompassed by
these maps would decrease because higher
thresholds would result in fewer positive stations,
and subsequently smaller estimated areas with
fire ants. We also hypothesized that the percent-
ages of actual nest locations that were encom-
passed by maps would be less if fire ant population
levels were low. A two-way analysis of variance
and Tukey's HSD test (SAS Institute 2001) were
used to compare the percentages of nest locations
encompassed by maps generated from the three
thresholds used to designate positive stations, and
between control and treated areas which reflected
areas of high and low fire ant populations.
The paths around 32 ponds were treated (Fig.
1) with fire ant bait that contained 0.5% of the in-
sect growth regulator (S)-Methoprene (Extin-
guishTM Professional Fire Ant Bait, Wellmark
International, Bensenville, IL 60106). This bait
was selected because of the low toxicity of the ac-
tive ingredient to fish and zooplankton (LC5,
trout: 760 ppb, bluegill: >370 ppb, and Daphnia
360 ppb). Formulations that contain a much
higher percentage of methoprene are used for
mosquito control in aquatic ecosystems (e.g.,
8.62% methoprene in Altosid).
Fire ant bait was applied to the paths between
ponds with a seed broadcaster (model GT-77 ATV,
Herd Seeder Co., Logansport, IN 46947) mounted
on an all-terrain vehicle. Bait was applied on
March 14, 2000, at a rate of 0.454 kg (1 lb) bait
per acre including ponds, which is within the la-
bel rate of 0.454 to 0.681 kg (1 to 1.5 lbs) per acre.
However, because bait was broadcast onto paths
between ponds, the majority of the bait landed on
the paths and banks. This resulted in an applica-
tion rate of approximately 10.4 kg (23 lbs) of bait
per acre if only path surfaces are considered. Two
areas encompassing eight ponds were used as un-
treated controls and were separated from the
treated area by approximately 21.3 m (70 ft).
Fire ant populations were determined with the
population index and the bait station methods the
day before treatment (March 13, 2000) and on 15
(June 29), 22 (August 17), and 39 weeks (Decem-
ber 14) after treatment. Surveys were conducted
within a week of moderate to heavy rains to en-
sure that soil was moist and allowed nest sites to
be distinctly visible as mounds. Percent change in
fire ant population indices and mound densities
from the pretreatment surveys were compared
between the treated and control areas.
RESULTS AND DISCUSSION
As the thresholds for positive monitoring sta-
tions increased, the percentages of nests that
were encompassed by maps that estimated fire
ant populations decreased significantly (F = 27.9;
df= 2, 18;P < 0.001) (Table 1). However, when fire
ant populations were low (<15 nests/ha) in
treated areas, only 10-30% of the nest locations
were encompassed by the maps based on the most
stringent threshold of 1 ant. In contrast, when
fire ant populations were above 35 nests per ha,
79 to 100% of the nests were included in the
mapped areas based on the same threshold (Fig.
2). The higher percentages were associated with
higher numbers of positive stations (Table 1). Sta-
tions with or without ants did not always reflect
the presence or absence of nearby nests as there
were several observations of fire ant nests adja-
cent to negative stations (Fig. 1). Perhaps when
December 2004
Oi et al.: Fire Ants on Tropical Fish Farms
TABLE 1. MEAN PERCENTAGES OF S. invicta NESTS THAT WERE ENCOMPASSED BY MAPS GENERATED BY INTERPOLA-
TION OF POSITIVE MONITORING STATIONS AND THE MEAN NUMBER OF POSITIVE STATIONS (IN PARENTHESES),
BASED ON MINIMUM THRESHOLDS OF 1, 25, OR 50 FIRE ANTS PER STATION IN METHOPRENE TREATED AND UN-
TREATED CONTROL AREAS AT THE UNIV. OF FLORIDA TROPICAL AQUACULTURE LABORATORY, RUSKIN, FLOR-
IDA. MEANS WERE OVER SAMPLING DATES (n = 4).
Threshold No. of S. invicta per station
Area 1 25 50 Treated vs Control'
Treated 52.8 7.2 1.8 20.6ab
(25.5) (10.5) (5.0)
Control 92.5 43.7 16.6 51.0 b
(48.3) (27.5) (16.3)
Comparison among thresholdsc 72.7 ad 25.5 b 9.2 b
"Comparison over all sampling dates and thresholds (n =12).
bMeans followed by a different letter within a column are significantly different by analysis of variance (F = 25.7; df = 1, 18; P < 0.001).
'Comparisons among thresholds over all sampling dates and areas (n = 8).
dMeans followed by a different letter within a row are significantly different by Tukey's HSD test (P < 0.05).
fire ant nest densities are low, areas may be more
accurately mapped by interpolating positive mon-
itoring stations and nest locations that are en-
countered during the servicing of stations.
The time required to service stations during
the March sampling, including the 30-45-min in-
terval to allow ants to forage at stations, was 1.6
person-h. In comparison, a walking survey to flag
nest locations required 2.8 person-h. Thus, servic-
ing the bait stations was 43% faster than the
walking survey when fire ant populations were
relatively high. On the March sampling date
when there were 35 and 52 nests per ha (14-21
nests/acre), the single fire ant threshold for a pos-
itive station resulted in maps that encompassed
over 92% of the nests.
While fish mortality relative to the bait applica-
tion was not formerly assessed, no obvious mortal-
ity was observed during the study (C. A. Watson,
pers. obs.). The half-life of technical methoprene in
pond water is less than 2 days, and approximately
10 days in soil (EXTOXNET 2001); thus, the accu-
mulation of the active ingredient is probably min-
imal. The equipment used to apply the bait was
calibrated to the recommended label rate. How-
ever, the amount of bait actually applied per area
of dry land (i.e., excluding water surfaces) was well
above the recommended rate. This reflected a typ-
ical reality of fire ant bait application, where cali-
brating commercially available broadcast bait
applicators to the very low recommended applica-
tion rates often is difficult. This difficulty was ex-
acerbated at the fish farm because extremely slow
speeds were required to navigate the application
equipment through the numerous turns on narrow
paths around ponds. Thus, the over-application of
fire ant bait represents a problem with bait appli-
cation that will require improvement.
Fire ant nests per hectare and population indi-
ces per hectare one day before bait application
were 35 and 564 (14 and 228/acre), respectively,
I 1 0 0 -. O
n" 80
so
S20 --treat
2.
0
) 40
0 .,0, control
13-Mar 29-Jun 17-Aug 14-Dec
Sloo
--1 treat B
80 0. control
60
| 60 o.
E .0
8 40 .. .
S20
20
z
13-Mar 29-Jun 17-Aug 14-Dec
100
S-- treat C
b 80 0 .tO control
60
1-Mar 29-Jun 1-Aug 14-Dec
8 0 0
13-Mar 29-Jun 17-Aug 14-Dec
Fig. 2. Percentages of S. invicta nests encompassed by
maps generated from the interpolation of positive bait
stations based on thresholds of(A) 1, (B) 25, or (C) 50 ants
between treated and untreated control areas among sam-
pling dates. Treatment was applied on March 14, 2000.
Florida Entomologist 87(4)
for the treated area, and 52 and 629 (21 and 255/
acre) for the control. Percent reductions of active
fire ant nests per hectare in the treated area from
the initial survey were 57 and 66% at 15 and 22
weeks after treatment. In contrast, the untreated
control areas had 85 and 62% increases in the den-
sities of active fire ant nests. These represent a
167 and 207% difference between the treatment
and control for the June and August surveys. Sim-
ilarly, there were 70 and 67% reductions in popu-
lation indices in the treated area, and increases of
120 and 99% in the control for the same dates rel-
ative to pre-treatment populations. Employees at
the facility also perceived a substantial reduction
in fire ant stings in treated areas (C. A. Watson,
pers. obs.). By the December survey, fire ants had
re-infested the treated area beyond the pre-treat-
ment number of active mounds by 83% and the
pre-treatment population index by 135% (Table 2).
Methoprene is an insect growth regulator
which results in an over production of the repro-
ductive caste and little or no production of adult
worker caste ants (Cupp & O'Neal 1973;Vinson &
Robeau 1974). Without the worker caste, ant col-
onies eventually die because food foraging and
other colony maintenance functions cease. Be-
cause the methoprene concentration in the fire
ant bait does not affect adults, colonies will sur-
vive until there is sufficient natural attrition of
existing adults. This process will generally take 8
to 10 weeks (Drees & Barr 1998) and is consistent
with the reduction in fire ants obtained at 15
weeks in this study. Drees & Barr (1998) reported
reductions of over 70% at application rates of 1.68
kg per ha, thus the large amount of bait applied in
this study probably was not needed to obtain the
level of control reported here. The estimated cost
of the bait needed for application to the paths in
the treatment area in this study was less than $3,
based on a cost of $37 per ha at the maximum
label rate (1.68 kg/ha).
Low precipitation during the study limited
population surveys to irregular intervals dictated
by episodes of enough rain to moisten soil and to
permit mound building by the ants. When soil is
dry, fire ant nests may not be visible, but popula-
tions can still increase. This was evident in the
December survey where there was a large in-
crease in fire ant mounds since the August survey.
Drees & Barr (1998) reported an increase in fire
ant mound densities between 6 and 12 months in
methoprene treated field plots. We observed a
greater concentration of large colonies in treated
areas near a buffer area, which indicated that re-
infestation was probably due to migration of ma-
ture colonies from the buffer and control areas. If
all areas were treated, it would be reasonable to
expect a slower reinfestation.
Surveying and mapping active mounds have
aided in delimiting areas that require baiting and
thus reduced the amount of bait that was applied
per year in managed landscapes (Cobb & Cobb
1995). Bait stations could be used on tropical fish
farms to delineate areas where fire ants are forag-
ing and could indicate where bait treatments may
be effectively applied. Using monitoring stations
was faster for us than searching for individual
nests, and may be advantageous during dry condi-
tions or in areas with overgrown vegetation,
where nests are difficult to see. Counting fire ants
at monitoring stations required the ability to dis-
tinguish fire ants from other ants. Information on
identifying common ants is available in several
extension publications (Vail et al. 1994; Drees et
al. 2000). Generating interpolated maps based on
monitoring stations required investments in GPS
equipment and GIS software. Continuing im-
provements in this technology have made it more
affordable and usable. Our use of an IDW interpo-
lator and a 1-ant threshold provided maps that es-
timated areas which contained over 78% of the fire
ant nests when their densities were >35 nests per
ha (14 nest/acre, Fig. 2a). Evaluations of other in-
terpolators and thresholds may provide more ac-
curate maps over a wider range of nest densities.
In summary, at low fire ant population levels,
improvements were needed in estimating nest loca-
tions based on the baited monitoring stations. How-
ever, when populations were high, the methods
utilized in this study detected the presence of most
of the fire ant nests. Suppression of fire ants on a
tropical fish farm was accomplished with the broad-
TABLE 2. FIRE ANT MOUND DENSITIES AND POPULATION INDICES OF METHOPRENE TREATED AND UNTREATED CONTROL
AREAS AT THE UNIV. OF FLORIDA TROPICAL AQUACULTURE LABORATORY, RUSKIN, FLORIDA.
Mounds/ha (% change) Pop. Index"/ha (% change)
Week" Treated Control Treated Control
0 35 52 564.1 629.1
15 15 (-57.1) 96 (+84.7) 169.7 (-69.9) 1383.2 (+220.0)
22 12 (-65.7) 84 (+61.5) 188.0 (-66.7) 1253.3 (+99.2)
39 64 (+82.9) 168 (+223.1) 1324.4 (+134.8) 3125.3 (+396.8)
"Population index.
'Week: 0 = March 13, 15 = June 29, 22 = August 17, 39 = December 14, 2000.
December 2004
|
RSS
