|
Armyworm Symposium '96: Rogers
DR. HARRY R. GROSS, JR.:
CONTRIBUTIONS TO ARMYWORM RESEARCH
C. E. ROGERS
USDA, Agricultural Research Service
Insect Biology and Population Management Research Laboratory
Tifton, GA 31793-0748
ABSTRACT
Dr. Harry R. Gross, Jr., USDA, ARS, IBPMRL (deceased) developed techniques
and methodology for rearing and augmenting biological control agents to assist in the
control of the fall armyworm, Spodoptera frugiperda (J. E. Smith), and the corn ear-
worm, Helicoverpa zea (Boddie), in the southeastern USA. Dr. Gross' career with the
USDA spanned 27 years, during which he published 75 scientific papers and pre-
sented 41 oral papers on the results of his research. Dr. Gross conducted pioneering
research on white fringed beetles, kairomones, and semiochemicals of beneficial in-
sects, and patented a hive-mounted device through which exiting honey bees autodis-
seminate Heliothis nuclear polyhedrosis virus to flowering plants for control ofH. zea
larvae. However, Dr. Gross' greatest contribution to entomology was his development
of rearing and augmentation technology to enhance the use of beneficial insects for
controlling H. zea and S. frugiperda. Dr. Gross firmly believed in, and researched in-
novative ways to, use biological control for managing armyworms and other pests.
Key Words: Population management, mass rearing, augmentation, biological control
RESUME
El doctor Harry R. Gross, Jr, USDA, ARS, IBPMRL (deceso) desarroll6 t6cnicas y
metodologias de cria de agents de control biol6gico para favorecer el control de Spo-
doptera frugiperda (J. E. Smith) y Helicoverpa zea (Bodie) en el sureste de los Estados
Unidos. La carrera del Dr. Gross en el USDA dur6 27 afos, durante los cuales public
75 articulos cientificos y present 41 ponencias sobre los resultados de su investiga-
ci6n. El Dr. Gross condujo investigaciones pioneras sobre escarabajos, kairomonas y
semioquimicos de insects beneficos, y patent un equipo para las colmenas que posi-
bilitaba a las abejas diseminar el virus de la polihedrosis de Heliotis a las plants con
flores para el control de larvas de H. zea. Sin embargo, la mayor contribuci6n del Dr.
Gross a la entomologia fue su desarrollo de la tecnologia de cria para aumentar el uso
de insects beneficos para el control de H. zea y S. frugiperda. El Dr. Gross firmemente
creyo en el control biol6gico e investig6 novedosas maneras de usarlo en el manejo de
los gusanos trozadores y otras plagas.
The Armyworm Symposium of the 70th Annual Meeting of the Southeastern
Branch of the Entomological Society of America (ESA) is dedicated to the memory of
Dr. Harry Gross, Jr. and his contributions to armyworm research. Dr. Gross was Su-
pervisory Research Entomologist, USDA, Agricultural Research Service (ARS), Tif-
ton, Georgia at the time of his death; his entire entomological career of 27 years was
spent within the geographical boundary of the Southeastern Branch, ESA.
Dr. Gross was born in New Orleans, Louisiana, on 16 March, 1939, and died in At-
lanta, Georgia, on 3 May, 1994, following a short illness. Dr. Gross obtained a passion
Florida Entomologist 79(3)
Fig. 1. Harry R. Gross, Jr., Supervisory Research Entomologist (1939-1994)
for entomology at an early age, as his father was a Pest Control Operator in the New
Orleans area. After graduating from high school, Dr. Gross enrolled at Louisiana
State University where he obtained the Bachelor of Science Degree (1960), Master of
Science Degree (1964), and Doctor of Philosophy Degree in Entomology (1967).
September, 1996
Armyworm Symposium '96: Rogers
Dr. Gross' graduate student research dealt with the effects of water emulsions on
pesticidal control of wood-boring beetles; the role of the American cockroach, Periplan-
eta americana (L), and striped earwig, Labidura riparia (Pallas) in the degradation of
household fabrics; flight habits of L. riparia; and the responses of L. riparia to hep-
tachlor and mirex applied for controlling fire ants, Solenopses invicta (Buren) and S.
molesta (Say). His research with L. riparia was used by the Louisiana Pest Control
Association to develop management recommendations for households in the early
1970s.
Dr. Gross' entire professional career was spent with the USDA, ARS. Dr. Gross
joined ARS in Gulf Port, Mississippi, in 1967 and for the following four years concen-
trated research on developing systems for collecting, rearing, storing and controlling
white fringed beetles, Graphognathus spp., which are devastating pests of seeds and
seedling-stage corn, cotton, peanut, soybean, and other crops in the Southeast. Dr.
Gross retained interest in white fringed beetles throughout his career, and served as an
expert witness in hearings by the Environmental Protection Agency in 1977 and the
State of California Department of Food and Agriculture Science Advisory Panel in 1988.
Dr. Gross transferred to the Insect Biology and Population Management Research
Laboratory (IBPMRL) (formerly Southern Grain Insects Research Laboratory) in Tif-
ton, Georgia, in 1971. At the IBPMRL, Dr. Gross pursued several research areas., e.g.,
double-cropping and host resistance in managing populations of the fall armyworm,
Spodoptera frugiperda (J. E. Smith); overwintering and spring emergence chronology
of the corn earworm, Helicoverpa zea (Boddie), and tobacco budworm, Heliothis vire-
scens (Fabricius); methods for separating individual eggs of S. frugiperda from egg
masses; developing an oviposition chamber for mass production ofH. zea; devising flo-
tation techniques for separating mature and immature pupae of H. zea and H. vire-
scens for irradiation in large-scale sterility programs; assessing the importance of
visual stimuli in enhancing trapping efficiency of H. zea; the effects of homogenates of
H. zea and S. frugiperda larvae on inhibiting their oviposition in corn and aggregating
the predators Coleomagilla maculata Timberlake and Geocoris punctipes Say; and
identifying and studying the bionomics of the first known pupal parasitoid of S. fru-
giperda [Dipetimorpha intreota (Cresson)] and of H. virescens [Cryptus albitarsas
(Cresson)] that he discovered. Other researchers are now actively pursuing research
on the biology and efficacy of the pupal parasitoids as an addendum to biological con-
trol organisms which may be used to reduce field populations of noctuid pests of
southeastern agriculture.
Dr. Gross was one of the pioneer researchers demonstrating the role of kairomones
in enhancing foraging behavior of parasitoids. He formulated the concept and demon-
strated that pre-release exposure of Trichogramma pretiosum (Riley) to host
kairomones and parasitization experiences increased its efficacy against H. zea eggs
in the field. In spite of the progress that Dr. Gross and his colleagues made in this area
of research in the early to mid-1970s, he became convinced that its use as an econom-
ical pest management strategy was not likely in the near future. Hence, Dr. Gross re-
verted to researching a more conventional approach for biological control ofH. zea and
S. frugiperda.
During the last 10 years of his career, Dr. Gross concentrated his research efforts
on mass rearing and augmentation techniques for field evaluation of the tachinidAr-
chytas marmoratus (Townsend) as a biological control agent for H. zea and S. fru-
giperda. Dr. Gross developed an efficient mechanical extraction technique for
removingA. marmoratus larvae from gravid females and an efficient method for ap-
plying them to whorl-stage corn. He also developed an efficient and economical mass-
rearing procedure forA. marmoratus using larvae of the greater wax moth, Galleria
Florida Entomologist 79(3)
mellonella (L.), as a factitious host, and demonstrated that laboratory-reared flies ef-
ficiently seek and parasitize H. zea larvae in the field independent of host density. At
the time of his death, Dr. Gross was in the second year of a large 3-year pilot program
evaluating field augmentation ofA. marmoratus to control H. zea larvae in whorl-
stage corn.
An innovative accomplishment by Dr. Gross shortly before his death was the de-
signing and testing of a hive-mounted device by which the honey bee, Apis mellifera
(L.), automatically contaminates itself with a talc formulation of Heliothis nuclear
polyhedrosis virus (HNPV). As contaminated bees forage among flowers for pollen
and honey, they distribute the HNPV. Helicoverpa zea larvae feeding on crimson clo-
ver, Trifolium incarnatum (L.), exposed to HNPV-contaminated honey bees experi-
enced a significantly higher mortality due to virus than larvae feeding on non-
contaminated clover. This development resulted in a patent and has generated con-
siderable international interest and further current testing.
Dr. Gross published his research results in 75 scientific papers and orally pre-
sented papers at 41 scientific meetings. In addition to his scientific accomplishments,
Dr. Gross assumed several leadership roles in ARS as well as in professional societies.
At the time of his death, Dr. Gross was Research Leader of the Insect Biology/Man-
agement Systems Research unit at Tifton, GA, and had served as a Laboratory Direc-
tor and Assistant to the Southeast Area Director (ARS). Dr. Gross served on several
committees of the Entomological Society of America and the Georgia Entomological
Society and served as Chair of this Symposium in 1982. Dr. Gross was President of the
Southeastern Biological Control Working Group in 1990. In 1980, Dr. Gross repre-
sented the USDA in Argentina, Brazil, Paraguay, and Uruguay on an exploration trip
to collect biological control agents for S. frugiperda and other pest species. Dr. Gross
worked with a core group in 1987 to develop the ARS National Biological Control Pro-
gram. In addition to his active participation in the Entomological Society of America,
Dr. Gross also was active in the Georgia Entomological Society, Florida Entomological
Society, Sigma Xi, American Registry of Professional Entomologists, Southeastern Bi-
ological Control Working Group, and Southern Regional Projects.
The void left by the sudden passing of Dr. Gross in May 1994 remains in the Insect
Biology and Population Management Research Laboratory, the University of Georgia
Coastal Plain Experiment Station, and the Tifton community. Dr. Gross' dedication to
his profession, his great empathy, his interpersonal communication skills, his willing-
ness to assist subordinates, peers, and superiors in solving problems, and his positive
outlook and jovial countenance are just a few of the characteristics of Dr. Gross that
will not be forgotten and will be long-appreciated by his many associates and friends.
Dr. Gross is survived by his wife, Marlene, Tifton, GA, and their daughter, Lisa, At-
lanta, GA.
September, 1996
Armyworm Symposium '96: Carpenter et al.
COMPATIBILITY OF F, STERILITY AND A PARASITOID,
COTESIA MARGINIVENTRIS (HYMENOPTERA:
BRACONIDAE), FOR MANAGING SPODOPTERA EXIGUA
(LEPIDOPTERA: NOCTUIDAE): ACCEPTABILITY AND
SUITABILITY OF HOSTS
J. E. CARPENTER, HIDRAYANI' AND W. SHEEHAN2
Insect Biology and Population Management Research Laboratory
Agricultural Research Service, U. S. Department of Agriculture,
Tifton, GA 31793-0748
'Andalas University, Padang, Indonesia
'E&C Consulting Engineers, Inc., 2175 Highpoint Road,
Snellville, GA 30278
ABSTRACT
The potential for combining two alternative pest management tactics, F, sterility
and a parasitoid, was examined in the laboratory and in the greenhouse. Studies com-
pared the acceptability and suitability of progeny from irradiated (100 Gy) and non-
irradiated beet armyworm, Spodoptera exigua (Hiibner), males as hosts for Cotesia
marginiventris (Cresson). Results from these studies revealed that progeny of irradi-
ated S. exigua males and nonirradiated S. exigua females are acceptable and suitable
hosts for C. marginiventris development. Cotesia marginiventris females showed no
oviposition preference for S. exigua progeny from females paired with either irradi-
ated or nonirradiated males. Cotesia marginiventris and F, sterility appear to be com-
patible tactics that potentially could be integrated into a preventative pest
management program for S. exigua.
Key Words: Spodoptera exigua, Cotesia marginiventris, F, sterility, inherited sterility,
biological control
RESUME
Fue examinado el potential de la combinaci6n de dos alternatives de manejo de
plagas, esterilidad de la F, y un parasitoide, en el laboratorio y en un invernadero. Los
studios compararon la aceptaci6n de la progenie del gusano de la remolacha, Spodop-
tera exigua (Hiibner), irradiado y no irradiado, como hospedante de Cotesia margini-
ventris (Cresson). Los resultados revelaron que la progenie de machos irradiados y de
hembras no irradiadas de S. exigua constitute un hospedante acceptable y adecuado de
C. marginiventris. Las hembras de C. marginiventris no mostraron preferencia ovopo-
sicional por la progenie de S. exigua proveniente de hembras pareadas con machos
irradiados o no irradiados. Cotesia marginiventris y la esterilidad de la F, parecen ser
tactics compatibles que potencialmente podrian ser integradas en un program pre-
ventivo de manejo de plagas para S. exigua.
The beet armyworm, Spodoptera exigua (Hiibner), is a serious pest in cotton in the
southeastern United States, especially during outbreak conditions (Smith & Freeman
Florida Entomologist 79(3)
1994). Although many factors have contributed to the outbreaks ofS. exigua in cotton,
an unusually high level of resistance to some pesticides is implicated (Sprenkel &
Austin 1994). Alternative management strategies, such as conservation of natural en-
emies (Ruberson et al. 1994), mating disruption with synthetic pheromone (Waka-
mura & Takai 1992), and inherited sterility, are being studied for their potential role
in an integrated pest management program for S. exigua.
The potential for using F, sterility as a component of regional management of lep-
idopteran pests has been suggested by Knipling (1970) and LaChance (1985), and nu-
merous laboratory and cage studies on pests around the world have supported these
ideas (LaChance 1985, Anonymous 1993). The successful application of the F, sterility
principle to a wild population ofHelicoverpa zea (Boddie) during a recent pilot test en-
couraged further development of this pest control strategy (Carpenter & Gross 1993).
However, the high cost of rearing lepidopterans, relative to the cost of rearing dipter-
ans, has moderated researchers' enthusiasm concerning the use of F, sterility for the
control of lepidopteran pests. Nevertheless, Carpenter & Gross (1993) revealed that
even a low irradiated:wild insect ratio could significantly reduce the seasonal increase
of H. zea. In addition, population models (Knipling 1992, Carpenter 1993) have sug-
gested that F, sterility would be more efficient if combined with other pest control
strategies. Therefore, recent studies have investigated the potential of integrating F,
sterility and parasitoids for increased efficiency in suppression of pest populations.
Mannion et al. (1994 & 1995) studied the compatibility of F, sterility in H. zea and
the tachinid parasitoid,Archytas marmoratus (Townsend). They found these two con-
trol strategies to be compatible and suggested that combining the two strategies may
be useful for managing early season populations of H. zea. However, caution should be
exercised in the extrapolation of these results to other lepidopteran pests, such as S.
exigua. Parasite/host relationships are highly variable as a result of the various re-
productive strategies of both host and parasite species. For example, H. zea females
lay individual eggs spaced some distance apart which reduces mortality from larval
cannibalism. Alternatively, S. exigua females lay eggs in masses, and larvae feed gre-
gariously, in a patch, during their first two instars. Also, A. marmoratus larviposits
many maggots in the vicinity of late instar H. zea, whereas a common parasitoid for
S. exigua, Cotesia marginiventris (Cresson), stings individual, early instars.
The objectives of the present laboratory and greenhouse studies were: (1) to compare
the acceptability and suitability of progeny from irradiated (100 Gy) and nonirradiated
S. exigua males mated with nonirradiated females as hosts for C. marginiventris, and
(2) to relate these findings to the potential of combining the F, sterility technique with
resident or released C. marginiventris for managing populations ofS. exigua.
MATERIALS AND METHODS
Insects
Cotesia marginiventris and S. exigua were obtained from laboratory colonies at the
Insect Biology and Population Management Research Laboratory, Tifton, GA. All S.
exigua larvae were reared in plastic cups (30 ml) containing a meridic diet (Burton
1969) at 27 1C with a photoperiod of 14:10 (L:D) h. Cotesia marginiventris was
reared on S. exigua larvae at the same temperature and photoperiod regime.
F, Sterility Technique
The "irradiation treatment" (I) larvae were those S. exigua larvae resulting from
the mating of a nonirradiated female moth with an irradiated male moth. The "nor-
September, 1996
Armyworm Symposium '96: Carpenter et al.
mal treatment" (N) larvae originated from crosses between nonirradiated moths. Lar-
vae from both treatments were reared as above. Adult males undergoing irradiation
were <24 h old and were irradiated (100 Gy) at about 20 C with a well-type "Co source
(Gammarad Irradiator, Model GR-12, U.S. Nuclear Division, Irvine, CA) delivering
about 65 Gy per min. Dose calibration with an X-ray monitor and probe indicated a
dose error of approximately 5%.
Suitability of Progeny from Irradiated Males as Hosts
Spodoptera exigua larvae from both treatments (I & N) were stung by C. margin-
iventris. After a single sting, the host larva was transferred to a one-ounce, clear cup
containing diet. A cardboard lid was placed on the cup and the host was allowed to
complete its development. Hosts were scored as producing either a pupa or a parasi-
toid cocoon. Comparisons between larval treatments were made when host larvae
were early 1st instars, late 2nd instars, early 3rd instars, and mid 3rd instars. Percent
parasitism was calculated as the number of hosts in a treatment producing parasitoid
cocoons divided by the total number of hosts in a treatment, multiplied by 100. Wasps
developing from larvae stung as 2nd instars (n=7 for I larvae, n=9 for N larvae) were
compared for time of development, adult weight, fecundity, and longevity. Data were
subjected to an unequal n, unequal variance t-test (Steel & Torrie 1980) to separate
differences between larval treatments (a=0.05).
Acceptability of Progeny from Irradiated Males as Hosts
A single C. marginiventris female was placed onto a cotton leaf in a large, open
petri dish. The leaf had feeding damage from the previous night and 10 unstung S. ex-
igua larvae. The following behavioral events were recorded: (1) time before flying off
the 'patch' (area containing a group of feeding larvae); (2) number of stings before fly-
ing off a patch (maximum = five); (3) number of 'walkbys' (walking past an unstung
host within one half wasp body length); and (4) number of 'rejections' (antennating
unstung host without stinging). Comparisons between larval treatments were made
when host larvae were early 1st instars, late 2nd instars, early 3rd instars, and mid
3rd instars. Behavioral responses from these "no-choice" experiments were recorded
from ten wasps for each larval treatment and each host age. Comparisons between
treatments were made using the t-test (Steel & Torrie 1980).
Oviposition Preference Test for Cotesia marginiventris
Four cotton plants (2-3 months old, DPL 90) were placed in a cage in the green-
house. I and N larvae were used as host treatments. A host "patch" for each treatment
was established by confining 40 1st instars in a circle (4 cm diam) on the surface of a
leaf and allowing the larvae to feed for one day. Within the cage, there were two host
patches for each treatment and one patch per plant. Eight female C. marginiventris
(2-3 days old) were released in the cage and allowed to search for and parasitize larvae
for approximately 4 h. Host larvae were removed from the cage and dissected to de-
termine the number parasitized. This test was replicated five times. Percent parasit-
ism was calculated as the number of hosts in a treatment containing one or more
parasitoid eggs divided by the total hosts in the treatment, multiplied by 100. Because
there were no differences between the percent parasitism among patches within each
larval treatment, patches for each treatment were pooled for analysis. Comparisons
between larval treatments were made using the pooled t-test (Steel & Torrie 1980).
Florida Entomologist 79(3)
September, 1996
TABLE 1. DAYS TO EMERGENCE, ADULT WEIGHT, FECUNDITY, AND LONGEVITY OF COTE-
SIA MARGINIVENTRIS DEVELOPING IN PROGENY OF IRRADIATED (I) (100 GY)
AND NON-IRRADIATED (N) SPODOPTERA EXIGUA MALES CROSSED WITH NOR-
MAL FEMALES, PARASITIZED AS SECOND INSTARS.
Mean (S. D.)1
Parent Cross of Days To Wasp Wt No. Eggs Longevity
Host Larvae Emergence (mg) Laid Days
NY xNd 10.8 a 1.03 a 70.7 a 17.9 a
(0.68) (0.08) (30.1) (1.8)
NY xNd 11.2 a 0.91 b 94.0 a 16.7 a
(0.47) (0.09) (45.3) (1.6)
Means followed by the same letter are not significantly different (P < 0.05) (t-test).
RESULTS AND DISCUSSION
Suitability of Progeny from Irradiated Males as Hosts
Results from this study revealed that progeny of irradiated male S. exigua were
generally suitable hosts for C. marginiventris. Wasps developing in progeny of irradi-
ated males had a mean weight that was significantly (P >0.05) lower than wasps de-
veloping in progeny of nonirradiated males. However, parasitoid development,
longevity, and fecundity were not significantly influenced by the host larval treatment
(Table 1). First and second instars ofS. exigua were more suitable as hosts for C. mar-
giniventris than third instars, but host suitability within instars was not significantly
affected by the larval treatment (Table 2). The percentage of stung larvae that pro-
duced C. marginiventris cocoons was similar for both larval treatments, regardless of
the host age.
TABLE 2. EFFECT OF LARVAL AGE OF PROGENY OF IRRADIATED (I) (100 GY) AND NON-IR-
RADIATED (N) SPODOPTERA EXIGUA MALES CROSSED WITH NORMAL FEMALES
AND AS SUITABLE HOSTS FOR COTESIA MARGINIVENTRIS
Mean S.E. % Stung Larvae Producing a Cocoon (n)'
Parental cross Early 1st Late 2nd Early 3rd Mid 3rd
of host larvae Instar Instar Instar Instar
NY xNd 66.7 11 90 6 39 9 30 +9
(18) (29) (28) (30)
N x Nd 83.3 9 87.1 6 44 9 29 9
(18) (31) (27) (29)
'There was no significant difference due to larval treatment (P < 0.05) (t-test).
Armyworm Symposium '96: Carpenter et al. 293
S+1 +1 +1 +
o cc o C Id 0
+1 +
S+1 + + +1
H N
S06
Q Zc +1 +1 +1
S+1 + o ++I
o
6-1 b+ +
LO
H +i 66
S6 -2 2 I 1
I cI 0 0 t
-5 1
| I ^ | X ^
i ^ 1^C-
Florida Entomologist 79(3)
TABLE 4. COTESIA MARGINIVENTRIS OVIPOSITION PREFERENCE BETWEEN PROGENY OF
IRRADIATED (I) (100 GY) AND NON-IRRADIATED (N) SPODOPTERA EXIGUA
MALES CROSSED WITH NORMAL FEMALES
Parental Cross of Host Mean % Parasitism of
Larvae n Larval Patch (S.D.)'
NY xNd 10 23.1 a
(12)
NY xNd 10 31.4 a
(16)
'Means followed by the same letter are not significantly different (P < 0.05) (t-test).
Acceptability of Progeny from Irradiated Males as Hosts
Host acceptability was not significantly affected by the larval treatment (Table 3).
The behavior of female C. marginiventris foraging in larval patches was similar for
both larval treatments, regardless of the host age.
Oviposition Preference Test for Cotesia marginiventris
Cotesia marginiventris females showed no oviposition preference between irradi-
ated and nonirradiated male S. exigua progeny. The mean percentage of larvae para-
sitized within N larvae and I larvae patches did not differ significantly (Table 4).
Fully successful integration ofF, sterility and parasitoids into a pest management
approach can occur only if parasitoid strategies do not negatively impact irradiated
insects and their progeny more than those of the wild population, and if F, sterility
does not negatively impact the efficacy of parasitoids. Results from these studies in-
dicate that C. marginiventris and F, sterility in S. exigua are compatible control strat-
egies. The compatibility of these two strategies is congruent with the findings of
Mannion et al. (1994, 1995). Carpenter (1993) suggested several scenarios in which
the integration of compatible strategies such as F, sterility and parasitoids might be
used to control lepidopteran pest populations. For example, sterile S. exigua larvae
could be field-reared on early season host plants or nursery crops. Cotesia margini-
ventris (native and/or released) could use these sterile larvae as hosts and, thereby, in-
crease the parasitoid's early season population. Other natural enemies of S. exigua
may also use these sterile hosts. Larvae that escaped the natural enemies would pro-
duce sterile adult S. exigua that would reduce the reproductive potential of the next
generation of S. exigua.
Cotesia marginiventris is considered the dominant parasitoid of S. exigua in the
eastern half of the United States (Tingle et al. 1978, Ruberson et al. 1994). This par-
asitoid is part of a large natural enemy complex that has the capacity to suppress S.
exigua populations in cotton. However, S. exigua can become a serious pest of cotton,
especially when the natural enemy complex has been disrupted (Ruberson et al.
1994). When S. exigua populations escalate, growers often are reluctant to postpone
insecticide applications until the natural enemy complex has brought the S. exigua
population under control. Pedigo (1995) suggested that the development of new inte-
grated pest management programs should provide a special focus on the identification
of preventative tactics. Ruberson (1994) emphasized the usefulness of conserving nat-
ural enemies for effective suppression of S. exigua. Because F, sterility and C. mar-
September, 1996
Armyworm Symposium '96: Carpenter et al.
giniventris are compatible and may provide synergistic effects, further studies are
warranted to test the practicality and efficacy of integrating these two tactics for con-
trolling S. exigua.
REFERENCES CITED
ANONYMOUS. 1993. Proc. Intl. symposium on management of insect pests: Nuclear
and related molecular and genetic techniques. Jointly organized by the Intl.
Atomic Energy Agency and the Food and Agric. Organization of the U.N., Vi-
enna, 19-23 October 1992. IAEA-SM-327/35. 1993.
BURTON, R. L. 1969. Mass rearing the corn earworm in the laboratory. U.S. Dep. Agric.
Agric. Res. Serv. ARS (Ser.) 33-134, Washington, D.C.
CARPENTER, J. E. 1993. Integration of inherited sterility and other pest management
strategies for Helicoverpa zea: Status and potential. Proceedings, Intl. sympo-
sium on management of insect pests: Nuclear and related molecular and ge-
netic techniques. Jointly organized by the Intl. Atomic Energy Agency and the
Food and Agric. Organization of the U.N., Vienna, 19-23 October 1992. IAEA-
SM-327/35, pp. 363-370.
CARPENTER, J. E., AND GROSS, H. R. 1993. Suppression of feral Heliothis zea (Lepi-
doptera: Noctuidae) populations following the infusion of inherited sterility
from released substerile males. Environ. Entomol. 22: 1084-1091.
KNIPLING, E. F. 1970. Suppression of pest Lepidoptera by releasing partially sterile
males: A theoretical appraisal. BioScience 20: 456-470.
KNIPLING, E. F. 1992. Principles of insect parasitism analyzed from new perspectives:
Practical implications for regulating insect populations by biological means.
Agricultural Handbook No. 693, United States Department of Agriculture,
Washington, DC.
LACHANCE, L. E. 1985. Genetic methods for the control of lepidopteran species.
USDA, Agric. Res. Serv. ARS. 28. 40 pp.
MANNION, C. M., J. E. CARPENTER, AND H. R. GROSS. 1994. Potential of the combined
use of inherited sterility and a parasitoid, Archytas marmoratus (Diptera: Ta-
chinidae), for managing Helicoverpa zea (Lepidoptera: Noctuidae). Environ.
Entomol. 23: 41-46.
MANNION, C. M., J. E. CARPENTER, AND H. R. GROSS. 1995. Integration of inherited
sterility and a parasitoid, Archytas marmoratus (Diptera: Tachinidae), for
managing Helicoverpa zea (Lepidoptera: Noctuidae): Acceptability and suit-
ability of hosts. Environ. Entomol. 24: 1679-1684.
PEDIGO, L. P. 1995. Closing the gap between IPM theory and practice. J. Agric. Ento-
mol. 12: 171-181.
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 ene-
mies. Florida Entomol. 77: 440-435.
SMITH, R. H., AND B. L. FREEMAN. 1994. Alabama plan for the management of beet ar-
myworms. Alabama Cooperative Extension Service, Auburn University Circu-
lar Anr-842, 4pp.
SPRENKEL, R. K, AND T. A. AUSTIN. 1994. 1994 Report of the Cooperative Beet Army-
worm Trapping Program. NFREC Extension Report 94-6. 50pp.
STEEL, R. G. D., AND J. H. TORRIE. 1980. Principles and procedures of statistics: a bi-
ometrical approach, 2nd ed. McGraw-Hill, New York.
TINGLE, F. C., T. R. ASHLEY, AND E. R. MITCHELL. 1978. Parasites of Spodoptera ex-
igua, S. eridania (Lepidoptera: Noctuidae) and Herpetogramma bipunctalis
(Lepidoptera: Pyralidae) collected fromAmaranthus hybridus in field corn. En-
tomophaga 23: 343-347.
WAKAMURA, S., AND M. TAKAI. 1992. Control of the beet armyworm in open fields with
sex pheromone. pp.115-125 in N. S. Talekar [ed.]. Diamondback moth and other
crucifer pests. Proc. Intern. Workshop, Tainan, Taiwan, 10-14 December 1990.
Asian Vegetable Res. and D. V. Center Pub. No. 92-368.
Florida Entomologist 79(3)
September, 1996
FACULTATIVE EGG-LARVAL PARASITISM OF THE BEET
ARMYWORM, SPODOPTERA EXIGUA (LEPIDOPTERA:
NOCTUIDAE) BY COTESIA MARGINIVENTRIS
(HYMENOPTERA: BRACONIDAE)
JOHN R. RUBERSON' AND JAMES B. WHITFIELD2
'Department of Entomology, University of Georgia, Tifton GA 31793
'Department of Entomology, University of Arkansas, Fayetteville AR 72701
ABSTRACT
The braconid parasitoid Cotesia marginiventris (Cresson) has long been known to
be a larval parasitoid of numerous lepidopteran species. Recent field observations,
however, indicated that C. marginiventris is also capable of functioning as an egg-lar-
val parasitoid of the beet armyworm, Spodoptera exigua (Hiibner). These field obser-
vations were corroborated by laboratory observations, demonstrating that C.
marginiventris is capable of ovipositing in S. exigua eggs, and of successfully develop-
ing and emerging from host larvae hatching from stung eggs. The mechanisms used
by the parasitoids to locate host egg masses in the field were not determined. These
results lend support to phylogenetic hypotheses of the Braconidae that indicate a
close relationship between the Cheloninae and the microgastroid taxa.
Key Words: Cotesia marginiventris, Spodoptera exigua, egg-larval parasitism, parasi-
toid, Braconidae, Microgastrinae
RESUME
Desde hace tiempo se conoce que el brac6nido Cotesia marginiventris (Cresson) es
un parasitoide de numerosas species de lepid6pteros. Sin embargo, observaciones re-
cientes de campo indicaron que C. marginiventris es ademas capaz de funcionar como
un parasitoide huevo-larval del gusano de la remolacha, Spodoptera exigua (Hiibner).
Tales observaciones de campo fueron corroboradas en el laboratorio, demostrando que
C. marginiventris es capaz de ovopositar en huevos de S. exigua y merger exitosa-
mente de las larvas del hospedero eclosionadas de huevos parasitados. Los mecanis-
mos usados por los parasitoides para localizar las masas de huevos del hospedante en
el campo no fueron determinados. Estos resultados soportan la hip6tesis filogenetica
de que en los Braconidae hay una estrecha relaci6n entire los Cheloninae y los taxa mi-
crogastroides.
The parasitoid Cotesia marginiventris (Cresson) is a common and important par-
asitoid in many agricultural systems (e.g., Kok & McEvoy 1989, McCutcheon et al.
1990, Ruberson et al. 1994). It is capable of attacking a wide range of hosts, chiefly
from the lepidopteran family Noctuidae (Krombein et al. 1979), although the suitabil-
ity of hosts varies with the species and age of the host attacked (A. Datema & J. Ru-
berson, unpublished data).
Among the braconid subfamily Microgastrinae, to which C. marginiventris be-
longs, larval parasitism is assumed to be the typical behavior, although at least one
species, Cotesia hyphantriae (Riley), appears to be an egg-larval parasitoid (Tadic'
Armyworm Symposium '96: Ruberson and Whitfield
1958). Several species of known larval parasitoids in this subfamily have been dem-
onstrated in laboratory tests to be capable of facultatively parasitizing host eggs and
emerging from the larvae (Johannson 1951, Wilbert 1960). But Johannson (1951),
working with Cotesia glomerata (L.), observed that the females only did so when held
in close proximity to the eggs, and concluded that in nature only first-instar larvae of
Pieris brassicae L. are attacked by this parasitoid. Thus, it appears that at least some
species are capable of successfully parasitizing both host eggs and larvae, although
the occurrence of such events in the field has never been documented. C. marginiven-
tris has been clearly shown to be a larval parasitoid (e.g., Boling & Pitre 1970, Kun-
nalaca & Mueller 1979, Braman & Yeargan 1991), but the extent to which this
parasitoid could function as an egg-larval parasitoid has not been examined, if indeed
it has ever been considered.
Although C. marginiventris has historically been considered a larval parasitoid,
our recent field observations, supported by laboratory data reported below, indicate
that this parasitoid, while typically a larval parasitoid, is also capable of parasitizing
eggs of the beet armyworm, Spodoptera exigua (Hiibner) and emerging from the lar-
vae. This paper presents data from field and laboratory studies demonstrating that C.
marginiventris is capable of facultatively parasitizing beet armyworm eggs, and that
such an event may not be unusual in the field. We will conclude by considering this be-
havior in light of phylogenetic relationships within the Braconidae.
MATERIALS AND METHODS
Field Collections
Sampling of beet armyworm egg masses was incidentally undertaken as a compo-
nent of other projects designed to characterize and quantify the impact of the parasi-
toid complex attacking this pest in cotton (see Ruberson et al. 1994). Most of the
collections in these studies focused on larvae, but occasionally egg masses were col-
lected, chiefly to assess the impact of the egg-larval parasitoid Chelonus insularis
Cresson. At the time egg masses were collected in the respective cotton fields, larval
populations of beet armyworms were very low. In 1993, a total of 75 egg masses were
collected on 3 different dates from 3 different locations [2 in Tift County (1 northern
and 1 southern) and 1 in Laurens County, GA]. In 1995, a single egg mass was col-
lected from a third location in Tift County, GA. After each egg mass had hatched, 30
randomly-selected larvae from each egg mass were individually placed in diet cups
with a semisynthetic diet (Burton 1969) and held in the laboratory. These larvae were
examined daily for parasitoid emergence.
Laboratory Trial
A single experiment was run in the laboratory to determine whether female C.
marginiventris would sting eggs of the beet armyworm and if the parasitoid's off-
spring could successfully develop after oviposition in the host egg. Female parasitoids
were obtained from a laboratory culture that had originated from parasitized beet ar-
myworm larvae collected in cotton fields. The culture had been in the laboratory ap-
proximately one year at the time of the experiment and had been maintained using
beet armyworm larvae as hosts. Three parasitoids were each given 1 egg mass (2-d
old) and their behavior observed for 1 hour. The parasitoids were then removed, and
the egg masses held until larvae hatched, at which time 30 randomly-selected larvae
Florida Entomologist 79(3)
from each egg mass were placed individually in diet cups as above. The larvae were
examined daily for parasitoid emergence.
RESULTS AND DISCUSSION
Initial parasitoid specimens collected in 1993 appeared to be C. marginiventris,
based on the appearance of the cocoons and the emerged parasitoids, but were dis-
carded before they could be determined to species. Nevertheless, the identification of
the parasitoids was not absolutely certain at that time. Parasitoids emerging from the
egg mass collected in 1995 were determined to be C. marginiventris by JBW. We as-
sume, therefore, with considerable confidence that the parasitoids reared from egg
masses in 1993 were also C. marginiventris. Laboratory data (see below) lend further
support to the validity of this assumption.
Field sampling indicated that egg-larval parasitism of beet armyworms by C. mar-
giniventris may not be an unusual phenomenon in the field (Table 1). In the largest
sample taken (19 August 1993 in Tift County), 23% of the collected egg masses yielded
larvae parasitized by C. marginiventris. Indeed, some parasitism was found in sam-
ples of egg masses from all locations. Rates of parasitism within parasitized egg
masses ranged from 15.4% to 100%, but typically were on the order of 30-40% (Table
1). These levels of parasitism suggest either that individual female C. marginiventris
parasitize the egg masses heavily after they locate them, or that multiple females are
discovering and exploiting the egg mass (i.e., exhibiting an aggregative response).
Beet armyworm egg masses typically consist of approximately 100 eggs, although the
number can range from 40 to over 250 (Ruberson et al. 1994); therefore, parasitism
rates of 30-40% in large egg masses may represent a substantial reproductive invest-
ment by female C. marginiventris which Braman & Yeargan (1991) demonstrated to
have a realized lifetime fecundity of 362.7 offspring.
In the laboratory test, 2 of the females exhibited considerable interest in their re-
spective egg masses, repeatedly probing the egg masses with their ovipositors during
the observation period. The third female showed no interest. Johannson (1951) and
Wilbert (1960) observed similar behavior with C. glomerata and Cotesia pieridis
(Bouche) females, respectively. Only 7 larvae from the 2 stung egg masses [5 (16.7%)
from one egg mass and 2 (6.7%) from the other] yielded C. marginiventris despite the
TABLE 1. PREVALENCE OF EGG-LARVAL PARASITISM BY COTESIA MARGINIVENTRIS
AMONG FIELD-COLLECTED EGG MASSES OF THE BEET ARMYWORM,
SPODOPTERA EXIGUA.
No. Egg Masses No. Egg Masses Mean % Parasitism
Collection Date/Locale Collected Parasitized within Egg Masses'
Tift Co., GA
29 July 1993 (Site 1) 2 2 32.8 5.80
19 August 1993 (Site 2) 61 14 42.1 22.64
3 July 1995 1 1 26.7
Laurens Co., GA
4 September 1993 11 2 31.7 7.07
With SD; means are for parasitized egg masses only.
September, 1996
Armyworm Symposium '96: Ruberson and Whitfield
high level of apparent ovipositional activity by the females. Nevertheless, these re-
sults demonstrate that C. marginiventris is capable of successfully functioning as an
egg-larval parasitoid. Age of the host eggs may also influence the success of parasit-
ism-Johannson (1951) noted that only host eggs nearing eclosion were suitable for
subsequent development of C. glomerata. A similar situation may also occur for C.
marginiventris, but this was not considered in the present experiments in which eggs
of intermediate age (2-d old) were used.
The facultative parasitization of host eggs by C. marginiventris raises some inter-
esting questions regarding the foraging behavior of this parasitoid. A body of litera-
ture has demonstrated that plant kairomones activated by the feeding of host larvae
are key foraging cues for this parasitoid (Loke et al. 1983, Turlings et al. 1989, 1990,
1991). However, such cues are lacking on plants with only egg masses present and are
limited or absent in fields with low or no larval populations. The prevalence of para-
sitism observed in our field collections suggests that the parasitoids were quite suc-
cessful at locating egg masses, even when cues induced by larval feeding were rare.
What cues are being used to locate the host egg masses? It is possible that volatiles
are released from the large mass of scales that the ovipositing female beet armyworm
deposits on the egg mass; perhaps volatiles similar to those left after oviposition by
other lepidopteran species (Beevers et al. 1981, Noldus & Van Lenteren 1983). It may
also be a result of beet armyworm sex pheromone remnants on the egg mass or leaf,
as Noldus et al. (1991) found with the parasitoid Trichogramma evanescens West-
wood. Further, the plant itself may respond to some minor disruption of the leaf cuti-
cle resulting from oviposition, by releasing some volatile. Another possible
explanation is that female parasitoids were drawn into the field by the feeding of the
few larvae present in the field and, thereafter, exploited the egg masses. Whatever the
ICHNEUMONIDAE
"Cyclostomes"
"HelconoIds"
- larval parasitism Helon
Neoneurinae
Proteropini
S chneutl __ae * Ichneutini
-- egg-larval parasitism Muesebeckiini
Dirrhopinae
S- Adeliinae
larval and egg-larval """ "". cheloninae
/ Cheloninae
Moendesellnae Mendesella
Uncertain Epsiogaster
Cardlochilinae
Khoikhoiinae
j Miracinae
S Microgastrinae
Ecnomllnae
Fig. 1. Phylogeny of the microgastroid lineage of Braconidae, based on morpholog-
ical data from Whitfield & Mason (1994), with the known distribution of egg-larval
parasitism shown. Biological data from Shaw & Huddleston (1991) and Whitfield &
Mason (1994).
Florida Entomologist 79(3)
cues, however, it is quite clear that larval damage is not the only source of cues to
which C. marginiventris is capable of responding in close-range host location.
The confirmation that at least some Cotesia spp. are indeed capable of ovipositing
into host eggs is of comparative physiological and phylogenetic interest. It has been
suspected for some time that the Microgastrinae (including Cotesia) are likely to be
closely related to the Cheloninae, the members of which all typically oviposit into host
eggs and emerge from the host larvae (Fig. 1). Many other biological similarities (in
addition to the structural similarities reviewed by Tobias (1967) between the two sub-
families have now been noted: typically they have three endoparasitic larval instars
(Shaw & Huddleston 1991), they both have associations with polydnaviruses for host
immune suppression (Fleming 1992, Stoltz & Whitfield 1992), and both groups attack
largely overlapping groups of Lepidoptera. Another subfamily, the Adeliinae, has been
proposed as being the actual sister-group to the Cheloninae (e.g., Nixon 1965, Du-
darenko 1974). Recent phylogenetic work based on comparative morphology (Whit-
field & Mason 1994) appears to establish fairly firm sister-group relationship between
both Adeliinae and Cheloninae and a close relationship between these two subfami-
lies and the "microgastroid" complex of subfamilies, including the Microgastrinae
(Fig. 1). Ongoing phylogenetic work based on mitochondrial DNA sequences (Whit-
field, in prep.) also appears to confirm these relationships.
In the Cheloninae (and predictably perhaps also the Adeliinae, see Wharton 1994),
the parasitoid eggs hatch relatively soon after oviposition, but development of the first
instar is delayed (Ullyett 1949, Narayanan et al. 1961, Powers & Oatman 1984,
Biihler et al. 1985, Kawakami 1985, Shaw & Huddleston 1991). The details of the de-
velopmental and immunological interactions and the effects of venoms and polydnavi-
ruses appear to differ to some degree between the studied species of chelonines (Jones
1985, 1987, Leluk & Jones 1989) and microgastrines (reviewed by Lavine & Beckage
1995), but too few species have been studied for firm conclusions to be drawn. It would
be interesting indeed to examine in further detail the aspects in which the two groups
differ or resemble one another in their physiological responses to the common prob-
lem of egg-larval parasitism.
ACKNOWLEDGMENTS
We appreciate the assistance of Melissa Dykes, Julie McFarland, Russ Ottens, Ray
Wilson, and Daniel West in collecting beet armyworm egg masses. The reviews of Drs.
Robert M. McPherson and James D. Dutcher (Univ. of Georgia, Tifton) are also appre-
ciated. This research was supported, in part, by the Georgia Agricultural Commodity
Commission for Cotton and Cotton Incorporated.
REFERENCES CITED
BEEVERS, M., W. J. LEWIS, H. R. GROSS, JR., AND D. A. NORDLUND. 1981. Kairomones
and their use for management of entomophagous insects. X. Laboratory studies
on manipulation of host-finding behavior of Trichogramma pretiosum Riley
with a kairomone extracted from Heliothis zea (Boddie) moth scales. J. Chem.
Ecol. 7: 635-648.
BOLING, J. C., AND H. N. PITRE. 1970. Life history ofApanteles marginiventris with de-
scriptions of immature stages. J. Kansas Entomol. Soc. 43: 465-470.
BRAMAN, S. K., AND K. V. YEARGAN. 1991. Reproductive strategies of primary parasi-
toids of the green cloverworm (Lepidoptera: Noctuidae). Environ. Entomol. 20:
349-353.
BOHLER, A., T. N. HANSLIK, AND B. D. HAMMOCK. 1985. Effects of parasitization of
Trichoplusia ni by Chelonus sp. Physiol. Entomol. 10: 383-394.
September, 1996
Armyworm Symposium '96: Ruberson and Whitfield
BURTON, R. L. 1969. Mass rearing the corn earworm in the laboratory. USDA-ARS 33-
134.
DUDARENKO, G. P. 1974. Formation of the abdominal carapace in braconids (Hy-
menoptera, Braconidae). Entomol. Rev. 53: 80-90.
FLEMING, J. G. W. 1992. Polydnaviruses: mutualists and pathogens. Annu. Rev. Ento-
mol. 37: 401-425.
JOHANNSON, A. S. 1951. Studies on the relation between Apanteles glomeratus L.
(Hym., Braconidae) and Pieris brassicae L. (Lepid., Pieridae). Norsk Entomol.
Tids. 8: 145-186.
JONES, D. 1985. Endocrine interaction between host (Lepidoptera) and parasite (Che-
loninae: Hymenoptera): is the host or parasite in control? Ann. Entomol. Soc.
America 78: 141-148.
JONES, D. 1987. Material from adult female Chelonus sp. directs expression of altered
developmental programme of host Lepidoptera. J. Insect Physiol. 33: 129-134.
KAWAKAMI, T. 1985. Development of the immature stages ofAscogaster reticulatus
Watanabe (Hymenoptera: Braconidae), an egg-larval parasitoid of the smaller
teat tortrix moth, Adoxophyes sp. (Lepidoptera: Tortricidae). Appl. Entomol.
Zool. 20: 380-386.
KUNNALACA, S., AND A. J. MUELLER 1979. A laboratory study of Apanteles margini-
ventris, a parasite of green cloverworm. Environ. Entomol. 8: 365-368.
KOK, L. T., AND T. J. McAvOY. 1989. Fall broccoli pests and their parasites in Virginia.
J. Entomol. Sci. 24: 258-265.
KROMBEIN, K. V., P. D. HURD, JR., D. R. SMITH, AND B. D. BURKS. 1979. Catalog of Hy-
menoptera of America north of Mexico. Smithsonian Instit. Press, Washington
D.C.
LAVINE, M. D., AND N. E. BECKAGE. 1995. Polydnaviruses: potent mediators of host in-
sect immune dysfunction. Parasit. Today 11: 368-378.
LELUK, J., AND D. JONES. 1989. Chelonus sp. near curvimaculatus venom proteins:
analysis of their potential role and processing during development of host Tri-
choplusia ni. Arch. Insect Biochem. Physiol. 10: 1-12.
LOKE, W. H., T. R. ASHLEY, AND R. I. SAILER 1983. Influence of fall armyworm,
Spodoptera frugiperda, (Lepidoptera: Noctuidae) larvae and corn plant dam-
age on host finding in Apanteles marginiventris (Hymenoptera: Braconidae).
Environ. Entomol. 12: 911-915.
MCCUTCHEON, G. S., S. G. TURNIPSEED, AND M. J. SULLIVAN. 1990. Parasitization of
lepidopterans as affected by nematicide-insecticide use in soybean. J. Econ. En-
tomol. 83: 1002-1007.
NARAYANAN, E. S., B. R. SUBBA RAO, AND K. R. THAKARE. 1961. The biology and some
aspects of morphology of the immature stages of Chelonus narayanai Subba
Rao (Braconidae: Hymenoptera). Proc. Natl. Inst. Sci. India (B) 27: 68-82.
NIXON, G. E. J. 1965. A reclassification of the tribe Microgasterini (Hymenoptera:
Braconidae). Bull. British Mus. Nat. Hist., Entomol. Suppl. 2: 1-284.
NOLDUS, L. P. J. J., AND J. C. VAN LENTEREN. 1983. Kairomonal effects on searching
for eggs of Pieris brassicae, Pieris rapae and Mamestra brassicae of the parasite
Trichogramma evanescens Westwood. Med. Fac. Landbouww. Rijksuniv. Gent
48/2:183-194.
NOLDUS, L. P. J. J., R. P. J. POTTING, AND H. E. BARENDREGT. 1991. Moth sex phero-
mone adsorption to leaf surface: bridge in time for chemical spies. Physiol. En-
tomol. 16: 329-344.
POWERS, N. R., AND E. R. OATMAN. 1984. Biology and temperature responses of Che-
lonus kelliae and Chelonus phthorimaeae (Hymenoptera: Braconidae) and
their host, the potato tuberworm, Phthorimaea operculella (Lepidoptera:
Gelechiidae). Hilgardia 52: 1-32.
RUBERSON, J. R., G. A. HERZOG, W. R. LAMBERT, AND W. J. LEWIS. 1994. Management
of the beet armyworm in cotton: role of natural enemies. Florida Entomol. 77:
440-453.
Florida Entomologist 79(3)
SHAW, M. R., AND T. HUDDLESTON. 1991. Classification and biology of braconid wasps
(Hymenoptera: Braconidae). Royal Entomological Society of London, London.
STOLTZ, D. B., AND J. B. WHITFIELD. 1992. Viruses and virus-like entities in the par-
asitic Hymenoptera. J. Hymenop. Res. 1: 125-139.
TADIC', M. D. 1958.Apanteles hyphantriae Riley, an egg parasite of the fall webworm.
Proc. X Internatl. Congr. Entomol. 4: 859-861.
TOBIAS, V. 1967. A review of the classification, phylogeny and evolution of the family
Braconidae (Hymenoptera). Entomol. Rev. 46: 387-399.
TURLINGS, T. C. J., J. H. TUMLINSON, W. J. LEWIS, AND L. E. M. VET. 1989. Beneficial
arthropod behavior mediated by airborne semiochemicals. VIII. Learning of
host-related odors induced by a brief contact experience with host by-products
in Cotesia marginiventris (Cresson), a generalist larval parasitoid. J. Insect Be-
hav 2: 217-225.
TURLINGS, T. C. J., J. H. TUMLINSON, AND W. J. LEWIS. 1990. Exploitation of herbi-
vore-induced plant odors by host-seeking parasitic wasps. Science 250: 1251-
1253.
TURLINGS, T. C. J., J. H. TUMLINSON, AND W. J. LEWIS. 1991. Larval-damaged plants:
source of volatile synomones that guide the parasitoid Cotesia marginiventris
to the micro-habitat of its hosts. Entomol. Exp. Appl. 58: 75-82.
ULLYETT, G. C. 1949. Distribution of progeny by Chelonus texanus Cress. (Hy-
menoptera: Braconidae). Canadian Entomol. 81: 25-44.
WHARTON, R. A. 1994. Bionomics of the Braconidae. Annu. Rev. Entomol. 38: 121-143.
WHITFIELD, J. B., AND W. R. M. MASON. 1994. Mendesellinae, a new subfamily of bra-
conid wasps (Hymenoptera, Braconidae) with a review of relationships within
the microgastroid assemblage. System. Entomol. 19: 61-76.
WILBERT, H. 1960. Apanteles pieridis (Bouche)(Hym., Braconidae), ein Parasit von
Aporia crataegi (L.)(Lep., Pieridae). Entomophaga 5: 183-211.
September, 1996
' ' ' '
' '
' '
' '
' ' ' '
Florida Entomologist 79(3)
September, 1996
GROWTH INHIBITION OF FALL ARMYWORM (LEPIDOPTERA:
NOCTUIDAE) LARVAE REARED ON LEAF DIETS OF
NON-HOST PLANTS
B. R. WISEMAN, J. E. CARPENTER AND G. S. WHEELER'
Insect Biology and Population Management Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture
Tifton, GA 31793-0748
'Aquatic Weed Research, Agricultural Research Service,
U. S. Department of Agriculture, Fort Lauderdale, FL 33314
ABSTRACT
Chemists and natural plant product scientists have shown an interest in com-
pounds from exotic plants that are considered non-hosts for the fall armyworm,
Spodoptera frugiperda (J. E. Smith), but may serve as sources of materials that re-
duce feeding and growth of herbivorous insects. We report here results of laboratory
bioassays with neonate and fifth instar fall armyworm fed on a standard diet alone
and on an amended diet with celufil or leaves from dogwood, Cornus florida L., hy-
drangea, Hydrangea macrophylla (Thunb.) Seringe, black cherry, Prunus serotina
Ehrh., or Bradford pear, Pyrus calleryna Decne. The neonates had reduced growth
Armyworm Symposium '96: Wiseman et al.
and frass production when fed diets containing leaves of dogwood, hydrangea, black
cherry and Bradford pear while fifth-instars had reduced consumption and weight
gain when fed the hydrangea leaf-diet. These results suggest a toxic component in the
leaves of hydrangea that reduces the development of the fall armyworm neonates and
fifth instars. The results also indicate that the leaves of dogwood, black cherry and
Bradford pear have growth inhibitors present in their leaves that adversely affect
growth of neonate fall armyworm. Because fifth instars performed similarly to con-
trols when fed dogwood, cherry or pear leaf-diets, older larvae may be able to over-
come the plants natural defenses better than neonates.
Key Words: Spodoptera frugiperda, antibiosis, deterrent, growth inhibition, protein
RESUME
Los quimicos y los estudiosos de los products de plants naturales han mostrado
interns en complejos derivados de plants ex6ticas no consideradas como hospedantes
de Spodoptera frugiperda (J. E. Smith) pero que pueden servir como fuente de mate-
riales que reduce la capacidad de alimentarse y el crecimiento en los insects herbi-
voros. Reportamos aqui los resutados de bioensayos de laboratorio con S. frugiperda
neonatas y del quinto instar alimentadas en una dieta estandar solamente y con una
dieta con hojas de Cornus florida L., Hydrangea macrophyla (Thunb.) Seringe, Pru-
nus serotina Ehrh. o Pyrus calleryna Decne. Los neonatos mostraron reducci6n del
crecimineto y de la producci6n de excretas cuando fueron alimentados con dietas con-
teniendo hojas de C. florida L., H. macrophylla, P. serotina y P. calleryna, mientras
que el quinto instar tuvo reducci6n del consume de alimento y de la ganancia de peso
cuando fue alimentado con una dieta de H. macrophylla. Estos resultados suguieren
que un component t6xico en las hojas de H. macrophylla reduce el desarrollo de los
neonatos y quintos instares de S. frugiperda. Los resultados ademas indican que las
hojas de C. florida, P. serotina y P. calleryna tienen inhibidores de creciemiento pre-
sentes en sus hojas que afectan adversamente el crecimiento de los neonatos. Debido
a que los quintos instares se comportaron como los testigos cuando se alimentaron con
C. florida, P. serotina o P. calleryna, las larvas mas viejas parecen ser capaces de ven-
cer las defenses naturales de las plants mejor que los neonatos.
The fall armyworm, Spodoptera frugiperda (J. E. Smith), continues to be an impor-
tant economic pest as was emphasized by participants in the 1996 Armyworm Sym-
posium at the annual meeting of the Southeastern Branch of the Entomological
Society of America. One of the important controls for the fall armyworm has been
plant resistance. This has led to questions such as: (1) Why don't larvae of the fall ar-
myworm feed on all plant species and (2) could fall armyworm larvae feed on a non-
host plant in no-choice situations? Doskotch et al. (1977) stated that it is evident from
field observations of the Gypsy moth, Lymantria dispar (L.), that phytophagous in-
sects utilize very few plant species as hosts, although many plant species may be ac-
ceptable to the Gypsy moth species. Bernays (1983) suggested that the use of natural
feeding deterrents is an intuitively satisfying but logistically questionable tactic in
crop pest management. Possibly, in the near future, chemicals from non-host plants
will be exploited as new environmentally safe pesticides that can be used in biotech-
nology to render domestic crop plants to a non-host status.
Warthen et al. (1982) and Wiseman et al. (1986) searched a number of non-host
plants and found several antifeedant responses by larvae of the fall armyworm. Both
research groups used one non-host plant in common, the dogwood tree, Cornus florida
Florida Entomologist 79(3)
L. Warthen et al (1982) evaluated twigs while Wiseman et al. (1986) evaluated leaves
for response to larvae of the fall armyworm. Wiseman et al. (1986) also found non-host
plants, such as Jalapeno peppers and banana leaves (unpublished) on which the lar-
vae of the fall armyworm performed well when plant materials were mixed in diets.
The dogwood is a small forest tree, common in the Southeastern United States.
Weiner (1982) reported that the Delaware Indians boiled the bark of the dogwood in
water and used the solution to reduce fever. Hostettmann et al. (1978) found a mol-
luscicidal saponin in the bark of the dogwood and that Biomphalaria glabratus snails
were killed within 24 h by a 6 to 12-ppm solution of the saponins. Miller (1978) found
that homogenized leaf solutions of dogwood applied to the soil were toxic to the root
lesion nematode, Pratylenchus penetrans (Cobb) Chitwood and Oteifa. Villani &
Gould (1985) found that crude plant extracts of dogwood leaves did not deter feeding
by the wireworm, Melanotus communis Gyll. Warthen et al. (1982) showed that 1.5 g
of dogwood twigs in 100 g of diet resulted in about 30% mortality of fall armyworm lar-
vae. However, when Wiseman et al. (1986) used 50 g of fresh dogwood leaves in 400 ml
of diet, 8-day-old fall armyworm larvae weighed an average of only 0.4 mg. In addi-
tion, none of the larvae fed the dogwood leaf diet pupated.
Chemists and natural product scientists have shown an interest in some of the ex-
otic plant materials that were reported by Wiseman et al. (1986). Other plant species
whose leaves have compounds that also have possibilities as feeding deterrents
against larvae of the fall armyworm are: hydrangea, Hydrangea macrophylla
(Thunb.) Seringe, black cherry, Prunus serotina Ehrh., and Bradford pear, Pyrus cal-
leryna Decne. Therefore, we report results of feeding tests with leaves of these plant
species incorporated in artificial diets to determine their effects on fall armyworm lar-
val performance.
MATERIALS AND METHODS
Fall armyworm larvae were obtained from a culture maintained on a modified
pinto-bean diet (Burton & Perkins 1989) at the Insect Biology and Population Man-
agement Research Laboratory, Tifton, GA.
Four experiments each were conducted in a randomized complete-block design
with 30 replications and 1 cup per replicate. An extra set of treatments of 15 cups each
was set up to measure the decrease in moisture of each diet during the test period.
Controls for each experiment were regular pinto bean diet (Burton & Perkins 1989)
and pinto bean diet diluted by combining 3 ml of prepared diet with 2 ml of water and
mixed at a rate of 50 mg of celufil per ml of dilute diet. Celufil, a fine mesh cellulose
filler with minimum nutritive value, binds with laboratory diets and does not readily
absorb water. Diets prepared with celufil as an inert material for the controls remain
fluid and are easily dispensed. For each experiment and leaf material diet, the pinto
bean diet was diluted by combining 3 ml of prepared diet with 2 ml water and 50 mg
of ground (1 mm screen) oven-dried leaves per ml dilute diet. All experiments were
held in a controlled environmental room maintained at 28 2 C and 65 2% RH with
a photoperiod of 14:10 (L:D) h.
Experiment 1
Treatments consisted of dogwood and hydrangea leaves mixed in dilute pinto bean
diet, regular pinto bean diet, and a dilute bean diet plus 50 mg celufil per ml dilute
diet. The diet treatments were dispensed about 10 ml per cup into 30-ml plastic diet
cups. The diets were allowed to cool for about 2 h, after which 1 neonate fall army-
September, 1996
Armyworm Symposium '96: Wiseman et al.
worm was introduced into each cup and the cup was capped. Weights of larvae and cu-
mulative frass produced (weighed immediately after collection from each cup) were
recorded at 9 d.
Experiment 2
Treatments consisted of 50 mg each of oven-dried cherry and pear leaves mixed
per ml dilute pinto bean diet, regular pinto bean diet, and a dilute pinto bean diet plus
50 mg celufil per ml dilute diet. The diet treatments were dispensed into 30-ml plastic
diet cups of about 10 ml per cup. The diets were allowed to cool for about 2 h, after
which 1 neonate fall armyworm was introduced into each cup and the cup was capped.
Weights of larvae and cumulative frass produced (weighed immediately after collec-
tion from each cup) were recorded at 8 d; the same larvae, frass accumulation and diet
consumption were weighed again 24 h later (9 d).
Experiments 3 and 4
Treatments consisted of 50 mg each of oven-dried dogwood or hydrangea leaves
per ml dilute pinto bean diet (Experiment 3) and cherry or pear leaves (Experiment
4). Regular pinto bean diet and a dilute pinto bean diet plus 50 mg celufil per ml dilute
diet served as controls. Treatments of diets and procedures were as described above
except that fifth instar fall armyworm were used. Larvae were reared to the fifth in-
star on the regular pinto bean diet. Initial weights of diets and larvae were recorded
as the larvae were placed on the diets. Weight of diets, larvae, and fresh frass was re-
corded after 24 h. Weight of dry frass and diets was recorded for each after they were
dried in an oven at about 50C for 7 d. The percentage of nitrogen in the dry diets and
frass from the latter 2 experiments was determined by an FP-228 nitrogen determi-
nator (Model 601-700, Leco, St. Joseph, MI). Protein content was estimated by the for-
mula 6.25 X % nitrogen (Helrich 1990).
Data were analyzed by PROC GLM analysis (SAS Institute 1989) and significantly
different means were separated by least significant difference (P < 0.05) (SAS Insti-
tute 1989).
RESULTS AND DISCUSSION
Experiment 1
Weight of fall armyworm larvae (Table 1) after having fed on dogwood- and hy-
drangea-leaf diets for 9 d was significantly (P < 0.05) different from the weight of lar-
vae that were fed the regular pinto bean diet or diluted diet plus celufil. In fact, none
of the larvae fed hydrangea-leaf diet were alive at 9 d. The dilute diet with celufil is
capable of supporting larval growth and development equivalent to the regular pinto
bean diet. Therefore, addition of leaf materials did not simply dilute the diet and
cause a reduction in growth, i. e. the dogwood-leaf diet. But, since the larvae died on
the hydrangea-leaf diet, the presence of a toxic factor was indicated. Frass production
during the 9 d feeding was significantly less for larvae that fed on dogwood- and hy-
drangea-leaf diets than for larvae that fed on the two control diets.
Experiment 2
Larvae weighed significantly (P < 0.05) less at 8 and 9 d when fed cherry- or pear-leaf
diets than when they were fed on the control diets (Table 2). Frass production for larvae
Florida Entomologist 79(3)
TABLE 1. WEIGHT OF FALL ARMYWORM LARVAE AND FRASS FROM NEONATES REARED ON
DIETS CONTAINING DOGWOOD OR HYDRANGEA LEAVES
Weight (mg) SEM at 9 Days
Food Source Larvae' Frass1
Reg. bean diet 208.8 8.3a 192.3 12.8b
Diet + celufil 193.0 + 8.1a 347.1 12.3a
Diet + dogwood 11.6 8.3b 11.6 12.5c
Diet + hydrangea 0.0 + 8.1b 0.0 12.3c
'Means within a column followed by the same letter are not significantly different (P < 0.05 LSD; SAS Insti-
tute 1989). SEM = standard error of mean was based on pooled error mean square in the analysis.
that were fed these diets was significantly (P < 0.05) less at both 8 and 9 d than that for
larvae fed the control diets. Both consumption and weight gain during the 24-h feeding
period between days 8 and 9 were significantly (P < 0.05) less for the larvae fed the black
cherry- and Bradford pear-leaf diets than for larvae that fed on the control diets.
Experiments 3 and 4
The initial weights of the fifth instars for the various treatments were not signifi-
cantly different before they were transferred to the test diets (Tables 3 and 4). How-
ever, after 24-h, larvae fed the hydrangea-leaf diet weighed significantly (P < 0.05)
less than larvae that were fed the dogwood-leaf diet or the control diets. The mean
weight of larvae that were fed the dogwood diet was not different from the weight of
larvae fed the control diets. Larvae fed the regular pinto bean diet gained significantly
(P < 0.05) more weight during the 24-h feeding period than did the larvae fed the diet
plus celufil or the dogwood-leaf diet. Weight gain during the 24-h feeding period was
significantly (P < 0.05) less for larvae fed the hydrangea-leaf diet than for larvae fed
the other three treatment diets. Larvae consumed more diet and excreted more frass
when fed the diet plus celufil and dogwood-leaf diet than for larvae fed the regular
pinto bean diet. Consumption and weight of frass were significantly less for the larvae
that fed on the hydrangea-leaf diet than for any of the other three treatments. Per-
centage nitrogen of the diet or frass as an indication of the amount of protein in the
diets or frass were also significantly (P < 0.05) different among treatments. The regu-
lar diet had the highest percentage of nitrogen in the diet but an intermediate amount
in the frass. The diet plus the celufil had the least nitrogen in the diet and the frass.
The hydrangea-leaf diet had an intermediate percentage nitrogen in the diet but the
greatest percentage nitrogen in the frass.
Larvae fed the regular pinto bean diet utilized more of the protein consumed than
did those that were fed on the other diets. Larvae feeding on the diet plus the celufil
consumed more diet than larvae fed the regular pinto bean diet. Nitrogen content was
lower in the diet plus celufil and in frass from larvae fed the diet plus celufil, indicat-
ing that the larvae that were fed this diet had to consume more diet and excrete less
nitrogen to obtain the necessary protein for the growth. However, since the larvae ex-
creted more frass, they also excreted more nitrogen (5.04 versus 3.46 mg) than did lar-
vae that were fed the regular pinto bean diet. Larvae that fed on the dogwood-leaf diet
excreted more than twice as much frass as those fed the regular diet. Larvae fed the
regular pinto bean diet and dogwood-leaf diet consumed almost equal amounts of pro-
September, 1996
Armyworm Symposium '96: Wiseman et al.
8~a
C
Cab
C C
+1 +1 +1
cq c
+ +1 +1
CC CC C+
+1 +1 +1
C '
+1 +1 +1
c-ce
+1 +1 +1
COe
+1 +1 +1
N 3^
m
GO
- a
~m
mcc
6 xS
3 8
0 !i g
8
~ccfl
& ^
Hiw
4m O
C
G)
C cc a>
in l
C) C Cc
li |
bi nc'
bb) a) o
p? a Qff
September, 1996
VI
C
cc~
C
Cl
E
+-'
(U
Qa
E
cc
cc
Florida Entomologist 79(3)
a
co
"c
I o
,
0 .
m .,
e,,
.m
6 s
in .
IJI
'-'4J
" "
1.?
Armyworm Symposium '96: Wiseman et al.
c)
C
C
icc
C)
(1)
QZ
^M
cc
rt
a,
u
cc
fc ^
a,
3!
+^E
a
m
I
- a
~m
Q "S
a -I -
0 !i g
8
~ccfl
& ^
Hi~
fc ^
liw
*^s
C
C)
C
C)g '3
St -g
bfl c2 ^
C) C)Q FQ
Florida Entomologist 79(3)
tein (29 versus 28 mg), but excreted more than twice as much nitrogen (3.45 versus
7.02 mg). Fifth-instar fall armyworms fed the hydrangea-leaf diet consumed less than
one-half the amount of diet consumed by larvae fed on the dogwood-leaf diet. And, the
nitrogen consumption was about one-half as much. However, the nitrogen excreted by
larvae fed the hydrangea-leaf diet was less than one-third as much as that excreted
by larvae fed the dogwood-leaf diet and similar to the nitrogen excreted by larvae fed
the regular pinto bean diet.
Weight of the larvae that fed on the pinto bean diet was significantly (P < 0.05)
greater after 24-h than the weight of larvae fed the diet plus celufil and pear-leaf diet
(Table 4). Larvae fed the cherry-leaf diet weighed significantly more than larvae that
fed on the diet plus celufil diet. However, consumption was significantly (P < 0.05)
greater for larvae fed the cherry-leaf diet than for larvae that fed on the other treat-
ment diets. Frass production was also significantly (P < 0.05) greater for larvae fed the
cherry-leaf diet than for the other treatments. Frass production was the least from
larvae that fed on the regular diet. Percentage nitrogen in the pinto bean diet was sig-
nificantly higher than the percentage nitrogen in the cherry, pear or celufil diets. Per-
centage nitrogen was lowest in the celufil diet as well as in the frass produced. The
consumption of nitrogen was highest by larvae fed the cherry-leaf diet (32.0 mg) and
least by larvae fed the celufil diet (20.0 mg). However, percentage nitrogen in the frass
produced by larvae fed the diets of regular bean, cherry and Bradford pear-leaf were
not significantly different. The amount of nitrogen in the frass varied considerably
from 11.2 mg for larvae that fed on the cherry leaf diets to those that fed on the pear-
leaf diet (7.94 mg), regular diet (5.18 mg) and celufil check (5.64 mg).
In summary, fall armyworm neonates did not perform well after feeding on the
dogwood, hydrangea, black cherry and Bradford pear-leaf diets. Larvae may have ex-
hibited compensatory feeding when offered the celufil and dogwood-leaf diets as their
consumption increased over larvae that fed on the regular pinto bean diet. The results
also indicated that the leaves of dogwood, black cherry and Bradford pear have
growth inhibitors present in their leaves that adversely affect the growth of neonate
fall armyworm. Fall armyworm fifth instars had reduced consumption and weight
gain when fed a hydrangea-leaf diet. These results suggested either a severe growth
inhibitor or a toxic component in the leaves of hydrangea that caused total mortality
as neonates and reduced the performance of the fall armyworm fifth instars. Larvae
that fed on the cherry- and pear-leaf diets excreted much higher amounts of nitrogen
than larvae that fed on either control diets, indicating a poor assimilation of the pro-
tein in these diets. Lastly, these non-hosts offer good possibilities for the discovery of
new chemicals that could be used in the management of the fall armyworm.
ACKNOWLEDGMENTS
We thank Johnny Skinner, Charles Mullis, and Robert Caldwell for their technical
support.
REFERENCES CITED
BERNAYS, E. A. 1983. Antifeedants in crop pest management. pp. 259-269 in D. C.
Whitehead and W. S. Bowers [eds.], Current themes in tropical science. Vol. 2.
Natural products for innovative pest management. Pergamon Press. New York.
BURTON, R. L., AND W. D. PERKINS. 1989. Rearing the corn earworm and fall army-
worm for maize resistance studies, pp. 37-45 In Toward insect resistant maize
for the third world: Proceedings of the international symposium on methodolo-
gies for developing host plant resistance to maize insects. International Maize
and Wheat Improvement Center (CIMMYT). El Batan, Mexico, D. F.
September, 1996
Armyworm Symposium '96: Wiseman et al. 311
DOSKOTCH, R. W., T. M. ODELL, AND P. A. GODWIN. 1977. Feeding responses of Gypsy
moth larvae, Lymantria dispar, to extracts of plant leaves. Environ. Entomol.
6: 563-566.
HELRICH, K. 1990. Official methods of analysis. Assoc. Off. Anal. Chem. 1: 71-72.
HOSTETTMANN, K., M. HOSTETTMANN-KALDAS, AND K. NAKANISHI. 1978. Mollusci-
cidal saponins from Cornus florida L. Helvetica Chemica Acta. 61: 1990-1995.
MILLER, P. M. 1978. Toxicity of homogenized leaves of woody and herbaceous plants
to root lesion nematodes in water and in soil. J. American Soc. Hort. Sci. 103:
78-81.
SAS INSTITUTE. 1989. SAS/STAT user's guide, version 6, 4th ed., vol. 1. SAS Institute,
Cary, NC.
VILLANI, M., AND F. GOULD. 1985. Screening of crude extracts as feeding deterrents of
the wireworm, Melanotus communis. Entomol. exp. appl. 37: 69-75.
WARTHEN, J. D., JR., R. E. REDFERN, E. C. UEBEL, AND G. D. MILL, JR 1982. Antifeed-
ant screening of thirty-nine local plants with fall armyworm larvae. J. Environ.
Sci. Health A17: 885-895.
WEINER, M. A. 1982. Earth medicine-earth foods. Plant remedies, drugs, and natural
forms of the North American Indians. MacMillian, New York.
WISEMAN, B. R., R. E. LYNCH, K. L. MIKOLAJCZAK, AND R. C. GUELDNER 1986. Ad-
vancements in the use of a laboratory bioassay for basic host plant resistance
studies. Florida Entomol. 69: 559-565.
4444444444444444444444444444444444444444444444444444
Armyworm Symposium '96: All et al.
CONTROLLING FALL ARMYWORM
(LEPIDOPTERA:NOCTUIDAE) INFESTATIONS IN WHORL
STAGE CORN WITH GENETICALLY MODIFIED BACILLUS
THURINGIENSIS FORMULATIONS
J. N. ALL', J. D. STANCIL', T. B. JOHNSON2 AND R. GOUGER2
'Department of Entomology
University of Georgia, Athens, GA 30602
'Ecogen Inc., Langhorne, PA 19047
ABSTRACT
Development of more potent strains of Bacillus thuringiensis (Berliner) (Bt) using
recombinant DNA technology may lead to improved insecticidal products for control-
ling pests like the fall armyworm, Spodoptera frugiperda (J. E. Smith), which have
been difficult to manage with conventional Bt-based products. EG1999, a variant Bt
strain developed by recombinant DNA technology from EG2348 (the active ingredient
of the bioinsecticide Condor"), showed improved control of FAW infestations in whorl
stage field corn in 1994 (oil flowable and granular formulations) and sweetcorn in
1995 wettablee granule (ECX9526) (WG)] and granular [ECX9526 (G) formulations]
as compared with Javelin"(WG), an insecticide derived from a naturally occurring Bt
strain, and methomyl (Lannate LV). In 1994, leaf samples taken 1 h, 1, 2, 3, 5, and 7
d after spraying and assayed in the laboratory with fall armyworm showed that the
Bt products (recombinant and natural) had less than 48 h residual toxicity to 1st in-
star larvae. In 1994 freshly hatched FAW were placed on plants in A-frame cages to
evaluate this as a method to establish artificial field infestations for use when natural
Florida Entomologist 79(3)
FAW populations are low. The plant damage and insecticide control in the cages were
similar to the heavy natural infestations outside the cages.
Key Words: Spodoptera frugiperda, Bt formulations, sweetcorn
RESUME
El desarrollo de cepas mas potentes de Bacillus thuringiensis (Berliner) (Bt) me-
diante el uso de la tecnologia del DNA recombinante puede contribuir a la mejora de
los products insecticides para el control de plagas tales como Spodoptera frugiperda
(J. E. Smith), que han sido dificiles de manejar con products comerciales a base de Bt.
EG1999, una variante de una cepa de Bt desorrollada mediante la tecnologia del DNA
recombinante a partir de EG2348 (el ingredient active del insecticide Condor ) mos-
tr6 un mejor control de S. frugiperda en maiz en estado de cogollo en 1994 (formula-
ciones de aceite flowable y granulado), en maiz dulce en 1995 [granulos humedecibles
(ECX9526 (WG)) y formulaci6n granular (ECX9526 (G))] al compararse con Javelin'
(WG), un insecticide derivado de una cepa natural de Bt, y con methomyl (Lannate'
LV). En 1994 las muestras tomadas a 1 hora, y a los 1, 2, 3, 5, y 7 dias depues de la
aplicaci6n y ensayadas en el laboratorio con S. frugiperda mostraron que los produc-
tos de Bt (recombinante y natural) tenian una toxicidad residual al primer instar me-
nor de 24 horas. En 1994 larvas reci6n eclosionadas fueron puestas sobre plants en
jaulas para evaluar un m6todo para establecer infestaciones de campo cuando las po-
blaciones naturales de S. frugiperda eran bajas. El dano a las plants y el control con
insecticide dentro de las jaulas fueron similares a las fuertes infestaciones naturales
que hubo fuera de las jaulas.
Naturally occurring Bacillus thuringiensis (Berliner) (Bt) based insecticides have
had low to moderate effectiveness against the fall armyworm (FAW), Spodoptera fru-
giperda (J. E. Smith) (Gardner & Fuxa 1980; Krieg & Langenbruch 1981; Teague
1993). Biopesticide products containing Bt continue to be improved with the use of ge-
netic technologies. ECX9399 (Ecogen, Inc., Langhorne, PA) showed improved toxicity
as compared with conventional Bt products to FAW in corn field tests in three south-
eastern states (All et al. 1994). ECX9399 was an oil flowable (OF) formulation of a re-
combinant DNA variant (EG1999) of EG2348, the active strain in the Bt product
Condor'. The present research evaluates field efficacy and stability of EG1999 in an
(OF), wettable granule (WG), and granular (G) formulation for FAW control in whorl
stage field corn (1994) and sweetcorn (1995).
MATERIALS AND METHODS
The field tests were conducted near Athens at the University of Georgia Plant Sci-
ences Farm using Dekalb 689 variety field corn in August 1994 and Silverqueen vari-
ety sweetcorn in August 1995. Plots were 4 rows x 6 m long with 95 cm row width and
1 m alleys and were arranged in a randomized complete block experimental design
with 4 replications.
Prior to spraying, a FAW baited pheromone trap was monitored weekly for moth
flights and the field was sampled for larvae and damage by examining 50 plants at
random at 3 to 5 day intervals. Spraying was initiated when 50% of the plants showed
damage by larvae scarificationn of leaves). In both years plants were in the two leaf
stage, about 25 cm tall, when insecticide application was initiated. Sprays were ap-
September, 1996
Armyworm Symposium '96: All et al.
plied with a CO, backpack sprayer in 190 liter/ha spray volume, at 3.2 kg/cm2 pres-
sure, with a TG3 nozzle about 20 cm over the plant whorl. Granular formulations
were applied in an 18 cm wide band over the row using a Noble granular spreader
and small plot procedures described in All & Hammes (1977). Four applications of the
recombinant DNA Bt insecticides were made at 7-day intervals. In 1994 (OF) and (G)
formulations of ECX9426 were tested and in 1995 (WG) and (G) formulations of
ECX9526 were evaluated (see Table 1). Methomyl (Lannate LV) was used as a con-
ventional insecticide standard in 1994 and 1995, and a Bt (subspecies kurstaki) stan-
dard, Javelin (WG), was used in 1994 at recommended rates for a conventional Bt
standard.
In 1994 artificial infestations using neonate FAW were conducted by placing an A-
frame cage 120 cm long x 60 cm wide x 105 cm high covered with lumite insect screen
(32 x 32 mesh) on top of a treated 5 x 10 cm wood base that had been buried 2.5 cm
deep in the middle of one row of each plot. The cage covered 6 to 8 plants and was re-
moved at the time of spraying. After the sprays dried, 5 freshly hatched FAW larvae
(obtained from the Insect Biology and Population Management Research Laboratory,
USDA-ARS, Tifton GA) were placed in the whorl of each plant; the plants were in-
fested again 3 days later. Infestations were continued after each spray application.
Efficacy was determined at selected intervals by examining all plants and rating
damage as 1 = no damage or very slight (<5%) defoliation, no feeding or excrement in
the whorl ("no whorl damage", NWD); 2 = light (5-10% defoliation), NWD; 3 = moder-
ate defoliation (up to 20%), NWD; 4 = light to moderate defoliation (up to 20%) and
light whorl damage (WD); 5 = moderate to heavy defoliation (20-40%) and moderate
WD; 6 = heavy (>40%) defoliation and severe WD; and 7 = only midribs of leaves left,
whorl destroyed. Five (1994) or 10 (1995) plants in each plot were examined for dead
or live larvae and their size was classified as 1 instarr 1 or 2); 2 instarr 3 or 4), or 3 (in-
star 5 or 6).
One month after the final rating, the remaining plants within each plot and in the
cages (1994) were harvested and weighed. Samples of the plants were then dried for
10 days at 88 C to obtain biomass estimates of the plants, and the weight data were
converted to dry weight.
In 1994 a study was conducted to determine the toxicity and residual effects of the
spray treatments on FAW larvae. Samples were taken from the midsection of leaves
in each plot about 1 h after the first spray application, as well as 1, 2, 3, 5, and 7 d post
spraying. Leaves (each treatment replicated 8 times) were cut into 5 cm sections and
placed in petri dishes containing moistened filter paper. Each leaf sample was in-
fested with 5 neonate larvae and the plates were sealed with parafilm and placed in
an incubator (31 C) for 24 h. The plates were then unsealed and larval mortality was
determined and defoliation rated on an ascending scale of 1 = no feeding to 6 = heavy
feeding. Surviving larvae were then placed in individual diet cups containing pinto
bean diet and their status was visually monitored for 7 d. Final mortality counts were
made, larval size measured under a dissecting microscope, and the larvae dried for 8
d at 88 C and weighed. The data were analyzed using analysis of variance and Dun-
can's new multiple range analysis procedures on SAS microcomputer based statistical
analysis software (SAS 1985).
RESULTS AND DISCUSSION
Natural infestations were higher in 1994 than 1995, but untreated check plants
suffered serious damage (damage rating of 6.3 in natural infestations and 6.7 in cages
in 1994 as compared with 4.1 natural infestation in 1995) both years. When plants
Florida Entomologist 79(3)
TABLE 1. DAMAGE RATINGS OF FAW INFESTATIONS IN WHORL STAGE FIELD CORN
(1994) AND SWEETCORN (1995) TREATED WITH RECOMBINANT DNA BT FOR-
MULATIONS AND SELECTED CONVENTIONAL INSECTICIDE STANDARDS.
% Control2 No. Live Larvae/Plant2
Cage Cage
Fields Plants Plants Field Plants Plants
Treatment and -
Rate' 18 d 28 d 21 d 18 d 28 d 21 d
1994
ECX9426 OF 1.2 32 a 36 a 52.5 a 3.0 ab 1.7 a 6.0 a
ECX9426 OF 2.4 64 b 44 a 79.1 b 2.3 ab 1.0 b 2.0 ab
ECX9426 G 5.6 32 a 28 a 50.0 a 2.5 ab 1.6 ab 4.8 ab
ECX9426 G 11.3 68 b 72 b 79.9 b 1.6 bc 0.5 bc 2.0 ab
Javelin" WG 1.1 40 a 28 a 31.3 a 3.5 a 1.0 bc 4.0 ab
Methomyl 0.5 80 c 64 b 86.1 b 0.5 c 0.5 c 1.8 b
1995
ECX9526 WG 1.1 60 a 33 b 0.3 a 0.45 a
ECX9526 WG 2.4 67 a 52 a 0.3 a 0.53 a
ECX9526 G 5.6 20 b 3 b 0.3 a 0.70 a
ECX9526 G 11.3 20 b 13 b 0.2 a 0.55 a
Methomyl 0.5 72 a 69 a 0.1 b 0.38 a
'Rates of ECX9426 (OF) are presented as volume (liters) of formulated product per ha, ECX9526 (WG), Jav-
elin (WG), ECX9426 (G) and ECX9526 (G) as weight (kg) of formulated product per ha and rate of methomyl as
weight (kg) of active ingredient per ha.
2Means followed by the same letter within a column and year are not significantly different in Duncan's new
multiple range test (P < 0.05). Percent control values for treatment are based on damage ratings compared with
untreated checks.
were examined for dead and live larvae, insects of all sizes were found in all of the
treatments. Table 1 shows that in 1994 the higher rate of both the ECX9426 (OF) and
(G) formulations gave superior percent control [1 (treatment damage rating/check
rating) x 100] as compared to Javelin (WG) at a rate of 1.1 kg/ha. However, only the
(G) formulation was statistically similar to methomyl at 0.5 kg active ingredient per
ha in the 28 d sample.
The high 1994 natural infestation rate resulted in levels of plant damage similar
to the injury that occurred in the infestations in cages. Relative efficacy of the various
treatments in the cages was similar to their performance on the non-caged plants in
plots subjected to natural infestations. These results indicate that use of artificial in-
festations in A-frame cages could be a useful method to simulate high natural infes-
tations in situations where indigenous FAW populations are low.
In 1995 the percent FAW control at the 2.4 kg rate of ECX9526 (WG) was similar
to that provided by methomyl, but efficacy of ECX9526 (G) was reduced substantially
compared with a similar formulation used in 1994 (Table 1). Overall, weather was
drier in 1995 during the test period, and this may have negatively influenced FAW con-
trol by the (G) formulation by reducing dispersion of the bacteria within the whorls.
September, 1996
Armyworm Symposium '96: All et al.
In the 1994 leaf sample assay of the two ECX9426 (OF) rates, Javelin" (WG), and
methomyl, all of the treatments were highly toxic to larvae exposed to leaves taken
one hour following spraying (Table 2). The granular formulation of ECX9426 was not
tested because of poor leaf coverage. Toxicity of all of the Bt products dropped dramat-
ically on the leaves collected 24 h after spraying, demonstrating a rapid loss in resid-
ual potency under field conditions. FAW mortality on Bt sprayed leaves was reduced
even further 2 days after spraying and was less than 8% (statistically similar to the
check leaves) on days 3, 5, and 7 after spraying (for this reason only data up to day 3
are presented in Table 2). Methomyl residues on sprayed leaves were highly toxic to
FAW larvae up to the 2 day sample, but substantially less mortality occurred on
leaves collected on day 3. Toxicity of methomyl residues was less than 5% on days 5
and 7 and was not significantly different from mortality on check leaves after day 3.
The growth of larvae surviving from exposure on treated leaves and then placed on
diet for 7 d was determined as an indicator of physiological inhibition resulting from
insecticide toxicity. Weight and length measurements followed similar trends, there-
fore only size measurements are presented in Table 3. Survivors from all of the insec-
ticide leaf samples taken on the same day as spraying were significantly smaller than
the larvae exposed on check leaves after 7 d on diet. The size of survivors was reduced
on the high, but not the low, rate of ECX9426 (OF) on 24 and 48 h leaf samples;
whereas the size of survivors from the Javelin treatment was not reduced in the 24
h leaf sample or thereafter. Reduced FAW growth by survivors on methomyl treated
leaves occurred for up to 3 d after spraying, but not from leaves collected at 5 and 7 d
after spraying. These results indicate that residues of ECX9426 (OF) sprayed on corn
leaves at a rate of 2.4 kg/ha reduce growth of FAW survivors and are significantly
more effective than Javelin (WG) at 1.1 kg/ha. However, the data also demonstrate
the weak residual activity (less than 2 d) of Bt products on corn leaves for FAW
TABLE 2. FAW LARVAL MORTALITY IN 1994 FOLLOWING FEEDING ON CORN LEAF SAM-
PLES COLLECTED AT SELECTED INTERVALS AFTER TREATMENT WITH BT-BASED
AND CONVENTIONAL INSECTICIDE PRODUCTS.
% Mortality'
Period Treatment and Rate2
Between
Spray and Observation ECX9426 (OF) Javelin Methomyl
Collection Time 1.2 2.4 1.1 0.5 Check
1h 24 h 47.5 c 55.0 c 72.5 b 97.5 a 0.0 d
7 d 65.0 c 90.0 ab 82.5 b 100.0 a 0.0 d
Id 24 h 25.0 b 22.5 b 2.5 c 90.0 a 0.0 c
7d 35.0 c 65.0 b 5.0 d 95.0 a 7.5 d
2d 24 h 12.5 be 22.5 b 7.5 c 97.5 a 0.0 c
7 d 15.0 bc 30.0 b 15.0 bc 97.5 a 0.0 c
3d 24 h 0.0 b 2.5 b 5.0 ab 17.5 a 2.5 b
7d 2.5 b 7.5 ab 7.5 ab 20.0 a 2.5 b
'Means followed by the same letter across a row are not significantly different in Duncan's new multiple
range test (P < 0.05).
2Rates of ECX9426 (OF) are presented as volume (liters) of formulated product per ha, Javelin (WG) as
weight (kg) of formulated product per ha and methomyl as weight (kg) of active ingredient per ha.
Florida Entomologist 79(3)
TABLE 3. GROWTH (AFTER 7 DAYS ON DIET) OF FAW LARVAE SURVIVORS FROM CORN
LEAF SAMPLES COLLECTED AT SELECTED INTERVALS FOLLOWING SPRAYING
WITH BT-BASED AND CONVENTIONAL INSECTICIDE PRODUCTS.
Mean Size (cm) of Survivors of 7 D'
Treatment and Rate2
Leaf Collection ECX9426 (OF) Javelin" Methomyl
Following Spray 1.1 2.4 1.1 0.5 Check
1 h 1.65 b 1.20 bc 1.33 c 1.00 d 3.00 a
1 d 2.25 ab 1.88 bc 2.95 a 1.10 d 2.83 a
2 d 2.70 ab 2.38 b 2.70 ab 1.05 c 3.00 a
3 d 2.90 ab 2.83 ab 2.78 ab 2.60 b 2.95 a
5 d 2.93 ab 3.00 a 2.93 ab 2.78 b 2.95 ab
7 d 2.75 a 2.75 a 2.90 a 2.73 a 2.95 a
'Means followed by the same letter across a row are not significantly different in Duncan's new multiple
range test (P < 0.05).
2Rates of ECX9426 (OF) are presented as volume (liters) of formulated product per ha, Javelin (WG) as
weight (kg) of formulated product per ha and methomyl as weight (kg) of active ingredient per ha.
Silage (dry wt) yield in the treatments paralleled efficacy trends. In 1994, plant
production in check and Javelin plots was 5229 and 6250 kg/ha, respectively, and
these were significantly less than yields in the methomyl plots (9408 kg/ha). Yield of
the ECX9426 (OF) treatments was 6232 and 6698 for the 1.2 and 2.4 kg/ha rates, re-
spectively. In comparison, the ECX9426 (G) treatments yielded 7619 and 6904 kg/ha,
respectively, at the 5.6 and 11.3 kg/ha rates, but these were not statistically different
from the other treatments. Yield in 1995 was 6368 kg/ha in check plots, which was sig-
nificantly less than all of the treatments except the 2.4 kg/ha rate of ECX9526 (G)
which averaged 7375 kg/ha. Methomyl at 0.5 kg active ingredient per ha had highest
yield with 15,943 kg/ha yield in dry plant matter followed by ECX9526 (WG), 2.4 kg/
ha (15,765 kg/ha), > ECX9526 (WG) 1.2 kg/ha (11,606 kg/ha), > ECX9526 (G) (11,445
kg/ha), all of which were statistically similar to each other (Duncan's new multiple
range test (P < 0.05)), but significantly higher than the yield in check plots.
Overall, the two years of tests with four formulated Bt products (ECX9426 (OF),
ECX9526 (WG), ECX9426 (G) and ECX9526 (G)) derived from a recombinant DNA
strain EG1999 demonstrated that at the rates tested the materials provided im-
proved control of FAW infestations in corn compared with a formulation of an indige-
nous Bt strain (Javelin). Under certain circumstances, the recombinant DNA Bt
products produced control of FAW similar to that of a conventional insecticide (meth-
omyl). It appears that the recombinant DNA Bt products have low residual potency
which has been a problem with most Bt-based insecticides (Krieg & Langenbruch
1981). These tests verified 1993 research with EG1999 formulations tested in three
southeastern states (All et al. 1994). Ecogen Corp. is currently pursuing registration
of an EG1999 product under a tentative trade name of Crystar.
REFERENCES CITED
ALL, J. N., AND G. G. HAMMES. 1977. Application technique for screening granular in-
secticides in row crops. J. Georgia Entomol. Soc. 12: 90-92.
September, 1996
Armyworm Symposium '96: All et al. 317
ALL, J. N., J. D. STANCIL, T. B. JOHNSON, AND R. GOUGER 1994. A genetically-modi-
fled Bacillus thuringiensis product effective for control of the fall armyworm
(Lepidoptera: Noctuidae) on corn. Florida Entomol. 77: 437-440.
GARDNER, W. A., AND J. R. FUXA. 1980. Pathogens for the suppression of the fall ar-
myworm. Florida Entomol. 63: 439-447.
KRIEG, A., AND G. A LANGENBRUCH. 1981. Susceptibility of arthropod species to Ba-
cillus thuringiensis, pp. 837-896 in H. D. Burges [ed.], Microbial control of pests
and plant diseases 1970-1980. Academic, New York.
SAS INSTITUTE. 1985. SAS Users Guide: Statistics, Version 6 ed. SAS Institute, Cary,
NC.
TEAGUE, T. G. 1993. Control of fall armyworm in sweet corn with Bacillus thuringien-
sis, 1991. Insecticide & Acaricide Tests 18: 127-128.
Armyworm Symposium '96: Davis et al.
INSECT COLONY, PLANTING DATE, AND PLANT GROWTH
STAGE EFFECTS ON SCREENING MAIZE FOR LEAF-FEEDING
RESISTANCE TO FALL ARMYWORM (LEPIDOPTERA:
NOCTUIDAE)
F. M. DAVIS', B. R. WISEMAN2, W. P. WILLIAMS', AND N. W. WIDSTROM2
USDA, ARS, Crop Science Research Laboratory
Mississippi State, MS 39762
'Crop Science Research Laboratory, USDA, ARS, Mississippi State, MS 39762
2IBPMRL, USDA, ARS, Tifton, GA 31793
ABSTRACT
Field experiments were conducted at Mississippi State, MS and Tifton, GA to de-
termine effects of laboratory insect colony, planting date, and plant growth stage on
screening maize, Zea mays L., for leaf-feeding resistance to the fall armyworm (FAW),
Spodoptera frugiperda (J. E. Smith). The experiments were conducted using a ran-
domized complete block design with treatments in a factorial arrangement with 6 rep-
lications. Treatments consisted of 2 insect colonies, an early and a late planting
period, 2 plant growth stages, and 4 single cross maize hybrids (2 susceptible and 2 re-
sistant to leaf-feeding by FAW) at each location. Each plant in an experiment was in-
fested with 30 neonate FAW larvae when the plants of the second planting within
each planting period reached the V4 (Tifton) or V, (Mississippi State) stage. Each plant
was visually scored for leaf damage 7 and 14 days after infestation. Statistical analy-
ses revealed interactions among factors resulting in inferences having to be made us-
ing nonmarginal means. Significant differences in rating scores within each factor
(insect colony, planting date, and plant growth stage) were found for some compari-
sons. However, none of these factors appreciably altered our ability to distinguish be-
tween resistant and susceptible genotypes which is the objective of screening.
Key Words: Spodoptera frugiperda, plant resistance, maize hybrids, screening
RESUME
Se llevaron a cabo experiments de campo en el estado de Mississippi y en Tifton,
Georgia, para determinar los efectos de una colonia de insects de laboratorio, fechas
Florida Entomologist 79(3)
de simbra y estado de crecimiento de las plants en el tamizaje de plants de maiz,
Zea mays L., resistentes al daio causado por Spodoptera frugiperda (J. E. Smith). Los
experiments fueron conducidos usando un diseio de bloques completamente aleato-
rizados con los tratamientos ordenados en un diseio factorial con 6 replicas. Los tra-
tamientos consistieron en dos colonies de insects, un period de siembra temprano y
otro tardio, dos estados de crecimiento, y cuatro cruces sencillos de maiz hibrido (dos
susceptibles y dos resistentes al gusano) en cada localidad. Cada plant en cada ex-
perimento fue infestada con 30 larvas neonatas de S. frugiperda cuando las plants de
la segunda siembra dentro de cada period de siembra alcanzaron los estados V4 (Tif-
ton) or V, (Mississippi). Cada plant fue visualmente evaluada en cuanto a daio a las
hojas a los 7 y 14 dias de la infestaci6n. Los analisis estadisticos revelaron interaccio-
nes entire los factors, lo que dio como resutado el tener que hacer inferencias me-
diante el uso de medias no marginales. En algunas comparaciones fueron encontradas
diferencias significativas en el valor de cada factor (colonia de insects, fecha de siem-
bra, y estado de crecimiento de las plantss. Sin embargo, ninguno de esos factors
permiti6 distinguir entire los genotipos resistentes y susceptibles, lo cual fue el obje-
tivo del tamizaje.
The fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith), is a polyphagous
insect that attacks many crops in the Americas (Ashley et al. 1989). In the southeast-
ern United States, the FAW possess a serious threat to maize, Zea mays L., produc-
tion, especially to late-season plantings (Scott et al. 1977). FAW often attack late-
season plantings of maize from seedling to mature stages of growth. Protection of
these plantings by multiple applications of insecticides is not practical. Therefore,
more economical management tactics must be pursued. The best tactic would be to
grow maize that naturally resists the feeding of the pest.
Entomologists and plant geneticists (USDA-ARS) located at Mississippi State, MS
and Tifton, GA have screened maize for leaf-feeding resistance to the FAW for many
years. Field screening techniques have been developed that have resulted in identifi-
cation, development, and release of FAW leaf-feeding resistant maize germplasm
(Davis et al. 1989, Williams & Davis 1989, and Widstrom et al. 1993). Occasionally, re-
sults from our screening experiments have been poorer than expected or desired.
When this happens, the research program suffers from loss of researchers) time,
funds, seed, and progress. The researchers at Tifton have expressed concern over the
virulence of their FAW culture since infusion of wild genes is not routinely done. Also,
they have experienced poor establishment of FAW on both susceptible and resistant
genotypes when the screening experiments were planted late in the growing season.
Both research groups have observed that smaller plants are often more heavily dam-
aged than larger plants of the same genotype. In 1993, the researchers at Mississippi
State experienced a poor screening year with the FAW. Possible causes were insect
colony, limited insect supply, and procedures used to process eggs and larvae for field
infesting or some combination of these factors. Environment is also a possible factor,
but is not under the control of the researcherss. The present study was conducted to
determine what effects insect colony, date of planting (early vs late), and plant growth
stage when infested may have on screening maize for FAW leaf-feeding resistance.
MATERIALS AND METHODS
Field experiments were conducted at Mississippi State, MS and Tifton, GA during
the 1995 growing season.
September, 1996
Armyworm Symposium '96: Davis et al.
Insect Colony. At both Tifton and Mississippi State, screening for leaf-feeding re-
sistance requires that each plant within the experiments be infested with approxi-
mately the same number of neonate FAW to provide a uniform infestation of the pest.
Neonates for infesting were obtained from laboratory colonies. Eggs from the Missis-
sippi colony were shipped by overnight mail to Tifton when plants there reached the
appropriate growth stage and vice versa for Georgia colony eggs to Mississippi. Per-
sonnel at each location have developed their own procedures for laboratory rearing of
FAW (Burton & Perkins 1989, Davis 1989). Their procedures differ primarily in larval
diets, handling of adults and eggs, and genetic maintenance of the laboratory colonies.
The Georgia colony has been maintained on a modified pinto bean diet for about six
years in the laboratory without any infusion of wild genes (personal communication:
W. D. Perkins), whereas, the Mississippi colony, which is reared on a wheatgerm-
casein diet, has an infusion of wild genes almost every year.
Planting Dates. Researchers at both locations have utilized early and late plant-
ings for screening. Sometimes inclement weather dictates later plantings and, on oc-
casion early and a late plantings of the same maize genotypes are used so that
experiments can be repeated within the same growing season.
In our experiments, there were two planting dates within each of the early and late
planting periods. This was done so that two whorl stages of growth would be available
at the time of infestation. Planting dates were as follows: early Georgia (Apr. 5 and 19),
Mississippi (Apr. 6 and 20), late Georgia (May 2 and 10), Mississippi (May 3 and 17).
Plant Growth Stages. Entomologists and plant geneticists normally select a pre-
ferred growth stage to infest when screening for resistance. At Tifton, the researchers
have selected the V4 stage, whereas, at Mississippi State the preferred stage is from
V, (based on the system devised by Ritchie & Hanway [1982] for identifying stages
of maize development). Sometimes, infestation is not possible at the desired plant
stage for various reasons. For example, if insect supplies are inadequate when plants
are ready, the plants will have grown to advanced stages before infestation can occur.
The infestations for the early and late planting periods were made when the plants
of the second planting within each period reached the V, stage at Mississippi State
and the V4 stage at Tifton. Growth stage of the plants of the first planting within each
planting period was recorded the day of infestation.
Maize Hybrids. Two susceptible and two resistant single cross maize hybrids were
selected to study the above factors and their effects on identifying different levels of
susceptibility. The susceptible hybrids were 'AB24E x Mp305' and 'SC229 x Tx601',
and the resistant hybrids were 'Mp496 x Mp708' and 'Mp704 x Mp707'.
Experimental Design. The experiments were conducted using a randomized com-
plete block design with treatments in a factorial arrangement with six replications.
Treatments were 2 insect cultures, early and late plantings, 2 plant growth stages,
and 4 single cross maize hybrids. The treatments were represented in each replication
by a single plot of maize 6.1 m long and 0.9 m apart. Plots were thinned to approxi-
mately 20 plants before infestation. Accepted recommended agronomic practices were
followed at both locations.
Infestation Technique. Each plant was infested with 30 neonate FAW mixed in 20/
40 size corn cob grits and delivered into each whorl by a hand-held plastic device
(Mihm 1983; Davis et al. 1989) when the plants of the second planting of each plant-
ing period (early or late) reached the specified growth stage. Plants of both plantings
within each planting period were infested on the same day. The infestation dates were
as follows: Mississippi State early period plantings on May 25 and late period plant-
ings on June 16; Tifton early period plantings on May 12 and late period plantings
on June 6.
Florida Entomologist 79(3)
Data. Each plant within a plot was visually scored for the degree of leaf-feeding
damage on the 7th and 14th day after infestation. The rating systems used were the
ones developed by Davis et al. (1992) specifically for the FAW. Both 7- and 14-day rat-
ing scales are based on the type and number of feeding lesions present on the leaves.
The scores range from 0 to 9 with 0 indicating no damage and 9 indicating heavy dam-
age. The two rating scales are highly correlated in their ability to separate resistant
from susceptible genotypes. The 7-day scale minimizes the effect of leaf damage
caused by migrating mid- instar FAW larvae, whereas, the 14-day scale gives the re-
searcher a view of the total feeding damage caused by the larvae and provides an op-
portunity to determine if the plants show any late resistance response to the insect
that would not be apparent at 7 days after infestation.
StatisticalAnalyses. Plot means were used in the analyses. Experiments were an-
alyzed by location because of the differences in growth stage at the time of infestation.
Data from the early and late planting periods were combined by the day (7 or 14) in
which leaf-feeding damage scores were taken and analyzed by use of ANOVA (SAS
1989). Means were separated by least significant difference test at the 5% level of
probability.
RESULTS
Mississippi State
An error was made when infesting the plants of the early planting period that re-
sulted in only 3 of the 6 replications being used in the statistical analyses. Three fac-
tor interactions were significant for the analyses of both 7- and 14-day leaf-feeding
damage rating scores. Therefore, inferences have been made utilizing nonmarginal
means.
Insect Colony. Differences between FAW colonies for the degree of leaf-feeding
damage were determined by comparing means within planting date periods, within
growth stages infested, and within and among hybrids. Differences in 7 days rating
scores between colonies occurred only within the susceptible hybrid 'Ab24E x Mp305'
for the early planting period (Table 1). In this case, the Mississippi colony produced
significantly more damage (at both leaf stages) than the Georgia colony. In the late
planting period, significant differences in leaf damage scores occurred for 6 of the 8
comparisons between colonies within hybrids and growth stages. Each time, the Mis-
sissippi colony produced more damage than the Georgia colony. However, both colo-
nies ranked the hybrids somewhat similarly when comparisons were made between
cultures within growth stages and planting dates (Table 1). Similar results were ob-
tained between colonies when 14-day rating scores were analyzed (Table 2).
Planting Date. The effect of planting date periods on screening results was deter-
mined by comparing means from 7- and 14-day ratings within growth stages at infes-
tation, within insect colonies, and within and among hybrids. When comparisons were
made between mean 7-day leaf-feeding damage scores of early and late planting peri-
ods (within the V, stage and within insect colonies and hybrids), only 1 of the 8 com-
parisons showed a significant difference (Table 1). However, when the same
comparisons were made for plants infested in the V,, stage, 5 of the 8 comparisons
were significantly different. The rating scores were higher for the early planting pe-
riod. Comparisons among hybrids (between early and late planting periods, within
growth stage and insect colony) revealed that the separation of susceptible from re-
sistant hybrids was basically the same for both planting periods.
September, 1996
Armyworm Symposium '96: Davis et al.
+1 +1
o
+1 +1
+1 +1
0o
+1 +1
+1 +1
00 L
x x
d co
cl c-
ra a
bo ao a
PT P
r- M- II
c4^ ct cc
a a a
a a a
'' ^ '*
>> '
'" '"
L Sr
Bt t
I^l lf
+^ +^ +
*? m
cc c c
ba a a
FI FI I
(U U (
g g
be be be
FI I I
0 0
gi gi ei
cS S c
cc c c
rt t r
0 0
VI I V
~ hci Sc
Florida Entomologist 79(3)
September, 1996
0
II
Ma II
bt fl
o a
c*5 *s
a
F- F-
a a
C +
a ab
rt rt
o '
'i? 'i?
a a
g g
0 0
a a
Armyworm Symposium '96: Davis et al.
The results of the 14 day ratings were similar to those obtained for the 7-day rat-
ings (Table 2). Significant differences between early and late planting periods (within
hybrids, insect colonies, and growth stages) occurred only when the plants were in the
V10 leaf stage at infestation. Again, plants of the early planting period suffered more
damage than those of the late planting period.
Plant Growth Stage. The effect of plant growth stage on screening results (damage
rating scores) was determined by comparing means from 7- and 14-day ratings within
planting date periods, within insect colonies, and within and among hybrids. Regard-
less of planting date period or insect colony, all comparisons between growth stages
within hybrids were significantly different for the 7 day ratings (Table 1). In each
case, the V,, plants sustained less damage than the V, plants. Comparisons among hy-
brids within growth stage, planting date period, and insect colony revealed that resis-
tant hybrids were separated from susceptible hybrids regardless of growth stage at
infestation.
When the same comparisons were made for 14-day ratings (Table 2), only 1 of 8
comparisons were significant during the early planting period, whereas all compari-
sons in the late planting period showed significant differences between growth stages.
Again, the V,, plants suffered less damage than the V, plants.
In general, the susceptible and resistant hybrids separated out as expected when
comparisons were made between growth stages within insect colony and planting
date period. However, the separation, especially of 'SC229 x Tx601' and 'Mp496 x
Mp708', was not as clearcut as occurred for 7-day ratings.
Tifton
Interactions among factors were also significant at Tifton. There was a significant
planting date period x insect colony interaction for the 7-day rating scores and a sig-
nificant planting date period x hybrid x insect colony interaction for the 14-day rating
scores. Since these interactions were encountered, our inferences were made using
nonmarginal means. The same comparisons as described for interpreting the Missis-
sippi State data were used to study the effects that insect colony, planting date period,
and plant growth stage may have on screening results.
Insect Colony. Only 3 of 16 comparisons between insect colonies (within planting
date period and plant growth stage) were significant when the 7-day rating data were
analyzed (Table 3). These significant comparisons were observed only in the early
planting period and involved only the resistant hybrids. In each case, the Mississippi
colony caused more damage than the Georgia colony. The hybrids were separated as
to level of susceptibility similarly by either insect colony.
Results similar to those found for 7-day ratings occurred when the 14-day ratings
were analyzed (Table 4). Significant differences between cultures occurred only in the
early planting period and within the resistant hybrids. Again, the Mississippi colony
caused more damage than the Georgia colony.
No differences were observed between insect colonies for separation of susceptible
from resistant hybrids in the late planting period. However, a few differences between
insect colonies did occur in the early planting period. For example, when plants were
infested in the V4 stage, there were no significant differences in rating scores among
the hybrids infested with the Mississippi colony, whereas hybrids infested at the same
growth stage with the Georgia colony separated out according to the expected resis-
tant and susceptible categories.
Planting Date. Only 2 of 16 comparisons within insect colonies, plant growth
stages, and hybrids showed significant differences between planting date periods
when 7-day ratings were analyzed (Table 3). Both occurred within the resistant hy-
Florida Entomologist 79(3)
p
0 c-
,co
us a
September, 1996
Armyworm Symposium '96: Davis et al.
p:
m
w
Fr;
m
a
W
Fr;
F
o
m
w~
S~
m +I
OIX
$A
O
00
E
F
Fr;
w
a
5
Fr;
0O
o
M
,i
bu
FIW
WU
Wr~
c-lm
i
Y
00 x
00
XX
a a.
Florida Entomologist 79(3)
brids with lower rating scores being given to the plants of the late planting. Generally,
the 7-day ratings from early and late planting periods resulted in no differences being
observed as to separation of hybrids into resistant and susceptible categories.
Four of the 16 comparisons between planting date periods were significant when
14-day ratings were analyzed (Table 4). The differences all occurred within the resis-
tant hybrids. As with the 7-day ratings, lower rating scores were attained in the late
planting period plants. Planting date did not appreciably affect the separation of hy-
brids into susceptible and resistant categories (Table 4).
Plant Growth Stage. When comparisons were made between V, and V, growth stages
within insect colonies, planting date periods, and hybrids (7-day ratings), all but one of
the comparisons were significant (Table 3). The V4 stage plants at infestation received
higher rating scores than those infested at the V, stage. There were some differences
that occurred at the 7-day rating as to separation of hybrids into resistant and suscep-
tible categories. These differences involved the resistant hybrid 'Mp496 x Mp708' when
infested primarily at the V, stage. In these cases, no significant differences in rating
scores were detected between this hybrid and the two susceptible hybrids.
Fewer differences between growth stages were detected when the 14-day ratings
were analyzed (Table 4). The only significant differences were found within the resis-
tant hybrid 'Mp704 x Mp707'. This difference was significant regardless of planting
date or insect colony. In each case, the V, stage plants sustained less feeding damage.
The separation of resistant from susceptible hybrids was not as clearcut when plants
were infested at the V, stage than when infested at the V, stage.
DISCUSSION
Concern over the effect that continuous laboratory rearing of an insect on an arti-
ficial diet can have on field screening for resistance has been expressed by many plant
resistance researchers. Guthrie et al. (1974) found that European corn borer, Ostrinia
nubilalis (Hiibner), laboratory colonies of up to one year in age (approximately 12 gen-
erations) could be used to successfully screen maize for resistance to this pest. How-
ever, European corn borer colonies maintained in the laboratory for longer periods of
time were observed to begin losing their virulence. By the 46th laboratory generation,
the borer larvae were found to cause little damage to field grown maize plants and
were deemed to be no longer usable for screening purposes.
Quisenberry & Whitford (1988) observed differences in damage to bermudagrass
cultivars depending upon the strain of FAW (bermudagrass/rice strain vs the maize
strain) and the diet used to maintain the laboratory colonies. They suggested that
FAW strain and laboratory diet be considerations when screening for bermudagrass
resistance.
In our study, some significant differences in damage rating scores within hybrids
were detected between FAW colonies. When this occurred, the Mississippi colony
(with routine infusion of wild genes) was always found to cause more leaf damage
than the Georgia colony. Both laboratory colonies were capable of causing sufficient
damage so that resistant genotypes could be separated from susceptible genotypes.
Our results did show that a FAW colony can be maintained in the laboratory on an ar-
tificial diet for at least 6 years (approximately 72 generations) without losing its abil-
ity to cause sufficient leaf damage to separate maize genotypes appropriately as to
susceptibility level. However, the Georgia colony appears to have lost some of its vir-
ulence when compared to the Mississippi colony.
Also, our results showed that planting date (early vs late) was not a factor at either
location in affecting the successful separation of resistant from susceptible genotypes.
September, 1996
Armyworm Symposium '96: Davis et al.
Therefore, a researcher could expect similar results if a screening experiment was re-
peated in the same year.
Williams et al. (1983) suggested that resistance mechanisms operating in FAW re-
sistant maize plants may have been more strongly expressed at the 10-leaf stage than
the 5-leaf stage. Videla et al. (1992) reported that FAW resistant maize was resistant
to this pest throughout the vegetative growth stages based on larval growth and sur-
vival. However, no consistent differences between larval growth or survival were
found among plant growth stages within the test single cross maize hybrids. Wiseman
& Isenhour (1993) observed that commercial maize hybrids in the 12-leaf stage, when
infested with neonate FAW, tolerated damage by this pest better than plants at the 8-
leaf stage at infestation.
Our results, especially the 7-day ratings at both locations (Tables 1 and 3), show
that FAW feeding damage occurred less on the older than on the younger plants for
both resistant and susceptible hybrids. In general, this difference in growth stage sus-
ceptibility did not alter the separation of resistant from susceptible hybrids. However,
the best separations among hybrids seemed to occur when the plants were V, stage
when infested. Researchers should be aware that rating scores within experiments
could reflect differences in plant stage because of dissimilar growth rates among gen-
otypes. Also, plants of different sizes within a plot caused by variation in germination
can result in a somewhat inaccurate rating score. When thinning or infesting plants
within a plot, plants not in the appropriate stage should be eliminated. These data on
differences in rating scores between plant growth stages confirm the suggestion by
Williams et al. (1983) concerning the increased expression of resistant mechanisms in
V, and older stage plants. Future research on elucidating the factors responsible for
this resistance should consider comparing magnitude of factor expression in plants of
different growth stages.
In conclusion, significant differences in rating scores within each factor (insect col-
ony, planting date period, and plant growth stage) were found for some comparisons.
However, none of these factors (at least in this study) were found to appreciably alter
the separation of resistant from susceptible genotypes, which is the objective of
screening.
When a screening failure occurs, it is important to determine, if possible, the factor
or factors responsible. Under other circumstances, the three factors studied here may
play a role. Since these data do not present a clear picture on the effect of insect colo-
nization on the ability of resulting FAW larvae to damage maize plants, we suggest a
regular infusion of wild genes into the laboratory colonies. Other factors should also
be investigated, such as breakdowns in the proper procedures for handling eggs and
neonate larvae for field infesting. We consider one of the best ways to avoid screening
failures is to have an ample supply of healthy, highly virulent larvae at the time the
plants reach the desired growth stage. Further, it is recommended that plants be in-
fested at or near the V, stage for the best separation of test genotypes for leaf-feeding
damage.
ACKNOWLEDGMENTS
We thank Johnny Skinner, Charles Mullis, and Mitchell Cook of the Insect Biology
and Population Management Research Laboratory and Thomas Oswalt and Susan
Wolf of the Crop Science Research Laboratory for their technical support in this study.
Also, appreciation is extended to our secretary, Edna Carraway, for preparation of this
manuscript. This article is a contribution of the Crop Science Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture in cooperation with
Florida Entomologist 79(3)
the Mississippi Agricultural and Forestry Experiment Station. It is published with
approval of both agencies as Journal No. 8914 of the Mississippi Agricultural and For-
estry Experiment Station.
REFERENCES CITED
ASHLEY, T. R., B. R. WISEMAN, F. M. DAVIS, AND K. L. ANDREWS. 1989. The fall army-
worm: A bibliography. Florida Entomologist 72: 150-202.
BURTON, R. L., AND W. D. PERKINS. 1989. Rearing the corn earworm and fall army-
worm for maize resistance studies, p. 37-45 in Toward insect resistant maize for
the third world: Proceedings of the international symposium on methodologies
for developing host plant resistance to maize insects. Mexico, D.F.: CIMMYT.
DAVIS, F. M. 1989. Rearing the southwestern corn borer and fall armyworm at Mis-
sissippi State, p. 27-36 in Toward insect resistant maize for the third world:
Proceedings of the international symposium on methodologies for developing
host plant resistance to maize insects. Mexico, D.F.: CIMMYT.
DAVIS, F. M., W. P. WILLIAMS, AND B. R. WISEMAN. 1989. Methods used to screen
maize for and to determine mechanisms of resistance to the southwestern corn
borer and fall armyworm, p. 101-108 in Toward insect resistant maize for the
third world: Proceedings of the international symposium on methodologies for
developing host plant resistance to maize insects. Mexico, D.F.: CIMMYT.
DAVIS, F. M., S. S. NG, AND W. P. WILLIAMS. 1992. Visual rating scales for screening
whorl-stage corn for resistance to fall armyworm. Mississippi Agric. and For-
estry Exp. Stn. Tech. Bull. 186. 9 pp.
GUTHRIE, W. D., Y. S. RATHONE, D. F. COX, AND G. L. REED. 1974. European corn
borer: Virulence on corn plants of larvae reared for different generations on a
meridic diet. J. Econ. Entomol. 67: 605-606.
MIHM, J. A. 1983. Techniques for efficient mass rearing and infestation of fall army-
worm, Spodoptera frugiperda (J. E. Smith), for host plant resistance studies.
CIMMYT, Mexico.
QUISENBERRY, S. S., AND F. WHITFORD. 1988. Evaluation of bermudagrass resistance
to fall armyworm (Lepidoptera: Noctuidae) influence of host strain and dietary
conditioning. J. Econ. Entomol. 81: 1463-1468.
RITCHIE, S. L., AND J. J. HANWAY. 1982. How a corn plant develops. Iowa State Uni-
versity Crop Ext. Serv. Spec. Rpt. 48. 21 pp.
SAS INSTITUTE. 1989. SAS/STAT users guide version 6, fourth edition, Vol. 1. SAS In-
stitute, Cary, N.C.
SCOTT, G. E., F. M. DAVIS, G. L. BELAND, W. P. WILLIAMS, AND S. B. KING. 1977. Host-
plant resistance is necessary for late-planted corn. Mississippi Agric. and For-
estry Exp. Stn. Res. Rpt. 13. 4 pp.
VIDELA, G. L., F. M. DAVIS, W. P. WILLIAMS, AND S. S. NG. 1992. Fall armyworm (Lep-
idoptera: Noctuidae) larval growth and survivorship on susceptible and resis-
tant corn at different vegetative growth stages. J. Econ. Entomol. 85: 2486-
2491.
WIDSTROM, N. W., W. P. WILLIAMS, B. R. WISEMAN, AND F. M. DAVIS. 1993. Registra-
tion of GT- FAWCC(5) maize germplasm. Crop Sci. 33: 1422.
WILLIAMS, W. P., AND F. M. DAVIS. 1989. Breeding for resistance in maize to south-
western corn borer and fall armyworm, p. 207-210 in Toward insect resistant
maize for the third world: Proceedings of the international symposium on
methodologies for developing host plant resistance to maize insects. Mexico,
D.F.: CIMMYT.
WILLIAMS, W. P., F. M. DAVIS, AND B. R. WISEMAN. 1983. Fall armyworm resistance
in corn and its suppression of larval survival and growth. Agron. J. 75: 831-832.
WISEMAN, B. R., AND D. J. ISENHOUR. 1993. Response of four commercial corn hybrids
to infestations of fall armyworm and corn earworm (Lepidoptera: Noctuidae).
Florida Entomologist 76: 283-292.
September, 1996
Armyworm Symposium '96: Wiseman et al.
RESISTANCE OF A MAIZE POPULATION, FAWCC(C5), TO FALL
ARMYWORM LARVAE (LEPIDOPTERA: NOCTUIDAE)
B. R. WISEMAN, F. M. DAVIS', W. P. WILLIAMS' AND N. W. WIDSTROM
USDA, ARS, IBPMRL
Tifton, GA 31793
'USDA, ARS, Plant Science Research Laboratory, Mississippi State, MS. 39762
ABSTRACT
Field tests at Mississippi State, MS, and Tifton, GA, were conducted to evaluate
the effect of resistance of a maize, Zea mays (L.), germplasm population, 'GT-
FAWCC(C5)', to feeding by larvae of the fall armyworm, Spodoptera frugiperda (J. E.
Smith). Plants of selected maize entries were infested at the 8 and 12 leaf stage with
two applications of 15 larvae per plant. Resistance traits measured were leaf damage
at 7 and 14 days after infestation and number and weight of surviving larvae per
plant at 7 and 10 days after infestation. Leaf damage ratings at both 7 and 14 days
after infestation and the number and weight of surviving larvae per plant on GT-
FAWCC(5) at 7 and 10 days after infestation on GT-FAWCC(C5) equalled the number
and weight of surviving larvae on 'MpSWCB-4', the resistant check. Both the resis-
tant check and GT-FAWCC(C5) were significantly more resistant to whorl damage
than the susceptible check,'Ab24E x SC229', for all resistance traits. It is evident that
antibiosis (low weight) and nonpreference (fewer larvae per plant and fewer larvae
preferring leaf samples) mechanisms of resistance are present in the GT-FAWCC(C5)
population as well as for MpSWCB-4.
Key Words: Leaf-feeding resistance, antibiosis, nonpreference, artificial infestations
RESUME
Fueron realizados studios de campo en el estado de Mississippi y en Tifton, Geor-
gia, para evaluar la resistencia de una poblaci6n de germoplasma de maiz, Zea mays
"GT-FAWCC(C5)" al daho producido por larvas de Spodoptera frugiperda (J. E.
Smith). Plantas de entradas seleccionadas de maiz fueron infestadas en el estado de
8-12 hojas con dos aplicaciones de 15 larvas por plant. Las variables de resistencia
medidas fueron el dano a las hojas y el numero y peso de las larvas sobrevivientes por
plant a los 7 y 14 dias despu6s de la infestaci6n. El dano foliar a los 7 y 14 dias des-
pu6s de la infestaci6n y el numero y peso de las larvas sobrevivientes por plant en
GT-FAWCC(C5) a los 7 y 10 dias despu6s de la infestaci6n igualaron al numero y peso
de las larvas sobrevivientes en "MpSWCB-4", el testigo resistente. Tanto el testigo re-
sistente como el GTT-FAWCC(C5) fueron significativamente mas resistentes al daio
del cogollo que el testigo susceptible, "Ab24E x SC229". Es evidence que los mecanis-
mos de resistencia mediante antibiosis (bajo peso) y no preferencia (pocas larvas por
plant y pocas larvas prefiriendo muestras de hojas) estan presents en las poblacio-
nes de GT-FAWCC(C5) y MpSWCB-4.
Maize, Zea mays (L.), is a crop upon which fall armyworm (FAW), Spodoptera fru-
giperda (J. E. Smith), infestations often reach devastating levels in the southeastern
United States. In 1975, losses in Georgia were estimated at over 20 million dollars
Florida Entomologist 79(3)
(Sparks 1979). Yield losses attributed to the fall armyworm for the U. S. have been es-
timated at 2% annually (Wiseman & Morrison 1981). Wiseman & Isenhour (1993)
showed that commercial maize hybrids suffered greater yield losses (32.4%) when
manually infested with 2 applications of 20 neonates per plant at the 8-leaf stage than
at the 12-leaf stage (15.4%).
Maize is the most valuable field cereal crop in the U. S. (Anonymous 1995). Al-
though the number of acres planted with maize and the value of the crop fluctuate an-
nually, 72.9 million acres were harvested in 1994 in the U. S. with a production of 10.1
billion bushels and a value of 22.2 billion dollars (Anonymous, 1995). Thus, the con-
tinued development and release of new maize germplasm with resistance to the FAW
is important. Understanding the mechanisms of resistance or the biological effects of
resistant cultivars on the FAW is important in managing the insect pest. Therefore,
the objective of this study was to determine leaf damage at 7 and 14 days after infes-
tation (DAI) and biological responses of FAW when feeding on a newly released resis-
tant germplasm population, 'GT-FAWCC(C5)' (Widstrom et al. 1993).
MATERIALS AND METHODS
Two dent maize entries, (resistant 'MpSWCB-4' or susceptible 'Ab24E x SC229'),
were selected for comparison with GT-FAWCC(C5) because of their response to FAW
in field tests (Scott & Davis 1981; Widstrom et al. 1993). The three entries were seeded
in separate experiments on 5 April, 1995 at Tifton, GA and on 17 April, 1995 at Mis-
sissippi State, MS. Test plots consisted of 3 rows 6.1 m long and 0.9 m apart. Plants
were thinned to about 30 cm apart. Recommended agronomic practices were followed
at both locations.
A split plot design with 6 replications was used with whole plots being leaf-stage
at the time of infestation: 8-leaf stage (V4 at Tifton and V8 at Mississippi State) or 12-
leaf stage (V8 at Tifton and V12 at Mississippi State) (Ritchie & Hanway 1982). Sub-
plots consisted of maize entries. At Tifton, leaf stage was determined by counting the
total number of emerged leaves, and at Mississippi State, leaf stage was determined
by counting only leaves with their collars exposed. Whole plots were bordered with 2
rows of a commercial maize hybrid. Each plant within the 8- or 12-leaf stage treat-
ments was infested with 2 applications of 15 neonate FAW on the same day (Wiseman
1989). By infesting with two applications, noninfested plants within the experiments
do not exist. From row one and two of each sub-plot, 10 plants were harvested at soil
level at both 7 and 10 DAI and taken to the laboratory to determine the number and
weight of surviving larvae per plant. Plants on the third row of each subplot were
rated at 7 and 14 DAI using a visual rating scale of 0-9 (Davis 1992) where 0 = no dam-
age and 9 = whorl destroyed.
FAW larvae used to infest plants were obtained from colonies maintained at the
Insect Biology and Population Management Research Laboratory, Tifton, GA and the
Crop Science Research Laboratory, Mississippi State, MS, respectively (Perkins 1979;
Davis 1989).
A laboratory test was designed to determine nonpreference for leaf samples of the
three maize entries. Whorl samples were obtained from the 8- leaf stage at Missis-
sippi State on May 25, 1995, taken to the laboratory and excised into disks of about 2
cm diam (Davis et al. 1989). The leaf disks of each entry were randomly placed on the
outer inside edge of a large dish (15.3 cm diam, 2.7 cm deep). One FAW egg mass in
the blackhead stage containing about 50 eggs was placed in the center of each dish
and the dish was placed in darkness. The experiment was arranged as a randomized
complete block with 22 replications. The experiment was held in a controlled environ-
September, 1996
Armyworm Symposium '96: Wiseman et al.
ment room maintained at 27 C and 50-60% RH. Total number of larvae found on each
maize entry leaf section (if larvae were not actually on the leaf section at the time of
recording, they were not counted) per dish was recorded 24 h after egg hatch.
Damage ratings, the number and weight of surviving larvae and number of larvae
preferring a leaf sample were analyzed by PROC GLM (SAS Institute 1989). When
significant differences were indicated, means were separated by least significant dif-
ferences (LSD) at P = 0.05 (SAS Institute 1989).
RESULTS AND DISCUSSION
A significant (P < 0.05) leaf-stage x entry interaction was found for the 7-day leaf
damage ratings at Mississippi State, but not at Tifton (Table 1). Plants from all three
entries at Mississippi State and two (Ab24E x SC229 and MpSWCB-4) at Tifton re-
ceived more damage at the 8-leaf stage than at the 12-leaf stage. Both, GT-
FAWCC(C5) and MpSWCB-4 had significantly (P < 0.05) less damage 7 DAI in the 8-
and 12-leaf stages than Ab24E x SC229 at both locations. GT-FAWCC(C5) was signif-
icantly less damaged at the 8-leaf stage than MpSWCB-4 at Tifton.
A significant (P < 0.05) leaf-stage x entry interaction also was found for leaf dam-
age ratings 14 DAI at Mississippi State, but not at Tifton (Table 1). At Tifton, plants
of all entries infested at the 8-leaf stage yielded significantly higher damage ratings
(8.2 vs 4.7) than plants of all entries at the 12-leaf stage. Plants from all three entries
at both locations received more damage at the 8-leaf stage than at the 12-leaf stage.
Both GT-FAWCC(C5) and MpSWCB-4 were damaged significantly (P < 0.05) less 14
DAI in the 8- and 12-leaf stages than Ab24E x SC229 at Mississippi State and Tifton.
MpSWCB-4 was damaged significantly less than GT-FAWCC(C5) at Mississippi State
at the 8-leaf stage but not at Tifton.
A significant (P < 0.05) leaf-stage x entry interaction was found for the number of
larvae recovered per plant 7 DAI at each location (Table 2). Significant differences in
the number of larvae per plant were found 7 DAI between the 8- and 12-leaf stages for
each entry at each location. In each case, plants infested at the 12-leaf stage had fewer
larvae 7 DAI than plants infested at the 8-leaf stage of each entry. Both GT-
FAWCC(C5) and MpSWCB-4 had significantly fewer larvae per plant than the sus-
ceptible check, Ab24E x SC229, at both leaf stages and locations.
A significant (P < 0.05) interaction (leaf-stage x entry) was found for number of lar-
vae recovered per plant 10 DAI at Tifton, but not at Mississippi State (Table 2). The
mean number of larvae recovered 10 DAI at Mississippi State at the 8-leaf stage was
significantly higher (4.7 vs 3.0) than the mean number recovered at the 12-leaf stage.
Significant differences between the number of larvae per plant 10 DAI at the 8- and
12-leaf stages for each entry occurred at Mississippi State, but differences between
plant growth stages were noted only for GT-FAWCC(C5) at the Tifton location. GT-
FAWCC(C5) and MpSWCB-4 had significantly fewer larvae per plant than the sus-
ceptible check, Ab24E x SC229, at the 8-and 12-leaf stages at Mississippi State, these
differences were noted only for the 12-leaf stage at Tifton.
Total weight of larvae per plant recovered 7 DAI from plants of entries infested at
the 8-leaf stage at Tifton was significantly (P < 0.05) greater than weight of larvae re-
covered from plants of entries infested at the 12-leaf stage (Table 3). No differences in
weight of larvae were found between the 8- and 12-leaf stages at Mississippi State.
Larvae feeding on and collected from 8- or 12-leaf stage plants of GT-FAWCC(C5) and
MpSWCB-4 weighed significantly less than larvae feeding on and collected from
Ab24E x SC229 at both locations.
Total weight of larvae per plant recovered 10 DAI from plants of entries infested
at the 8-leaf stage at Mississippi State (164 vs 79) was significantly (P < 0.05) heavier
Florida Entomologist 79(3)
NN00CQ
00 Q0
Q0 0
L-i i o
c o -
cN o
00CO ^0
r t u c
1^ Crl 1 N
CO CO CO
-s-PscOs
'3i Q ^-
c 0 c<]d -
c~ ir T<
September, 1996
a
o
II
m
oj
Sci
5-,
I
B
B
00
S
a
a
OS
8oj
r
"3
&c~
a
trjF
>i
a*t
is?
,,
Armyworm Symposium '96: Wiseman et al.
ito co e
4e5e66
ici -i
t &1&1
c~ -N -
t Nhj
ci -t -
t N~
-c -i ,
'p N
t Nl
Cl
Florida Entomologist 79(3)
0 e^^O
-CpC
rt ^ u c--
N- 0 0
S&m 00
tMr
eh: er &
o to e0 to
c M M to&
qC
ct -3 -3 tq
'u to j
00 N^ CTS e
to &Cl CO I
c-(
p N 0
l-u -I C
^' &
^-(
"3 p i 0
^ cTS cdS
C~l f-
t-lm
September, 1996
S
I
a
S
ic
0.a
F
-&0
Armyworm Symposium '96: Wiseman et al.
than weight of larvae recovered from plants of entries infested at the 12-leaf stage,
but not at Tifton (Table 3). Weight of larvae per plant was significantly more for larvae
feeding on, and collected from, Ab24E x SC229 from the 8-leaf stage at Mississippi
State than from the 12-leaf stage. The total weight of larvae per plant or weight per
larva from both GT-FAWCC(C5) and MpSWCB-4 was significantly less 10 DAI than
the weight of larvae from the susceptible check (Ab24E x SC229) for both the 8- and
12-leaf stages and locations.
Results from the laboratory nonpreference study indicated that leaf sections of GT-
FAWCC(C5) were less preferred by FAW neonate than those for Ab24Ex SC229 [Ab24E
x SC229 15.46a larvae per dish; MpSWCB-4 10.09ab larvae per dish and GT-
FAWCC(C5) 7.96b larvae per dish; number of larvae followed by the same letter are not
significantly different P < 0.05]. MpSWCB-4 and AB24E x SC229 were equally pre-
ferred as was MpSWCB-4 and GT-FAWCC(C5). Wiseman et al. (1981) reported that
FAW larvae preferred MpSWCB-4 similar to the susceptible check but possessed higher
antibiosis than the susceptible check and that 'Antigua 2D-118' was more nonpreferred
by FAW larvae than MpSWCB-4. This appears to be the case here for GT-FAWCC(C5)
except that both MpSWCB-4 and GT-FAWCC(C5) are similar in antibiosis.
In summary, though the 8- and 12- leaf stages at Mississippi State and Tifton were
measured differently, the leaf damage ratings 7 and 14 DAI, number and weight of
larvae per plant at 7 and 10 DAI at both locations for GT-FAWCC(C5) equalled those
of MpSWCB-4, the resistant check. MpSWCB-4 and GT-FAWCC(C5) were signifi-
cantly more resistant than the susceptible check, Ab24E x SC229, when weight per
larva was also calculated. It is evident that antibiosis (low total weight of larvae per
plant or weight per larva) and nonpreference (less establishment per plant and num-
bers preferring leaf samples) mechanisms of resistance are present in the GT-
FAWCC(C5).
ACKNOWLEDGMENTS
We thank Johnny Skinner, Charles Mullis, and Mitchell Cook of the Insect Biology
and Population Management Research Laboratory and Thomas Oswalt and Paul
Buckley of the Crop Science Research Laboratory for their technical support in this
study.
REFERENCES CITED
ANONYMOUS. 1995. The world of corn: Agriculture's innovators. National corn growers
association. National corn development foundation. 39 p.
DAVIS, F. M. 1989. Rearing the southwestern corn borer and fall armyworms at Mis-
sissippi State, pp. 27-36 in Toward insect resistant maize for the third world:
Proceedings of the international symposium on methodologies for developing
host plant resistance to maize insects. Mexico, D. F.: CIMMYT. 327 p.
DAVIS, F. M., S. S. NG, AND W. P. WILLIAMS. 1989. Mechanisms of resistance in corn
to leaf feeding by Southwestern corn borer and European corn borer (Lepi-
doptera: Noctuidae). J. Econ. Entomol. 82: 919-922.
DAVIS, F. M., S. S. NG, AND W. P. WILLIAMS. 1992. Visual rating scales for screening
whorl-stage corn for resistance to fall armyworm. Mississippi Agricultural &
Forestry Experiment Station. Technical Bulletin 186. 9 pp.
PERKINS. W. D. 1979. Laboratory rearing of the fall armyworm. Florida Entomol. 62:
87-91.
RITCHIE, S. W., AND J. J. HANWAY. 1982. How a corn plant develops. Iowa State Uni-
versity Coop. Ext. Ser. Spec. Rpt. 48. 21 pp.
336 Florida Entomologist 79(3) September, 1996
SAS INSTITUTE. 1989. SAS/STAT user's guide version 6, fourth edition, Vol. 1. SAS In-
stitute, Cary, N. C.
SCOTT, G. E., AND F. M. DAVIS. 1981. Registration of MpSWCB-4 population of maize.
(Reg. No. GP 87). Crop Science 21: 148.
SPARKS, A. N. 1979. A review of the biology of the fall armyworm. Florida Entomol. 62:
82-87.
WIDSTROM, N. W., W. P. WILLIAMS, B. R. WISEMAN, AND F. M. DAVIS. 1993. Registra-
tion of GT-FAWCC(C5) maize germplasm. Crop Science 33: 1422.
WISEMAN, B. R. 1989. Technological advances for determining resistance in maize to
insects, pp. 94-100 in Toward insect resistant maize for the third world: Pro-
ceedings of the international symposium on methodologies for developing host
plant resistance to maize insects. Mexico, D. F.: CIMMYT. 327 p.
WISEMAN, B. R., AND D. J. ISENHOUR. 1993. Response of four commercial corn hybrids
to infestations of fall armyworm and corn earworm (Lepidoptera: Noctuidae).
Florida Entomol. 76: 283-292.
WISEMAN, B. R., AND W. P. MORRISON. 1981. Components for management of field
corn and grain sorghum insects and mites in the United States. USDA-ARS
ARM-S-18.18p.
WISEMAN, B. R., W. P. WILLIAMS, AND F. M. DAVIS. 1981. Fall armyworm: Resistance
mechanisms in selected corns. J. Econ. Entomol. 74: 622-624.
Florida Entomologist 79(3)
September, 1996
BEET ARMYWORM (LEPIDOPTERA: NOCTUIDAE) CONTROL
ON COTTON IN LOUISIANA
V. J. MASCARENHAS', B. R. LEONARD', E. BURRIS3 AND J. B. GRAVES'
Louisiana State University Agricultural Center
'Department of Entomology, Baton Rouge, LA 70803;
'Macon Ridge Location of the Northeast Research Station, Winnsboro, LA 71295; and
3Northeast Research Station, St. Joseph, LA 71366.
ABSTRACT
Efficacy of selected labeled and experimental insecticides against beet armyworm,
Spodoptera exigua (Hiibner), populations from Louisiana were determined in both a
laboratory diet bioassay and in replicated field plots. Significantly higher LC0,'s for
chlorpyrifos and thiodicarb were observed for one of two field-collected strains relative
to a laboratory-reference strain in the laboratory diet bioassays. No significant differ-
ences in susceptibility between the reference strain and field-collected strains were
observed for chlorfenapyr (proposed common name), spinosad or tebufenozide. For the
reference strain, LC0,'s (ppm) for tebufenozide, spinosad, chlorfenapyr, chlorpyrifos,
and thiodicarb were 2.6, 2.8, 4.8, 4.9, and 319.8, respectively. In two field tests, all
three experimental insecticides (chlorfenapyr, spinosad, and tebufenozide) as well as
chlorpyrifos significantly reduced the numbers of beet armyworm larvae relative to
the untreated control at all sampling periods (3, 5, 7, and 10 days after treatment), ex-
cept for Test 2 at 3 days after treatment. Thiodicarb provided satisfactory control of
Armyworm Symposium '96: Mascarenhas et al.
larvae in Test 1; however, in Test 2 thiodicarb did not significantly reduce the numbers
of beet armyworm compared with the untreated control. The microbial insecticide
Spod-X provided inadequate larval control in both tests.
Key Words: Beet armyworm, insecticides, insecticide efficacy, insecticide bioassay
RESUME
Fue determinada la eficacia de insecticides registrados y experimentales contra
poblaciones del gusano de la remolacha, Spodoptera exigua (Hiibner). Los ensayos se
realizaron en el laboratorio, con dietas, y en parcelas replicadas en el campo. En los
ensayos de laboratorio fueron observadas CL0, significativamente mas altas para
chlorpyrifos y thiodicarb en una de las dos cepas colectadas en el campo en relaci6n
con una cepa de referencia de laboratorio. No fueron halladas diferencias significati-
vas entire la cepa de referencia y las colectadas en el campo, en cuanto a la suscepti-
bilidad a chlorfenapyr (nombre comun propuesto), spinosad o tebufenozide. Para la
cepa de referencia, las CL0, para tebufenozide, spinosad, chlorfenapyr, chlorpyrifos y
thiodicarb fueron 2.6, 2.8, 4.8, 4.9 y 319.8 ppm, respectivamente. En dos ensayos de
campo, los tres insecticides experimentales (chlorfenapyr, spinosad, y tebufenozide),
asi como chlorpyrifos, redujeron significativamente los ndmeros de larvas del gusano
de la remolacha en relaci6n con el testigo sin tratar en todos los muestreos (3, 5, 7 y
10 dias despues del tratamiento), except en el ensayo 2 a los tres dias despues del tra-
tamiento. Thiodicarb produjo un control satisfactorio de las larvas en el ensayo 1; sin
embargo, en el ensayo 2 thiodicarb no redujo significativamente los ndmeros del gu-
sano de la remolacha comparados con los del testigo no tratado. El insecticide micro-
biano Spod-X produjo un control larval inadecuado en ambas pruebas.
The beet armyworm, Spodoptera exigua (Hiibner), has historically been viewed as
a secondary pest of cotton in most of the southeastern United States. However, popu-
lation outbreaks experienced in the 1980's and early 1990's in Alabama, Georgia, Lou-
isiana, Mississippi (Douce & McPherson 1991, Burris et al. 1994, Layton 1994, Smith
1994), and more recently in Texas (Arrillago 1995) have demonstrated the potential
damage associated with this pest and the ineffective control provided by most cur-
rently labeled insecticides. During the outbreak of 1993, beet armyworms infested
60% of the total cotton acreage in the mid-south and southeastern states, with approx-
imately 35% of this acreage having infestations above an economic level (Smith 1994).
The economic impact of beet armyworm infestations varies from region to region
and includes both the direct yield loss caused by insect injury and the high production
costs associated with frequent and costly insecticide usage. In some regions of the
southeast, such as Alabama and Mississippi, where the beet armyworm outbreak of
1993 was particularly devastating, large cotton acreages were abandoned after the
control costs exceeded $250-370 per ha (Layton 1994, Smith 1994).
The studies reported here were designed to evaluate the susceptibility of field pop-
ulations of beet armyworm to standard and experimental insecticides in laboratory
and field experiments.
MATERIALS AND METHODS
Laboratory and field experiments were conducted to evaluate the efficacy of stan-
dard and experimental insecticides against the beet armyworm. Insecticides tested
Florida Entomologist 79(3)
included commercial formulations of two currently recommended insecticides, thiodi-
carb [Larvin3.2F (flowable powder), Rhone-Poulenc Ag. Co., Research Triangle Park,
North Carolina] and chlorpyrifos [Lorsban 4EC (emulsifiable concentrate), Dow-
Elanco, Indianapolis, Indiana], as well as three experimental compounds,
tebufenozide (Confirm 2F, Rohm & Haas Co., Philadelphia, Pennsylvania), chlor-
fenapyr (Pirate 3F, American Cyanamid Co., Wayne, New Jersey) and spinosad
(Tracer 4F, DowElanco, Indianapolis, Indiana). Spod-X" (Crop Genetics Interna-
tional, Wilmington, Delaware), a NPV (nuclear polyhedrosis virus) product which is
labeled for beet armyworm control in cotton, was included in the field tests.
Diet Bioassay
Susceptibility of a laboratory-reference and two field-collected strains of beet ar-
myworms was evaluated using a surface-treated diet bioassay similar to that de-
scribed by Joyce et al. (1986) and Chandler & Ruberson (1994). Both field strains were
collected from northeast Louisiana, one near Newellton and the other near St. Joseph.
These strains were bioassayed using individuals from the F,-F4 laboratory-reared
generations. Three ml of an artificial wheat germ/pinto bean diet were pipetted into
individual 30 ml diet cups (Schneider Paper Product Inc., New Orleans, Louisiana)
and allowed to cool. Serial dilutions were prepared for each insecticide tested, and 100
1l of each concentration (four concentrations and a water control) were individually
pipetted onto the surface of the diet and allowed to air dry at room temperature for ap-
proximately 1 h. Diet cups were shaken to evenly distribute the insecticide solution
over the diet's surface.
One first instar (approximately 3-d old) beet armyworm larva was placed on the
treated diet and cups were then capped. A minimum of 50 larvae per dose were bioas-
sayed for each insecticide and mortality was observed at 48, 72, 96, and 120 h. Larvae
were considered dead if they did not respond to prodding with a paint brush. Bioas-
says were conducted under constant light at 221C and 405% RH. Data from the
diet bioassay were analysed by probit analysis using POLO-PC (LeOra Software
1987). LC0,'s of field-collected strains were considered to be significantly different
than that of the reference strain if the 95% confidence limits did not overlap. Toxicity
ratios (TR) were calculated by dividing the LC0,of a field strain by that of the reference
strain. Field strains bioassayed were collected from the same region where field stud-
ies were conducted. The reference strain of beet armyworms was obtained from the
USDA-ARS, Southern Insect Management Laboratory (SIML) at Stoneville, Missis-
sippi.
Field Experiments
Field tests were conducted at the Northeast Research Station (Test 1) near St. Jo-
seph, Louisiana and at the Macon Ridge location of the Northeast Research Station
(Test 2) near Winnsboro, Louisiana during the summer of 1995. Both tests were ar-
ranged in a randomized complete block design with 4 replications. Plots measured
four rows (approximately 1 m centers) by 15.25 m. Test 1 was planted to'Stoneville LA
887' cotton on 16 May and Test 2 was planted to'DPL 5690' cotton on 20 June.
Larval densities in each block were estimated before insecticide application by
taking 6-10 drop cloth (approximately 1 row meter each) samples. Treatments for Test
1 and 2 were applied on 15 and 30 August, respectively, with a high clearance sprayer.
In Test 1, the sprayer was calibrated to deliver 93.5 liters total spray volume per ha
through Teejet X-12 hollow cone nozzles (2 per row) at 3.6 kg/cm2. In Test 2, the sprayer
September, 1996
Armyworm Symposium '96: Mascarenhas et al.
was calibrated to deliver 105.5 liters total spray volume per ha through Teejet X-8 hol-
low cone nozzles (2 per row) at 3.1 kg/cm2.
Treatment effect was measured by taking 2 drop cloth samples in each plot and
counting the number of live larvae. Sampling was done in areas within a row where
evidence of a'hit' (recently hatched egg mass) and/or larval feeding was observed. This
sampling procedure was adopted because randomly sampling for a clump-distributed
pest population would not appropriately reflect larval densities in field plots. At each
sampling period [2, 5, 7, and 10 days after treatment (DAT)], one row of each plot was
sampled, so that rows 1, 2, 3, and 4 were sampled at 3, 5, 7, and 10 DAT, respectively.
This sampling pattern was used to avoid sampling an individual'hit', which may have
been disturbed during an earlier sampling period. Recently deposited egg masses
were avoided during the last two sampling dates, because these 'hits' represented in-
festations which would not have received the full treatment effect. With this sampling
approach, neonates through second instar larvae were not included in samples taken
at 7 and 10 DAT. Total number of live larvae per 0.3 m of row was used in the data
analysis. Data were analyzed by ANOVA and means were separated according to Fis-
cher's protected LSD (SAS Institute 1988).
RESULTS
Diet Bioassay
In the thiodicarb bioassays, LC0,'s ranged from 320 to 641 ppm (parts per million)
and a significantly higher LC0,was obtained with larvae of the Newellton strain com-
pared with the SIML reference strain (Table 1). In the chlorpyrifos bioassays, Newel-
lton strain larvae had a significantly higher LC0,than both the St. Joseph and the
reference strain. Furthermore, the toxicity ratio for Newellton strain (TR = 7.1) was
much higher for chlorpyrifos than for all other insecticides. There were no significant
differences in the LC0,'s between either of the field-collected strains and the SIML
strain for all three experimental insecticides. For the tebufenozide bioassays, LC0,'s
for all strains evaluated ranged from 2.6 to 5.5 ppm. A similar range in LC0,'s was ob-
served for spinosad (2.1-4.8 ppm). Slightly higher LC0,'s, which ranged from 4.0 to 6.1
ppm, were obtained in the chlorfenapyr bioassays. Slopes of dosage-mortality lines for
chlorfenapyr were steeper compared with slopes for the other chemicals.
Field Experiments
The average number of beet armyworm larvae per 0.3 m of row in Test 1 and 2
prior to application of the various treatments was 5.1 and 12.5, respectively. In Test
1, numbers of beet armyworm larvae were significantly lower than that of the un-
treated control for all treatments at 3 and 5 DAT, except for Spod-X (Table 2). Similar
results were observed at 7 DAT when all treated plots, except for Spod-X and thiodi-
carb, had fewer live larvae than the untreated control. At the final observation (10
DAT), all treatments had significantly reduced the number of beet armyworm larvae
relative to the untreated control.
In Test 2, no significant differences among treatments were observed at 3 DAT (Ta-
ble 3). At 5 DAT, only the chlorfenapyr and spinosad treatments significantly reduced
the number of beet armyworm larvae compared with the untreated control. All
treated plots, except for thiodicarb and Spod-X, had significantly fewer larvae than
the untreated control at 7 DAT. By 10 DAT, all treatments, except for thiodicarb, had
significantly fewer larvae relative to the untreated control (Table 3).
Florida Entomologist 79(3)
OO l =-1 c- C 1- C =o
L cc d ( cO
't q N er 4 t6
o e] No No to to &
C~ co N r O C
~~~~ii ~iM
N &1< &
M~ 00c
&lodtd
Lc~ c<] c<
00 0t
ccdc
CO n 't
M U5) *
c-S 0 l
44&44
cc
o oo oo oo t o o o o
O to t ~~o ~t o oNt t o0 oNt
-n -
4cc
c~q C
cc c
cc ~4c
mC) C
d~~
September, 1996
I Q
A
i~ E
0I
I~ I
1,1 I
-ll S
"' s' s
g^ na
6^ se
H "3 i- i-l
? ^ 5 ^ e
l-|ll
1 I Is
*^;31
usmm~cous~mm
dihjdihjhjCjd;Q;di
r( r(r(
MCD
Armyworm Symposium '96: Mascarenhas et al.
TABLE 2. EFFICACY OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM AT 3, 5, 7,
AND 10 DAYS AFTER TREATMENT IN TEST 1 AT THE NORTHEAST RESEARCH
STATION, ST. JOSEPH, LOUISIANA.
Numbers of Beet Armyworm Larvae per 30.5 cm
Treatment Rate (kg AI/ha) 3 DAT' 5 DAT 7 DAT 10 DAT
Chlorfenapyr 0.22 0.5 b 0.3 c 0.6 b 0.3 b
Chlorpyrifos 1.12 0.9 b 0.5 c 1.2 b 0.5 b
Spinosad 0.08 0.7 b 0.2 c 0.4 b 0.2 b
Spod-X 185.52 6.1 a 3.7 ab 4.0 a 1.1 b
Tebufenozide 0.14 2.1 b 0.6 c 1.1 b 0.5 b
Thiodicarb 0.67 1.7 b 2.7 b 1.9 ab 0.9 b
Untreated 5.1 a 5.1 a 3.9 a 2.4 a
P>F 0.01 0.01 0.01 0.01
'Means within a column not followed by a common letter are significantly different (P = 0.05; LSD).
2Milliliters formulated material per ha.
DISCUSSION
Variation in susceptibility of field-collected strains of beet armyworm to chlorpyri-
fos have been reported by Chandler & Ruberson (1994); three (Bartow Co., Georgia,
Macon Co., Alabama and Yazoo Co., Mississippi) of seven field strains had signifi-
cantly higher LC0,'s than the SIML reference strain. LC0,'s of field strains reported by
Chandler & Ruberson were 2- to 29-fold higher than the highest LC0, (Newellton, Lou-
isiana) observed in the bioassays reported herein. As with chlorpyrifos, ranges of
LCs,'s reported for thiodicarb by Chandler & Ruberson (1994) were larger than those
observed in bioassays of Louisiana strains. LCs,'s observed for all three experimental
compounds generally were much lower than those observed with the standard insec-
ticides tested (Table 1). Significantly higher LCs,'s observed with the Newellton field
strain in both the chlorpyrifos and thiodicarb bioassays likely are due to the fact that
this strain was collected from an area which typically receives high insecticide inputs.
Although the St. Joseph strain also was collected from a high-input cotton producing
region, the actual collection was made from a field within the Northeast Research Sta-
tion, which typically does not receive intensive insecticide applications.
Susceptibility of the beet armyworms on the Northeast Research Station (both the
St. Joseph and Macon Ridge locations) to standard insecticides was also observed in
the field tests in which chlorpyrifos (Tables 2 and 3) and thiodicarb (Table 2) signifi-
cantly reduced the numbers of beet armyworm larvae relative to the untreated con-
trol. Considerable variation in control of beet armyworms with chlorpyrifos and
thiodicarb is reported in the literature. Smith (1985) reported 65 and 76% control at
3 DAT in Texas with chlorpyrifos and thiodicarb, respectively. In South Carolina, Sul-
livan et al. (1991) reported 90% control with thiodicarb, while in Mississippi Reed et
al. (1994) reported less than 50% larval control with either insecticide. In both field
tests, chlorfenapyr and spinosad provided excellent and rapid beet armyworm control
and performed as well as, or better than, the standard, chlorpyrifos. Efficacy of chlo-
rfenapyr against beet armyworm also has been documented in numerous EUP trials
Florida Entomologist 79(3)
TABLE 3. EFFICACY OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM AT 3, 5, 7,
AND 10 DAT IN TEST 2 AT THE MACON RIDGE RESEARCH STATION, WINNS-
BORO, LOUISIANA.
Numbers of Beet Armyworm Larvae per 30.5 cm1
Treatment Rate (kg AI/ha) 3 DAT' 5 DAT 7 DAT 10 DAT
Chlorfenapyr 0.22 2.8 a 0.6 c 1.1 b 0.4 c
Chlorpyrifos 1.12 4.7 a 3.6 bc 1.2 b 0.3 c
Spinosad 0.08 5.8 a 0.7 c 1.3 b 0.4 c
Spod-X 185.5 10.7 a 7.9 a 3.4 ab 0.7 bc
Tebufenozide 0.14 14.5 a 1.4 bc 0.4 b 0.7 bc
Thiodicarb 0.67 12.6 a 4.4 ab 5.2 a 1.4 ab
Untreated 9.2 a 4.4 ab 5.4 a 1.8 a
P>F 0.41 0.02 0.04 0.05
Means within a column not followed by a common letter are significantly different (P = 0.05; LSD).
Milliliter formulated material per ha.
throughout the southeast (Wiley et al. 1995) where 90-98% control was reported.
Tebufenozide, an insect growth regulator, provided satisfactory control of beet army-
worm. However, this compound generally had a slower mode of action that required 5
days or more to obtain maximum control. Similar findings were reported by Furr &
Harris (1995) where maximum control (83%) was achieved with tebufenozide at 9
DAT. Although this product has a slightly slower mode of action than chlorfenapyr
and spinosad, it appears to be well suited for integration into an overall pest manage-
ment program. Commercialization of these new compounds for beet armyworm con-
trol in cotton may lower the insecticide inputs (kg AI/ha) required to maintain this
pest under an economic threshold level.
ACKNOWLEDGMENTS
The financial support of Cotton Incorporated is gratefully acknowledged. Appreci-
ation is expressed to American Cyanamid Company, Crop Genetics International,
DowElanco, Rhone-Poulenc Agricultural Company and Rohm & Haas Company for
supplying the insecticides. We thank Don Cook, Chad Comeaux, Larry Daigle, Hunter
Fife, Joe Pankey, Rosanne Mascarenhas and Karla Torres for their assistance in many
aspects of these studies. This manuscript is approved for publication by the Director
of the Louisiana Agricultural Experiment Station as Manuscript No. 96-17-0125.
REFERENCES CITED
ARRILLAGO, P. 1995. Farmers want to end boll weevil spraying, pp. Ic and 7c in The
Advocate (October 10), Baton Rouge, Louisiana.
BURRIS, E., S. MICINSKI, B. R. LEONARD, J. B. GRAVES, AND R. D. BAGWELL. 1994. The
performance of cotton insecticides in Louisiana 1994 Louisiana Agric. Expt.
Sta. Mimeo Series No. 91, 84 pp.
CHANDLER, L. D., AND J. R. RUBERSON. 1994. Comparative toxicity of four commonly
used insecticides to field-collected beet armyworm larvae from the southeast-
September, 1996
Armyworm Symposium '96: Mascarenhas et al.
ern United States, pp. 860- 864 in Proc. Beltwide Cotton Conference, National
Cotton Council, Memphis, Tennessee.
DOUCE, G. K., AND R. M. MCPHERSON. 1991. Summary of losses from insect damage
and cost of control in Georgia, 1989. Georgia Agric. Expt. Sta. Spec. Publ. 70.
FURR, E. R., AND F. A. HARRIS. 1995. Evaluation of insecticide efficacy for beet army-
worm management in Mississippi, pp. 902- 903 in Proc. Beltwide Cotton Conf.,
National Cotton Council, Memphis, Tennessee.
JOYCE, J. A., R. J. OTTENS, G. A. HERZOG, AND M. H. BASS. 1986. A laboratory bioas-
say for thiodicarb against the tobacco budworm, bollworm, beet armyworm and
fall armyworm. J. Agric. Entomol. 3: 207-212.
LEORA SOFTWARE. 1987. POLO-PC a user's guide to Probit or Logit analysis. LeOra
Software, Berkley, California 94707.
LAYTON, M. B. 1994. The 1993 beet armyworm outbreak in Mississippi and future
management guidelines, pp. 854-856 in Proc. Beltwide Cotton Conf, National
Cotton Council, Memphis, Tennessee.
REED, J. T., B. LAYTON, AND C. S. JACKSON. 1994. Evaluation of insecticide and insec-
ticide combinations for beet armyworm control. Arthropod Management Tests
19: 237-238.
SAS INSTITUTE. 1988. SAS/STAT users guide, version 6.03, [ed.] SAS Institute, Cary,
North Carolina. 1028 pp.
SMITH, R. H. 1985. Fall and beet armyworm control, pp. 134-136 in Proc. Beltwide
Cotton Conf., National Cotton Council, Memphis, Tennessee.
SMITH, R. H. 1994. Beet armyworm: a costly caterpillar, pp. 13-14 in Proc. Beltwide
Cotton Conf., National Cotton Council, Memphis, Tennessee.
SULLIVAN, M. J., S. G. TURNIPSEED, AND T. W. SMITH. 1991. Beet armyworm control
in South Carolina, pp. 777 in Proc. Beltwide Cotton Conf., National Cotton
Council, Memphis, Tennessee.
WILEY, G. L., L. R. DESPAIN, K. KALMIWITZ, T. CAMPBELL, F. WALLS, T. HUNT, AND K.
TREACY. 1995. Results of the PirateM insecticide-miticide EUP program on fo-
liage feeding insects on cotton, pp. 925-928 in Proc. Beltwide Cotton Conf., Na-
tional Cotton Council, Memphis, Tennessee.
Armyworm Symposium '96: Rogers and Marti
BEET ARMYWORM (LEPIDOPTERA: NOCTUIDAE): EFFECTS
OF AGE AT FIRST MATING ON REPRODUCTIVE POTENTIAL
C. E. ROGERS AND O. G. MARTI, JR.
Insect Biology and Population Management Research Laboratory
Agricultural Research Service, U.S. Department of Agriculture,
Tifton, GA 31793-0748
ABSTRACT
The effects of age at the first mating on the reproductive potential of the beet ar-
myworm, Spodoptera exigua (Hiibner), was studied in the laboratory. Both the fecun-
dity and fertility of eggs laid were significantly affected (P < 0.01) by age of males and
females at the time of mating. Delayed mating by females increased longevity but de-
creased fecundity and fertility (P < 0.05). Delayed mating by males increased longev-
ity (P < 0.05) but decreased the number of spermatophores they transferred to
females (P < 0.01). The number of spermatorphores transferred during mating af-
fected female fecundity and fertility (P < 0.01). The optimum age for the first mating
Florida Entomologist 79(3)
for both males and females was 1-2 days post-emergence. Delaying the first mating by
either sex beyond 3-4 days post-emergence significantly, and adversely, impacted the
reproductive potential of the beet armyworm.
Key Words: Spodoptera exigua, fecundity, fertility, longevity
RESUME
Fueron estudiados en el laboratorio los efectos de la edad en el moment del primer
apareo sobre el potential reproductive del gusano de la remolacha, Spodoptera exigua
(Hibner). La fecundidad y la fertilidad de los huevos puestos fueron significativa-
mente afectadas (P < 0.01) por la edad de los machos y hembras en el moment del
apareo. El retardo en el apareo en las hembras aument6 la longevidad pero disminuy6
la fecundidad y la fertilidad (P < 0.05). El retardo en el apareo en los machos aument6
la longevidad (P < 0.05) pero disminuy6 el numero de espermat6foros transferidos a
la hembra (P < 0.01). El numero de espermat6foros transferidos a la hembra durante
el apareo afect6 la fecundidad y la fertilidad de las hembras (P< 0.01). La edad optima
para el primer apareo en machos y hembras fue 1-2 dias despues de la emergencia. El
retado del primer apareo mas alla de los 3-4 dias de la emergencia redujo significati-
vamente el potential reproductive del gusano de la remolacha en ambos sexos.
The beet armyworm, Spodoptera exigua (Hiibner), was introduced into Oregon in
1876 and within a few years had migrated across the southern latitudes of the United
States (Mitchell 1979). In recent years, the beet armyworm has become a serious eco-
nomic pest of vegetables and cotton in the Southeast and mid-South (Douce &
McPherson 1991, Layton 1994). The beet armyworm wreaked havoc on the cotton crop
in the Lower Rio Grande Valley of Texas in 1995 (Summy et al. 1996). In the absence
of heavy insecticide pressure on crops, populations of the beet armyworm often are
maintained below economic levels by larval and pupal natural enemies (Ruberson et
al. 1994). The beet armyworm has no effective natural enemies of its adult stage; how-
ever, it is a good factitious host for the ectoparasitic nematode, Noctuidonema guyan-
ense Remillet and Silvain (Nematoda: Acugutturidae) in the laboratory (Rogers &
Marti 1996). Before studying the effects ofN. guyanense on the reproductive potential
of S. exigua, we investigated the effects of age at mating on the reproductive potential
of nematode-free moths.
MATERIALS AND METHODS
Experimental insects were from an established colony of the beet armyworm
maintained on a pinto bean-based diet at the ARS Insect Biology and Population Man-
agement Research Laboratory, Tifton, GA (Burton 1967). As moths emerged, they
were placed individually in a 0.6-liter cardboard container, provided 10% honey solu-
tion for nourishment, and maintained at 27C and 80% RH in a 14:10 h (L:D) photo-
period. To determine the effect of age at first mating on fecundity and fertility in the
beet armyworm, pairs of different male/female age combinations were established.
Tests encompassed 19 trials, each of which contained 19-24 replicated pairs ranging
in age from 1 to 9 days post-emergence. Mating pairs were maintained as mentioned
above in 0.6-liter cardboard cages with the top and bottom covered with netting and
paper towel lids. Cages were lined with waxed paper to facilitate removal and count-
ing of eggs. Cages were examined daily to record data on moth mortality/longevity
September, 1996
Armyworm Symposium '96: Rogers and Marti
and to collect eggs. Eggs were incubated 4 days at 30'C. The number of unhatched
eggs and viable larvae was used to compute daily fecundity and fertility for each pair.
Moths were maintained in their respective cages until both died. Dead females were
dissected to determine the number of harbored spermatophores.
All data were subjected to an analysis of variance by the General Linear Model
Procedure, and significantly different means were separated by the least significant
difference (LSD) test (SAS Institute 1989). Pearson correlation coefficients also were
computed to document significant interactions among biological parameters.
RESULTS AND DISCUSSION
Among the 443 pairs, 112 females received no spermatophores, 145 females re-
ceived a single spermatophore, and one female each received as many as 7, 8, or 9
spermatophores (Table 1). An average of 1.55 spermatophores was transferred to fe-
males among all matings. Discounting the 112 pairs with which no spermatophores
were transferred, the remaining 331 pairs transferred an average of 2.07 spermato-
phores. The number of spermatophores transferred during mating was significantly
affected by male age (r = -0.69; P < 0.01; F = 7.70; d.f. = 8,424), female age (r = -0.56;
P < 0.05; F = 6.45; d.f. = 8,424), and combined pair ages (r = -0.76; P < 0.01; F = 7.87;
d.f. = 10,422) at the first mating. Also, the number of spermatophores transferred dur-
ing mating significantly affected the number of eggs laid (r = 0.73; P < 0.01; F = 1.68;
d.f. = 9,423), the percentage of eggs hatching (r = 0.81;P < 0.01; F = 24.66; d.f. = 9,423),
and the number of resulting viable larvae (r = 0.84; P < 0.01; F = 23.26; d.f. = 9,423).
The lone female receiving 7 spermatophores laid 1,993 eggs and the 6 females receiv-
ing 6 spermatophores laid an average of 1,464.8 eggs, of which 99.4 and 66.4%
hatched, respectively (Table 1). One female received 9 spermatophores. However,
these were small and shriveled, and none of her eggs hatched. Females receiving only
1 or 2 spermatophores laid an average 500.1 to 831.2 eggs, of which 16.3 to 50.9%
hatched. Females receiving no spermatophores laid an average of only 263.9 eggs,
none of which hatched. Thus, with the beet armyworm, it appears that multiple sper-
matophores must be received by females for maximum reproductive capacity to be re-
alized. The number of spermatophores transferred during mating seemed to have no
effect on the longevity of females (r = -0.37, P > 0.05) or males (r =-0.22, P > 0.05). A
correlation between spermatophore transfer at mating and female fecundity and fer-
tility also has been reported for Trichoplusia ni (Hiibner) (Noctuidae) (Ward &
Landolt 1995), S. frugiperda (J. E. Smith) (Rogers & Marti 1994), andAtteva punctella
(Cramer) (Yponomeutidae) (Taylor 1967). It appears that unsuccessful matings (evi-
denced by transfer of viable sperm) and low fertility among lepidopteran species are
not uncommon, especially if the initial mating after emergence is delayed by 3 or more
days (Unnithan & Paye 1991, Proshold 1996, Ellis & Steele 1982, Proshold et al. 1982,
Lingren et al. 1988).
The age of females at their first mating significantly affected fecundity (r = -0.69;
P < 0.01; F = 5.84; d.f. = 8,435), fertility (r = -0.69; P < 0.01; F = 15.45; d. f = 8,435),
longevity (r = 0.74; P <= 0.01; F = 8.21; d.f. = 8,431), and the number of viable larvae
(r = -0.72; P < 0.01; F = 14.88; d.f. = 8,435). Females mating on the first or second day
post-emergence received significantly more spermatophores (P < 0.01; F = 6.45; d.f. =
8,432) than females mating at an older age (Table 2). Percentage hatch of eggs laid by
females mating during their first 3 days was significantly higher than for eggs laid by
older females. Although mating by females at an early age enhances their fecundity
and fertility, it also significantly shortens their life span (Table 2). Females mating by
day 3 post-emergence lived an average 12.4 to 17.4 days, while those mating at an
Florida Entomologist 79(3)
t3NrO &1 t 0 e
0 ~ toe
September, 1996
Armyworm Symposium '96: Rogers and Marti
bfi
C,1
m +1
'V
+1
z
z
rA
o +|
Cr &1N &1N & t 1
Q & eC t o eN NC
Florida Entomologist 79(3)
older age lived for 20 days or more. The number of eggs laid by females also signifi-
cantly affected their longevity (r = -0.51; P < = 0.05; F = 1.36; d.f. = 37,402). Also, egg
fertility was inversely correlated with female longevity (r = -0.67; P < 0.05; F = 5.32;
d.f. = 37,402). Females laying an average of 676.8 or more eggs had an average lon-
gevity of up to 12 days, whereas females laying an average of fewer than about 400
eggs lived an average of more than 15 days. The fertility of eggs laid by females living
fewer than 12 days averaged 56.4 to 93.7%, whereas the fertility of eggs from females
living more than 15 days averaged less than 27%. These relationships among repro-
ductive parameters of females are not unique to S. exigua. These same relationships
exist in other lepidopteran species, e.g., Spodoptera littoralis (Boisduval) (Ellis &
Steele 1982), Lymantria dispar (L.) (Lymantridae) (Proshold 1996), Chilo partellus
(Swinhoe) (Pyralidae) (Unnithan & Paye 1991), Heliothis virescens (F.) (Noctuidae)
(Proshold et al. 1982), Pectinophora gossypiella (Saunders) (Lingren et al. 1988), and
S. frugiperda (Rogers & Marti 1994).
The age of males at first mating significantly affected the number of spermato-
phores transferred (r = -0.69; P < 0.01; F = 7.70; d.f. = 8.424), the number of eggs laid
by their mates (r = -0.59; P < 0.01; F = 14.09; d.f. = 8,435), fertility of mate's eggs (r =
-0.65; P < 0.01; F = 14.09; d.f. = 8,435), the number of resulting viable larvae (r = 0.61;
P < 0.01; F = 9.94; d.f. = 8,435), and the longevity of males after their first mating (r
= -0.65; P < 0.01; F = 8.77; d.f. = 8,425 (Table 3). Males mating on days 1-4 post-emer-
gence transferred significantly more spermatophores (x = 1.82-2.23) than males mat-
ing on days 6-9 (x = 0.7-1.2). Significantly more eggs were laid by females mating with
1- to 3-day old males (x = 701.1-787.4) than females mating with 7- to 9-day old males
(x = 343.6-371.6). The percentage of eggs hatching from eggs of females mating with
1- to 3-day old males averaged about 45%, compared with 5.2-19.9% hatching of eggs
from females mating with 4- to 9-day old males. Although the total longevity of males
was only weakly correlated with mating age (r = 0.30; P > 0.05; F = 3.54; d.f. = 8,425),
their longevity after mating averaged 15.8 to 17.6 days for those mating on day 1-3
post-emergence, but only 10.6 to 14.5 days for those mating 4-9 days post-emergence.
Female longevity was only weakly correlated (r = 0.35; P > 0.05; F = 5.72; d.f. = 8,431)
with male age at mating.
Although the reproductive status of males affected the reproductive potential of
the beet armyworm, their contributions had less impact than those of females. For the
fall armyworm, the age of males at first mating had little impact on the reproductive
potential of female mates (Rogers & Marti 1994). However, in H. virescens and L. dis-
par, the age of males at first mating affected spermatophore and/or sperm transfer to
females during mating (Proshold 1996, Proshold & Bernon 1994). InA. punctella, the
failure of female fertility is usually due to the failure of males to transfer sperm dur-
ing mating, but there was no relationship between potency of sperm and age of males
at mating (Taylor 1967). In C. partellus, the age of males had no effect on mating ac-
tivity up to 5 days post-emergence, nor did it affect the fecundity or fertility of female
mates (Unnithan & Paye 1991).
Just as the age of females and males at mating independently affected the repro-
ductive potential of the beet armyworm, the combined age of mating pairs interacted
to significantly affect the number of spermatophores transferred (r = -0.76; P < 0.01;
F = 7.87; d.f. = 10,422), the number of eggs laid (r = -0.78; P < 0.01; F = 12.45; d.f. =
10,433), fertility of eggs (r = -0.82; P < 0.01; F = 22.34; d.f. = 10,433), the number of re-
sulting viable larvae (r = 0.81; P < 0.01; F = 19.36; d.f. = 10,433), and the longevity of
female partners (r = 0.67; P < 0.01; F = 5.83; d.f. = 10,429). Mating moths with a com-
bined age of 2 days transferred significantly more spermatophores (x = 3.3) than pairs
with a combined age of 4 days or more (x = 0.7-2.1). Pairs with a combined age of 2 or
September, 1996
Armyworm Symposium '96: Rogers and Marti
bh
r, +1
rA
bfl +1
ggx
C
C
P
^^j
vi^
E +1
Ha"l^
d
C
E
cc~"
Ix+
m%`
d ""
Florida Entomologist 79(3)
1300
SY 0970.7-41.9*X
1100 95% Confidence Bands
r --0.78 p<0.001
S900
0 700 ...........
Z .............
3 00
z" --+ .. .
100
0 4 8 12 16 20
COMBINED AGE
Fig. 1. Fecundity of the beet armyworm as a function of combined age of mating
pairs on their first mating.
4 days at mating produced significantly more eggs (x? = 907.4) than pairs with a com-
bined age greater than 6 days (x = 238.7-610.9) (Fig. 1). Eggs laid by females from
pairs with a combined age of 2 or 4 days at first mating produced eggs having a sig-
nificantly higher percentage hatching (x = 57.7-61.1%) than eggs from older pairs (x
= 1.8-44.4%). If either mate of the pair were aged, the contributions of its younger
mate were greatly minimized (Fig. 2). The number of viable larvae produced was sig-
nificantly greater in mating pairs with a combined age of 2-4 days (x = 637.1-767.7)
than the number produced by pairs with a combined age greater than 6 days (x = 9.8-
316.7). Females from pairs having a combined age of 2-6 days at first mating lived sig-
nificantly fewer days (x = 15.1-16.2) than females of pairs having a combined age
greater than 8 days (x = 19.6-21.9).
In summary, the reproductive potential of the beet armyworm is significantly im-
pacted by the age of both males and females at their first mating. However, there is a
trade-off between enhanced reproductive potential of young moths and longevity of
ovipositing females, i.e., more fecund/fertile moths have a shorter life span than fe-
males with a reduced reproductive potential. One to 2 days post-emergence is the op-
timum age for the first mating for both sexes for maximum reproductive potential.
Hence, to evaluate the effects of adult mortality factors, e.g., parasitic nematodes, on
the reproductive potential of the beet armyworm, it will be necessary to standardize
the age of young moths.
REFERENCES CITED
BURTON, R. L. 1967. Mass rearing the fall armyworm in the laboratory. USDA-ARS-
33-177. Govt. Print. Office, Washington, DC.
DOUCE, G. K., AND R. M. MCPHERSON. 1991. (eds.) Summary of losses from insect
damage and costs of control in Georgia, 1989. GA Agric. Exp. Sta. Special Pub-
lication 70, 46 pp.
September, 1996
Armyworm Symposium '96: Rogers and Marti
2
10
MALE A
Fig. 2. Fertility of beet armyworm eggs as a function of male and female age at the
first mating of a pair.
ELLIS, P. E., AND G. STEELE. 1982. The effects of delayed mating on the fecundity of
females ofSpodoptera littoralis (Boisduval) (Lepidoptera: Noctuidae). Bull. En-
tomol. Res. 72: 295-302.
LAYTON, M. B. 1994. The 1993 beet armyworm outbreak in Mississippi and future
management guidelines. Beltwide Cotton Conf, Vol 2: 854-856.
LINGREN, P. D., W. B. WARNER, AND T. J. HENNEBERRY. 1988. Influence of delayed
mating on egg production, egg viability, mating, and longevity of female pink
bollworm (Lepidoptera: Gelechiidae). Environ. Entomol. 17: 86-89.
MITCHELL, E. R. 1979. Migration by Spodoptera exigua and S. frugiperda, North
American style, pp. 386-393 in R. L. Rabb and G. E. Kennedy [eds.], Movement
of highly mobile insects: concepts and methodology in research. NC State Univ,
Raleigh, 456 pp.
PROSHOLD, F. I. 1996. Reproductive capacity of laboratory-reared gypsy moths (Lepi-
doptera: Lymantriidae): effect of age of female at time of mating. J. Econ. En-
tomol. (accepted 1/15/96).
PROSHOLD, F. I., AND G. L. BERNON. 1994. Multiple mating in laboratory-reared gypsy
moths (Lepidoptera: Lymantriidae). J. Econ. Entomol. 87: 661-666.
Florida Entomologist 79(3)
September, 1996
PROSHOLD, F. I., C. P. KARPENKO, AND C. K. GRAHAM. 1982. Egg production and ovi-
position in the tobacco budworm: effect of age at mating. Ann. Entomol. Soc.
America 75: 51-55.
ROGERS, C. E., AND O. G. MARTI, JR. 1994. Effects of age at first mating on the repro-
ductive potential of the fall armyworm (Lepidoptera: Noctuidae). Environ. En-
tomol. 23: 322-325.
ROGERS, C. E., AND O. G. MARTI, JR. 1996. Beet armyworm (Spodoptera exigua) as a
host for the ectoparasitic nematode, Noctuidonema guyanense. J. Agric. Ento-
mol. (Accepted 1/22/96).
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 ene-
mies. Florida Entomol. 77: 440-453.
SAS INSTITUTE. 1989. SAS/STAT User's guide, version 6, 4th ed., Vol. 1 and Vol. 2.
SAS Institute, Cary, NC.
SUMMY, K. R., J. R. RAULSTON, D. SURGEON, AND J. VARGAS. 1996. An analysis of the
beet armyworm outbreak on cotton in the Lower Rio Grande Valley of Texas
during the 1995 production season. Beltwide Cotton Conf. (accepted 1/7/96).
TAYLOR, 0. R., JR. 1967. Relationship of multiple mating to fertility in Atteva punc-
tella (Lepidoptera: Yponomeutidae). Ann. Entomol. Soc. America 60: 583-590.
UNNITHAN, G. C., AND S. O. PAYE. 1991. Mating longevity, fecundity, and egg fertility
of Chilo partellus (Lepidoptera: Pyralidae): effects of delayed or successive mat-
ings and their relevance to pheromonal control methods. Environ. Entomol. 20:
150-155.
WARD, K. E., AND P. J. LANDOLT. 1995. Influence of multiple matings on fecundity and
longevity of female cabbage looper moths (Lepidoptera: Noctuidae). Ann. Ento-
mol. Soc. America 88: 768-772.
Buckingham and Bennett: Biology of Parapoynx diminutalis 353
LABORATORY BIOLOGY OF AN IMMIGRANT ASIAN MOTH,
PARAPOYNX DIMINUTALIS (LEPIDOPTERA: PYRALIDAE),
ON HYDRILLA VERTICILLATA (HYDROCHARITACEAE)
GARY R. BUCKINGHAM' AND CHRISTINE A. BENNETT2
'Agricultural Research Service
U.S. Department of Agriculture
Florida Biological Control Laboratory
Gainesville, FL 32614-7100
'Department of Entomology and Nematology
Institute of Food and Agricultural Sciences
University of Florida, Gainesville, FL 32611
ABSTRACT
The Asian moth Parapoynx diminutalis Snellen is an immigrant in Florida and
Panama where it attacks hydrilla, Hydrilla verticillata (L. fil.) Royle, an immigrant
submersed weed from Asia. Field populations of P. diminutalis are occasionally heavy
on hydrilla but are rarely found on other plant species, including those that are labo-
ratory hosts. Larvae build portable cases from which they feed on leaves and stems.
The 7 instars can be differentiated by head capsule widths. Measurements are pre-
sented of other immature stages. In the laboratory at 26.7C, eggs developed in 4-6 d,
larvae in 21-35 d, prepupae in 1-2 d, and pupae in 6-7 d. Adults lived 3-5 d at 24.4C.
Key Words: Biological control, fecundity, degree-days, host plants, Bacillus thuring-
iensis
RESUME
La polilla asiatica, Paraponynx diminutalis Snellen es un inmigrante de la Florida
y Panama donde ataca a Hydrilla verticillata (L. fil.) Royle, una hierba sumergida in-
migrante de Asia. Las poblaciones de campo de P. diminutalis son ocasionalmente al-
tas en H. verticillata pero raramente son encontradas en otras species de plants,
incluidas las que son hospedantes de laboratoirio. Las larvas fabrican una funda por-
tatil desde la cual se alimentan de las hojas y los tallos. Los siete instares pueden ser
diferenciados mediante el ancho de la capsula cefalica. Las medidas de los estadios in-
maduros son presentadas. En el laboratorio, a 26.7C, los huevos se desarrollaron en
4-6 dias, las larvas en 21-35 dias, las prepupas en 1-2 dias y las pupas en 6-7 dias. Los
adults vivieron 3-5 dias a 24.4C.
Parapoynx diminutalis Snellen is an immigrant Asian moth that attacks the sub-
mersed aquatic plant hydrilla, Hydrilla verticillata (L. fil.) Royle, in Florida. Hydrilla,
also an Asian immigrant, is the state's most important aquatic weed and is found
throughout much of the Southern USA and along the East Coast to Maryland (Lange-
land 1990). The native range of hydrilla includes Asia, Australia, the Rift Valley area
ofAfrica, and Europe, where only relict populations occur. The moth was first reported
in Florida in 1976, when it was discovered in hydrilla plantings at a research station
in Ft. Lauderdale (Del Fosse et al. 1976). Subsequently, it was found in north Florida
Florida Entomologist 79(3)
(Balciunas & Habeck 1981) and in Panama, where it infested hydrilla in the Panama
Canal. It has also been reported colonizing glasshouses at aquatic plant nurseries in
England and Denmark (Agassiz 1978, 1981). Its native range includes much of Asia
from Afghanistan to the Philippines and north to Shanghai, People's Republic of
China, Africa from Ethiopia to South Africa (Speidel 1984) and Australia (Yoshiyasu
1992). Prior to its discovery in Florida, P. diminutalis was the subject of a study in Pa-
kistan to determine its potential for introduction into the United States for biological
control of hydrilla (Baloch & Sana-Ullah 1974).
Baloch & Sana-Ullah (1974) presented a brief description of the biology along with
results of preliminary laboratory host range tests. Additional biological studies, appar-
ently with this species, were reported by Varghese & Singh (1976) in Malaysia in sup-
port of the USA biological control program. They referred to their subject asNymphula
sp. in the text but as N. diminutalis in the table headings. Nymphula diminutalis is a
synonym of P. diminutalis. Chantaraprapha & Litsinger (1986) included a table with
life history data for P. diminutalis in their study of the host range in the Philippines.
We originally began studies with P. diminutalis collected in Panama and held in
quarantine at the Florida Biological Control Laboratory, Division of Plant Industry
(DPI), Florida Department of Agriculture and Consumer Services in Gainesville, FL.
It was thought to be a little known Neotropical species, P. rugosalis Moschler. Balciu-
nas & Center (1981) attempted to study P. rugosalis in Panama before its introduction
into our quarantine and reported results of tests with larvae ofParapoynx sp. The spe-
cies tested by them was probably not P. rugosalis but rather a third, as yet unidenti-
fled species. Three subsequent attempts by researchers, including one of us (CAB), to
collect and carry to our quarantine laboratory the unidentified Parapoynx sp. yielded
only P. diminutalis. We continued our studies after determining the correct identity of
our material in order to better document the laboratory biology and host range of this
interesting species. We report here the results of our biological studies. Our host
range studies have been reported elsewhere (Buckingham & Bennett 1989).
MATERIALS AND METHODS
All experiments were conducted in the quarantine facility at the Florida Biological
Control Laboratory from 1980 to 1982. Hydrilla, the laboratory host plant, was col-
lected periodically from several sites in Levy and Alachua counties or was grown in
outdoor pools at the laboratory. The majority of test insects were from a colony estab-
lished from eggs and larvae collected in November 1980 in the Rio Chagres near Gam-
boa, Panama, by J. K. Balciunas. Additional insects were collected at Orange and
Lochloosa Lakes near Gainesville. Adult and larval specimens have been deposited in
the Florida State Collection ofArthropods, DPI, Gainesville and in the National Mu-
seum of Natural History, Washington, D.C. The rearing colony was maintained in a
quarantine greenhouse in a large wooden cage described by Buckingham & Bennett
(1989). Larvae were also reared in 3.8 liter jars filled with water and hydrilla and
capped with nylon organdy. The jars were held either in a greenhouse (12.2-31.1 C,
50-90% RH) or in a temperature- and humidity-controlled room (25 1 C, 50-60%
RH). Both had fluorescent lights on 16 h photophase. Plants and water were added
when necessary. The rearing material was transferred at least weekly to new jars
sterilized with bleach.
Biological Studies
All studies were conducted in environmental chambers at 26.7 C, 16 h photophase,
unless otherwise indicated. Individual larvae on hydrilla were confined after hatching
September, 1996
Buckingham and Bennett: Biology of Parapoynx diminutalis 355
in small plastic cups, 29.6 ml, with plastic lids. These small cups were used in many
of the studies. The cups were examined daily for shed head capsules to determine the
number and the durations of the stadia. Head capsule measurements were made with
larvae preserved in 75% isopropyl alcohol. All measurements were made with a ste-
reomicroscope and an ocular micrometer and are reported as follows: mean stan-
dard deviation (range, number).
Neonate mortality outside of water on both dry and moist filter paper was tested
by placing single larvae (n=30) without hydrilla in small plastic cups with lids. Neo-
nate mortality in aerated water was tested in small plastic cups which had organdy
glued over holes in the bottoms (n=20). The cups sat partially submerged on gravel in
water-filled, aerated pans. These were compared with water-filled cups without holes
(n=50). The neonates were tested at 22.2 1C, and those in water were provided with
hydrilla.
Most larval observations were made in small plastic cups, in 266 ml styrofoam
cups, or in 0.95 liter or 3.8 liter jars. Hydrilla and water were added as needed. Larval
and larval-pupal development times were determined at various temperatures in en-
vironmental chambers with 16 h photophases.
Pupal development times were determined by observing pupae removed from co-
coons and held on moist cotton in small plastic cups or in plastic petri dishes.
Newly emerged females were confined individually with 2 newly emerged males in
plexiglass cylinders (42 cm high, 14.5 cm ID) to determine longevity and realized fe-
cundity, the number of eggs laid. We did not record matings, and deaths were recorded
daily. Small amounts of water and hydrilla were placed in the bottom of each cylinder,
which was capped with nylon organdy. Some females were dissected at death to deter-
mine the number of eggs remaining in the ovaries.
RESULTS AND DISCUSSION
Egg Stage
Description. Dorsoventrally compressed; outline circular at deposition, elliptical at
maturity; chorion appears smooth at 50x; bright yellow at deposition changing to
whitish and then transparent at hatching; length and width of mature egg presented
in Table 1. The height of one detached mature egg was 0.26 mm.
Development. Eggs were deposited in various size masses on leaves or stems of
plants lying at the water surface, on moist filter paper in petri dishes, and on the sides
of plexiglass cylinders at the waterline. The egg masses on hydrilla (Table 1) were gen-
erally smaller than masses placed on filter paper. Eggs were arranged in uneven rows
within a mass. Most of the developing larvae were oriented in the same direction;
however, often the larvae in one row, or portions of a row, were oriented in the opposite
direction. Larval head position was initially indicated by two small dark eyespots;
later the head was light brown with dark mandible tips. The curled embryo developed
lying on its side, but within a few hours prior to hatching it actively moved within the
egg. The development time is presented in Table 1. Usually all or most eggs in a mass
hatched on the same day. Normal egg development occurred both underwater and out-
side of water on moist filter paper.
Larval Stage
Description. Seven (I-VII) instars: I whitish almost transparent; no tracheal gills;
orange malpighian tubules noticeable; one pair of long anal setae (0.20-0.22mm);
Florida Entomologist 79(3)
smaller setae in longitudinal rows dorsally and along each side of abdomen; head cap-
sule light brown with dark brown ocelli and epicranial suture; unspotted; mandibles
reddish; light brown pronotal shield about equal to width of head capsule (Table 1);
length of neonates about 1.00 mm. II -VII whitish, mature larva turns yellowish just
before pupation; tracheal gills present along each side of abdomen, number increasing
on successive instars; head capsule light brown with scattered small dark brown spots
at base of setae (Fig. 1A.), length increasing with each instar to 7.4-14.1 mm for last
instar; instars best separated by widths of head capsules (Table 1). Head capsule
widths were not compared between sexes. A detailed description and an illustration
of the head capsule of the fourth instar was presented by Yoshiyasu (1985) for Japa-
nese specimens.
Development. The duration of the larval stage and pupal stage at 26.7 C is pre-
sented in Table 1, along with the approximate duration of each stadium. The days re-
quired for development from neonate to adult at 4 constant temperatures is presented
in Table 2. A plot of the development rates (1/days to develop) against temperatures
revealed that 36.1C was near the upper threshold. Although adults were produced at
that temperature, the development rate slowed. The estimated lower threshold for de-
velopment obtained by the "linear approximation" method (Wilson & Barnett 1983)
was 12.7 C. The threshold was identical with linear regression of the 3 lower temper-
atures (y = -0.0308 + (0.00243x), r = 0.99). The estimated degree-days for mean devel-
opment times and for the ranges calculated using 12.7 C were 424 (351-551), 388
(294-490), 418 (340-483), and 494 (397-608), respectively, for the four temperatures
listed in Table 2. Although these are approximations, they should be useful when
planning additional field or laboratory studies with this species.
Behavior. Neonates were active crawlers. Although some began feeding immedi-
ately, most wandered about the containers before settling on plants. It was not un-
usual to find that some had crawled from the water and out of containers that were
not covered. Neonates fed on leaves, either by scraping the surface and rendering the
leaf transparent, or by completely eating portions of the leaf. Most fed without a shel-
ter but some made simple shelters by cutting small (about 2mm long) pieces of leaf
and attaching them to the leaf surface. Most second instars made these simple shel-
ters, and all later instars constructed tubular cases by tying together pieces of leaves
or stems. Their cases might be mistaken for those of some caddisflies. Balciunas &
Minno (1985) included P. diminutalis in a key to the larval cases of insects on hydrilla
in the United States. Larvae generally fed on leaves by partially exiting from the
cases they carried with them. At night, however, larvae were observed crawling with-
out cases. On one occasion, many naked larvae were clustered together near the sur-
face in an aerated jar that had a fluorescent light lying across the mouth. This
suggested an attraction to light, but we did not test this hypothesis. Leaves were
eaten most readily, but portions of the stem were also eaten during heavy feeding.
Mortality. Neonates often crawled from their containers, but they were unable to
develop outside water. All died within 1 h in dry cups, whereas in cups with moist fil-
ter paper all survived for 23.5 h. Survival in moist cups without hydrilla was 83% at
32 h, 57% at 46.5 h, 30% at 55 h, and 0 at 72 h. Two of an additional three neonates
tested in moist cups with hydrilla were still alive at 72 h but both were dead at 6 d.
These relatively long periods of survival out of water suggest that neonates might suc-
cessfully disperse with hydrilla carried on boat trailers, boat propellers, etc. When ne-
onates were held in water without food, forty-two percent were alive after 6 d in the
greenhouse at 14.3-26.7 C, but none was alive after 14 d.
Early instar mortality (within the first 2 weeks) ranged from 30 to 100% and was
greater than, or equal to, 50% in ten of 21 diverse experiments where individual lar-
September, 1996
Buckingham and Bennett: Biology of Parapoynx diminutalis 357
TABLE 1. BIOLOGICAL STATISTICS FOR P. DIMINUTALIS.
Statistic n Mean SD Range
Measurements
Mature egg (mm)
Length
Width
Egg mass size (Nos.)
Larval head capsule (mm)
Instar I
II
III
IV
V
VI
VII
Female pupa (mm)
Length
Width
Width at spiracles
Male pupa (mm)
Length
Width
Width at spiracles
Wingspan (mm)
Female
Male
10 0.44 0.02
10 0.34 0.02
112 29.7 19.4
18 0.21 0.01
5 0.29 0.02
1 0.42
8 0.62 0.03
16 0.81 0.05
30 1.02 0.02
18 1.15 0.04
11 7.82 0.34
11 2.09 0.10
11 2.31 0.09
10 6.63 0.20
10 1.67 0.11
10 1.84 0.11
10 18.12 0.98
10 13.40 0.93
Development Times (Days @ 26.7C)
Egg
Larva + Pupa
Prepupa
Pupa
Total (Estimated)
4-6
17 27.1 3.8 21-351
1-2
10 6-7
25-41
Longevity (Days)
22.2-25.6'C
Female
Male
33 5.2 2.1
26 5.9 3.2
'Instar 1=4, 11=3, III=3-5, IV=2-3, V=2-5, VI=3-8, VIII=3-9 days.
0.42-0.48
0.32-0.34
2-99
0.20-0.24
0.26-0.30
0.60-0.66
0.75-0.90
0.98-1.06
1.10-1.20
7.14-8.33
1.96-2.30
2.21-2.47
6.14-6.80
1.45-1.79
1.62-1.96
16.50-19.67
12.34-14.84
Florida Entomologist 79(3)
TABLE 1. (CONTINUED) BIOLOGICAL STATISTICS FOR P. DIMINUTALIS.
Statistic n Mean SD Range
24.4C
Female 3 4.3 0.6 4-5
Male 6 4.0 1.4 3-5
'Instar =4, 11=3, III=3-5, IV=2-3, V=2-5, VI=3-8, VIII=3-9 days.
vae were monitored. There were no obvious causes for this; however, initial mortality
(4-9 d) was generally less [41 30% (0-100, 18 experiments)] when eggs containing ac-
tive larvae were transferred to the test containers rather than neonates [61 20%
(30-90, 6)]. In various experiments (Buckingham & Bennett 1989), more adults were
generally produced in the greenhouse than in the laboratory.
Aeration of the water did not significantly improve survival in the small contain-
ers with individual neonates and hydrilla. After 5 days, 16 of 20 neonates were alive
in aerated cups versus 34 of 50 in non-aerated cups (X', p=.48). After 13 days, 15 lar-
vae were still alive in the aerated cups versus 30 in the non-aerated cups (p=.36).
There was no apparent increase in survival in various other experiments in which the
small cups were aerated versus those experiments in which they were not aerated.
Aeration, however, did decrease mortality in rearing jars containing older larvae with
too high a density of hydrilla stems. Several jars were found with moribund larvae
that recovered when the water was aerated. This mortality with densely packed hyd-
rilla was probably due to the decrease in oxygen at night (measured at 1.4 ppm)
caused by plant respiration. Chantaraprapha & Litsinger (1986) also reported that
larval mortality was high (50-60%) during their experiments with this species. Mor-
tality was generally higher when large numbers of larvae, especially the early instars,
were present. Because there was an abundance of leaves for the early instars, the
Fig. 1. Parapoynx diminutalis Snellen: A) mature larva with characteristic dark
spots on the head capsule, B) male.
September, 1996
Buckingham and Bennett: Biology of Parapoynx diminutalis 359
TABLE 2. PERCENTAGE SURVIVAL AND TIME (DAYS) REQUIRED OF P. DIMINUTALIS FOR
DEVELOPMENT FROM NEONATE TO ADULT AT FOUR CONSTANT TEMPERATURES.
Temp ( C) % Survival n Mean SD Range
22.2 70 21 44.6 5.1 37-58
26.7 40 20 27.7 3.8 21-35
30.6 60 24 23.4 2.6 19-27
36.1 40 20 21.1 2.8 17-26
higher mortality suggests that larvae killed each other rather than competed for re-
sources. However, no observations were made to confirm that.
Larvae are often quite damaging in research plantings of hydrilla in pools or other
small protected containers in Florida. We immersed larvae with hydrilla in a commer-
cial preparation of Bacillus thuringiensis, Dipel Hg, (Sunnyland Corp., Sanford FL) at
10% of the dosage recommended as a garden spray and produced 80% mortality
within 4 days compared to no mortality in the controls. The surviving larvae in the
Dipel Hg did not feed and were dead when examined after 10 days. Larvae were found
relatively often on other plant species associated with hydrilla in our research plant-
ings, but in addition to hydrilla, we have found immatures in the field only on coon-
tail, Ceratophyllum demersum L. (1 larva, 1 pupa), southern naiad, Najas
guadalupensis (Sprengel) Magnus (2 larvae), and Illinois pondweed, Potamogeton il-
linoensis Morong (5 larvae, 3 pupae). This is in sharp contrast to the 14 plant species
in 13 genera that produced adults in our host range tests (Buckingham & Bennett
1989).
Pupal Stage
Description. Similar to those of other species of Parapoynx (Lekic 1971, Virak-
tamath et al. 1974, Yoshiyasu 1985). Narrow, elongate, with 3 distinct spiracular tu-
bercles along each side and 2 strong dark setae on top of the head. Female
differentiated most easily from male by the length of the antennae. Female antennae
extending to abdominal segment A4 just anterior to the wing tips; male antennae ex-
ceeding the wing tips and extending to A5. Females generally larger than males (Ta-
ble 1). The abdomen expands in length when the pupa darkens prior to emergence.
Lengths of expanded darkened pupae: female 8.97 0.05 mm (8.93-9.01, 4); male -
7.9 mm (n=l).
The pupa was enclosed in an air-filled white silken cocoon. The cocoon was firmly
attached along one side to the submersed stem or occasionally was attached at one
end and perpendicular to the stem. The cocoons were covered with leaves or bare stem
sections and were similar to the larval cases. Pupae obtained air through the cocoon
from 1-4 excavations of various sizes made in the stems by the larvae. Air was held be-
tween the layers of the multilayered silk cocoon in a broad silvery band surrounding
the spiracles. Lekic (1971) illustrated similar holes in the plant associated with co-
coons ofP. stratiotata L., and Yoshiyasu (1985) illustrated a silvery band in the cocoon
of Parapoynx vittalis (Bremer). The length of female cocoons was generally greater
than that of male cocoons (Table 1).
Development. Prepupae remained immobile in the cocoon for 1-2 d before pupat-
ing. Pupal development times are presented in Table 1. Distinct color changes were
associated with pupal development. Body color changed from white to yellow while
Florida Entomologist 79(3)
the eyes darkened to red and then to dark brown. The wing pattern was visible shortly
before emergence. The pupa actively moved its abdomen when disturbed during the
first few days but was immobile during the last few days. Pupae died in the cocoons if
the stems were only a few cms long or if the stem became waterlogged. Pupae devel-
oped normally when removed from cocoons in water to moist sphagnum or to filter pa-
per in closed containers. All cocoons were constructed on submersed stems. This
contrasts with the observation ofVarghese & Singh (1976) that the anterior end of the
pupa often remained above the water surface and of Chantaraprapha & Litsinger
(1986) that they pupate out of water on plant vegetation. These differences might be
due to water level changes exposing the cocoons or to a complex of species rather than
one widely distributed species.
Adults
Description. Adults are white with sinuous light-brown or tan bands on the wings
(Fig. 1B). Variable amounts of black scales mask portions of the bands. The body has
transverse tan stripes. Females had a wider wingspan (Table 1), more pointed fore-
wings, a more robust abdomen, and relatively shorter antennae than the males. The
tip of the male abdomen has a tuft of white setae that is more noticeable than that of
the female. With experience, we could easily distinguish males by this character. De-
tailed taxonomic descriptions along with illustrations of the genitalia can be found in
Agassiz (1978), Speidel (1984), and Yoshiyasu (1985). The North American species
most easily confused with P. diminutalis, especially the male, is Parapoynx allionealis
(Walker). However, the light-brown or tan middle (postmedial) band on the forewing
of P. diminutalis is much more sinuous than that on P. allionealis, which was illus-
trated by Monroe (1973). The wing patterns of both sexes ofP. diminutalis are similar
although the male generally has more black scales on the wings. The amount of black
tended to vary inversely in both sexes with the temperature at which the larva was
reared and with the type of host plant. Both sexes generally held the wings outspread
and pressed to the substrate while resting. When disturbed, however, they often
landed with the forewings folded over the hindwings but not overlapping the body.
Development and Longevity. The development time from neonate to adult at 26.7C
is presented in Table 1. No cohorts were followed from egg deposition to adult; how-
ever, total development time at 26.7C should be 25-41 days based upon an egg stage
of 4-6 d and the larval/pupal times.
Mean longevities of males and females were the same (Table 1). In a greenhouse
at fluctuating temperatures, mean longevities were slightly longer: females 5.2 2.1
days (n=33, max 8 days), males 5.9 3.2 days (n=26, max 17 days). However, deaths
were not recorded on the weekends in that experiment.
Fecundity. In oviposition experiments females that had been reared at various tem-
peratures from 21.1 C to 26.7C laid a mean of 222.9+ 140.6 eggs (3-524, 26). The mean
without four females who laid less than 22 eggs each was 261.2+ 116.3 eggs (108-524,
22). Fourteen of the preceding 22 females were dissected at death. Eleven had no eggs
in their ovaries and three had only 24.7 + 20.5 eggs (4-45). Differences among individ-
uals in egg deposition apparently reflected true differences in the egg complements ex-
cept for individuals that laid very few eggs. For example, three females that laid 0, 3,
and 19 eggs had 26, 199, and 196 eggs, respectively, in their ovaries at death.
Behavior. Adults began emerging in the large greenhouse cage approximately 30
min after dusk and began flying about 30 min later. After emergence, they sat above
or on the water surface while their wings expanded. If disturbed before their wings ex-
panded, they ran quickly across the hydrilla or the water surface, their long legs pro-
viding speed as they remained high above the substrate. Even though unable to fly
September, 1996
Buckingham and Bennett: Biology of Parapoynx diminutalis 361
during these first few minutes, they would not be easy prey for most predators. We ob-
served a few adults emerge shortly before dusk in India in 1985 while we were con-
ducting another project, although the majority of adults emerged after dusk. One of
us (CAB) also observed pre-dusk emergence in Panama in 1982.
Mating was not observed in the greenhouse cage until about 3 h after dusk. Pairs
in copula rested facing in opposite directions for at least 30 min. The maximum time
in copula was not determined. We did not test females for multiple matings. No fe-
males oviposited the night they emerged. In the cage, they oviposited within the first
hour after dusk. The female sat above the water surface and usually inserted her ovi-
positor into the water to oviposit on leaves or stems. Eggs were placed occasionally,
just above the water surface.
Because of the 1-day preoviposition period, there is probably heavy preoviposi-
tional mortality in the large water bodies in Panama and Florida. Hydrilla mats ex-
tend considerable distances from shore, thus many moths rest during the day exposed
on the mats. Balciunas & Habeck (1981) reported heavy bird predation on moths at a
large Florida lake. We have also observed large numbers of detached moth wings on
the water surface above hydrilla mats.
Although the adult proboscis was greatly reduced, it was apparently functional.
Adults that had been held in a refrigerator for several days without water and adults
collected at outdoor lights extended their probosci to the water when presented wet
leaves and wet paper toweling. We did not test longevity without water, but adults
provided sugar water lived no longer than those with water.
Adults in the greenhouse cage were attracted to both incandescent white lights
and UV blacklights. They did not respond to incandescent red light which was used to
observe them without disturbance.
The sizes of both the egg and the last instar head capsule of our population were
considerably smaller than those reported by Varghese & Singh (1976) in Malaysia. Al-
though these differences might be merely populational or technique differences, they
suggest that two different species might have been studied. Our 4th instar head cap-
sule width was equal to that of a Japanese 4th instar illustrated by Yoshiyasu (1985).
The number of larval instars that we found (7), was different from that reported by
Varghese & Singh (1976) (6) and by Chantaraprapha & Litsinger (1986) (4). Fecundity
and development times also varied, but those are often greatly influenced by tech-
niques. Taxonomic studies of these populations and others of this widely distributed
species appear to be warranted.
Parapoynx diminutalis is apparently well established in both Panama and Flor-
ida, but its potential to extensively damage or control hydrilla is probably greater in
Panama because of the milder climate. In Florida, at least in north-central Florida,
near Gainesville, the larval populations are reduced to almost undetectable levels
during late winter and early spring probably by the cool water temperatures in au-
tumn and winter and the annual late winter decline in hydrilla. However, some pop-
ulations near Gainesville rebounded quickly in some years, so that by late summer
the defoliation of surface hydrilla stems was extensive. Early summer augmentation
of moth populations by the release of immatures or adults might allow the popula-
tions to increase fast enough to at least slow the growth of hydrilla. Extensive defoli-
ation might also increase the plants sensitivity to herbicides or pathogens. If plants
attacked were more susceptible, an integrated program might be feasible.
ACKNOWLEDGMENT
The following individuals helped greatly with this study: Joe Balciunas, Al Cof-
rancesco, Mary Davis, Dale Habeck, Bonnie Ross, Steve Hampton, and Russell The-
Florida Entomologist 79(3)
September, 1996
riot. Personnel at Manatee Springs State Park, Florida Department of Natural
Resources, allowed us to collect hydrilla. Plants were identified by Kenneth Langdon
and Carlos Artaud, Florida Department of Agriculture and Consumer Services, Divi-
sion of Plant Industry, Gainesville. The moths were identified by D.C. Ferguson, SEL,
ARS/USDA. The Aquatic Plant Control Research Program, U.S. Army Corps Engi-
neers, Waterways Experiment Station, Vicksburg, MS, provided partial funds for this
project and the Division of Plant Industry, Florida Department of Agriculture and
Consumer Services provided laboratory and quarantine facilities and support. Our
sincere thanks to all of these organizations and individuals, especially Bonnie Ross.
Mention of a proprietary product does not constitute an endorsement or a recommen-
dation for its use by USDA.
REFERENCES CITED
AGASSIZ, D. 1978. Five introduced species, including one new to science, of China
mark moths (Lepidoptera: Pyralidae) new to Britain. Entomol. Gaz. 29: 117-
127.
AGASSIZ, D. 1981. Further introduced China mark moths (Lepidoptera: Pyralidae)
new to Britain. Entomol. Gaz. 32: 21-26.
BALCIUNAS, J. K., AND T. D. CENTER 1981. Preliminary host specificity tests of a Pan-
amanian Parapoynx as a potential biological control agent for hydrilla. Envi-
ron. Entomol. 10: 462-467.
BALCIUNAS, J. K., AND D. H. HABECK. 1981. Recent range extensions of a hydrilla-
damaging moth, Parapoynx diminutalis (Lepidoptera: Pyralidae). Florida En-
tomol. 64: 195-196.
BALCIUNAS, J. K., AND M. C. MINNO. 1985. Insects damaging hydrilla in the U.S.A., J.
Aquat. Plant Manage. 23: 77-83.
BALOCH, G. M., AND SANA-ULLAH. 1974. Insects and other organisms associated with
Hydrilla verticillata (L.f.) L.C. (Hydrocharitaceae) in Pakistan, pp. 61-66 in A.
J. Wapshere [ed.], Proc. 3rd Int. Symp. Biol. Control Weeds, Montpellier,
France, 1973. Commonw. Inst. Biol. Control Misc. Publ. No. 8.
BUCKINGHAM, G. R., AND C. A. BENNETT. 1989. Laboratory host range of Parapoynx
diminutalis (Lepidoptera: Pyralidae), an Asian aquatic moth adventive in Flor-
ida and Panama on Hydrilla verticillata (Hydrocharitaceae). Environ. Ento-
mol. 18: 526-530.
CHANTARAPRAPHA, N., AND J. A. LITSINGER 1986. Host range and biology of three rice
caseworms. IRRN 11(5): 33-34.
DEL FOSSE, E. S., B. D. PERKINS, AND K. K. STEWARD. 1976. A new record for Parapo-
ynx diminutalis (Lepidoptera: Pyralidae), a possible biological agent for Hyd-
rilla verticillata. Florida Entomol. 59: 19-20.
LANGELAND, K. A. 1990. Hydrilla (Hydrilla verticillata (L. fil.) Royle), a continuing
problem in Florida waters. Univ. Florida Coop. Ext. Serv. Circ. 884.
LEKIC, M. 1971. Ecology of the aquatic insect species Parapoynx stratiotata L.
(Pyraustidae, Lepidoptera). J. Sci. Agric. Res. 23: 49-63.
MUNROE, E. 1973. Pyraloidea. Pyralidae (Part), Fasc. 13.1C, pp. 251-304 in R. B. Do-
minick et al., The Moths of America North of Mexico. E. W. Classey, Ltd., Lon-
don.
SPEIDEL, W 1984. Revision der Acentropinae des palaearktischen Faunengebietes
(Lepidoptera: Crambidae). Neue Entomol. Nachrichten 12: 1-157.
VARGHESE, G., AND G. SINGH. 1976. Progress in the search for natural enemies of hy-
drilla in Malaysia, pp. 341-352 in C. K. Varshney and J. Rzoska [eds.], Aquatic
Weeds in South East Asia. W. Junk, The Hague.
VIRAKTAMATH, C. A., M. PUTTARUDRIAH, AND G. P. CHANNA-BASAVANNA. 1974. Stud-
ies on the biology of rice case-worm, Nymphula depunctalis Guenee (Pyraus-
tidae: Lepidoptera). Mysore J. Agric. Sci. 8: 234-241.
Buckingham and Bennett: Biology of Parapoynx diminutalis 363
WILSON, L. T., AND W. W. BARNETT. 1983. Degree-days: an aid in crop and pest man-
agement. California Agric. 37: 4-7.
YOSHIYASU, Y. 1985. A systematic study of the Nymphulinae and the Musotiminae of
Japan (Lepidoptera: Pyralidae). Kyoto Prefectural University, Agriculture, Sci-
entific Report 37: 1-162.
YOSHIYASU, Y. 1992. Nymphulinae, pp. 79-80 in J. B. Heppner and H. Inoue [eds.],
Lepidoptera of Taiwan, Vol. 1, Pt. 2: checklist. Scientific Publishers, Inc.,
Gainesville, Florida.
Jansson: Infectivity and Reproduction of Nematodes
INFECTIVITY AND REPRODUCTION OF THREE
HETERORHABDITID NEMATODES (RHABDITIDA:
HETERORHABDITIDAE) IN TWO INSECT HOSTS
RICHARD K. JANSSON1
Merck Research Laboratories
P.O. Box 450, Hillsborough Road
Three Bridges, NJ 08887
'Previous address, University of Florida, Institute of Food and Agricultural Sciences,
Tropical Research and Education Center, Homestead, FL 33031
ABSTRACT
The infectivity, incubation time, and reproduction of three heterorhabditid nema-
todes, Heterorhabditis sp. Bacardis and FL2122 strains and H. bacteriophora Poinar
HP88 strain were studied in two insect hosts, an apionid weevil, Cylas formicarius
(F.), and a pyralid moth, Galleria mellonella (L.). Two of the nematodes, Heterorhab-
ditis sp. Bacardis and FL2122 strains, were of tropical or subtropical origin, whereas
the third nematode, H. bacteriophora HP88 strain, was of temperate origin. Infectiv-
ity did not differ among nematodes within each host; however, it did differ between
hosts for the Bacardis strain. Cylas formicarius was more susceptible to this nema-
tode than G. mellonella. Incubation times also did not differ among nematodes within
hosts; however, incubation times were 3.2-4.3 d shorter in C. formicarius than in G.
mellonella. Progeny production differed (although not significantly consistent) among
nematodes and was highest for Heterorhabditis sp. Bacardis followed by the Heter-
orhabditis sp. FL2122 and H. bacteriophora HP88 in both hosts. Percentages of in-
fected cadavers that produced progeny were consistently higher for the tropical and
subtropical nematodes, Heterorhabditis sp. Bacardis and FL2122, than for the tem-
perate nematode, H. bacteriophora HP88, in both hosts. Patterns of emergence from
cadavers were consistent in G. mellonella; most progeny emerged by 23 d after inocu-
lation and emergence lasted for up to 48 d after inoculation. Conversely, emergence
patterns varied markedly in C. formicarius. Emergence lasted for up to 29 d after in-
oculation and peak emergence varied between 12 and 28 d after inoculation. Progeny
production in C. formicarius was not related to the biomass of the host cadaver.
Key Words: Heterorhabditis spp., Cylas formicarius, Galleria mellonella, infectivity
Florida Entomologist 79(3)
RESUME
Fueron estudiadas la infectividad, el tiempo de incubaci6n y la reproducci6n de
tres nematodos heterorhabditidos, las cepas Bacardis y FL2122 de Heterorhabditis
sp., y la cepa HP88 de H. bacteriophora Poinar, en dos insects hospedantes, un gor-
gojo api6nido, Cylas formicarius (F.), y una polilla piralida, Galleria mellonella (L.).
Dos de los nematodos, las cepas Bacardis y FL2122 de Heterorhabditis sp., fueron de
origen tropical o subtropical, mientras que el tercer nematodo, la cepa HP88 de H.
bacteriophora Poinar, fue de origen templado. La infectividad no difirio entire los ne-
matodos dentro de cada hospedante; sin embargo, difirio entire los hospedantes para
la cepa Bacardis. Cylas formicarius fue mas susceptible a este nematodo que G. me-
llonella. Los periods de incubaci6n tampoco difirieron entire los nematodos dentro de
los hospedantes; sin embargo, los tiempos de incubaci6n fueron 3.2-4.3 dias mas cor-
tos en C. formicarius que en G. mellonella. La producci6n de progenie difirio (aunque
no con significaci6n consistent) entire los nematodos y fue mas alta para Heterorha-
bditis sp. Bacardis, seguida por Heterorhabditis sp. FL2122 y por H. bacteriophora
HP88 en ambos hospedantes. Los porcentajes de cadaveres infectados que produjeron
progenie fueron consistentemente mas altos para los nematodos tropicales y subtro-
picales, Heterorhabditis sp. Bacardis y Heterorhabditis sp. FL2122, que para el tem-
plado, H. bactriophora HP88, en ambos hospedantes. Los patrons de emergencia de
los cadaveres fueron consistentes en G. mellonella; la mayoria de la progenie emergio
antes de los 23 dias posteriores a la inoculaci6n y la emergencia dur6 al menos 48 dias
despues de la inoculaci6n. Por el contrario, los patrons de emergencia variaron mar-
cadamente entire los 12 y los 28 dias despues de la inoculaci6n. La producci6n de pro-
genie en C. formicarius no estuvo relacionada con la biomasa del cadaver del
hospedante.
Two population parameters of entomopathogenic nematodes that affect their suit-
ability as a biological control agent against specific target insects are their level ofin-
fectivity and reproductive capacity. Infectivity refers to the ability of nematodes to
cause infection in a target insect (Tanada & Fuxa 1989) and has been shown to vary
among nematodes within specific target hosts (Bedding et al. 1983, Molyneux et al.
1983, Morris et al. 1990, Mannion 1992) and among hosts for a given nematode spe-
cies or strain (Bedding et al. 1983, Morris et al. 1990). The reproductive capacity of
nematodes was also shown to differ among nematodes within target insects (Morris et
al. 1990, Mannion & Jansson 1992), and among hosts within specific nematode spe-
cies or strains (Morris et al. 1990). Nematodes with higher levels of infectivity and re-
production within a specific target host may be more effective at controlling a
particular insect under field conditions. These two population parameters are also
central to long-term persistence. Morris et al. (1990) noted that a high infection rate
of soil insects followed by a high rate of reproduction is critical to ensure reinfestation
of the habitat by nematode progeny.
Recent studies conducted in the laboratory that used a variety of different bioas-
say systems showed that heterorhabditid nematodes, especially an undescribed nem-
atode isolated in Florida, Heterorhabditis sp. FL2122 strain, were most suitable as
biological control agents of the sweetpotato weevil, Cylas formicarius (F.) (Mannion
1992). She found that heterorhabditid nematodes had some of the lowest LC,, and
LC0, values, produced more progeny per cadaver, had higher levels of infectivity in
sand, soil, and Petri plates, killed more hosts within sweet potato storage roots, and
had a greater ability to exit infected weevil cadavers within storage roots and infect
new hosts in the soil than steinernematid nematodes. Her data concur with previous
September, 1996
Jansson: Infectivity and Reproduction of Nematodes
reports (Jansson 1991, Jansson et al. 1990, 1992, 1993) that found heterorhabditid
nematodes to be more efficacious against C. formicarius. Jansson et al. (1992, 1993)
also found that heterorhabditid nematodes persisted longer than steinernematids in
the field.
A recent study that determined the potential for applying nematode-infected wax
moth, Galleria mellonella (L.), cadavers for controlling C. formicarius in the field
found that an undescribed nematode isolated from Puerto Rico, Heterorhabditis sp.
Bacardis strain, produced more progeny per G. mellonella cadaver than H. bacterio-
phora Poinar HP88 strain (Jansson et al. 1993). The present studies were conducted
to more fully compare the suitability of certain nematodes as biological control agents
of C. formicarius and determined if fitness parameters of three heterorhabditid nem-
atodes differed between two insect hosts, C. formicarius and G. mellonella.
MATERIALS AND METHODS
Two insect hosts were used in these experiments: late instar wax moth, G. mel-
lonella, and third instar sweetpotato weevil, C. formicarius. Galleria mellonella lar-
vae were obtained from a commercial supplier (JA-DA Bait, Antigo, Wisconsin). Cylas
formicarius were reared in the laboratory using methods described previously (Man-
nion & Jansson 1992).
Three heterorhabditid nematodes were tested: H. bacteriophora HP88 strain, and
two undescribed nematodes, Heterorhabditis sp. FL2122 and Bacardis. The two latter
nematodes were of tropical or subtropical origin. The FL2122 isolate was found in cen-
tral Florida and the Bacardis nematode was isolated near the San Juan harbor in Pu-
erto Rico (R. K. J. et al., unpublished). These two undescribed isolates are presumably
local variants of a previously unrecorded species of Heterorhabditis (J. Curran, per-
sonal communication) reported earlier as Heterorhabditis sp. B in Poinar (1990). The
HP88 strain was of temperate origin (Poinar 1990). Nematodes were reared in vivo
(Dutky et al. 1964) in G. mellonella larvae as previously described (Mannion & Jans-
son 1992). Infective juveniles used in these tests were less than two weeks old at the
time of the experiment.
Infectivity
The level of infectivity was determined in each host using a single nematode/single
host bioassay system (Miller 1989). Single infective juveniles were removed from an
aqueous suspension taken from laboratory cultures and pipetted (0.3 ml) onto a dou-
ble layer of filter paper in individual wells (1.5 cm diam) of a MultiwellT 24 well, flat
bottom tissue culture plate (FalconR, model 3047, Becton Dickinson and Co., Lincoln
Park, New Jersey). One host larva was then placed on the filter paper in each well. Fil-
ter paper treated with deionized water served as the control. Microsoap (Interna-
tional Products Corp., Burlington, New Jersey) was added (1.25 ml/liter) to nematode
suspensions to improve efficiency of nematode transfer. Tissue culture plates were
covered, sealed with parafilm and stored in the dark at 252C. Larvae were checked
daily for nematode-induced mortality for four consecutive days. Two trials were con-
ducted against G. mellonella larvae and five trials were conducted against C. formi-
carius larvae. A total of four (G. mellonella) and twelve (C. formicarius) tissue culture
plates (96 and 288 larvae, respectively) per nematode treatment were used to deter-
mine infectivity of these nematodes.
Efficiency of nematode transfer was determined by removing an aliquot (0.3 ml)
with a single nematode from suspensions, dispensing each aliquot onto a Petri plate,
Florida Entomologist 79(3)
and then counting the number of infective juveniles in each aliquot. A total of 80 ali-
quots were removed from each suspension on the day of the experiment. The efficiency
of transferring single nematodes were 93.73.7, 97.51.4, and 95.02.9 for H. bacte-
riophora HP88, Heterorhabditis sp. FL2122, and Heterorhabditis sp. Bacardis, re-
spectively. These efficiencies were then used to adjust infectivity data to estimate real
infectivity of each nematode.
Incubation Time and Reproduction
All nematode-infected G. mellonella larvae and only those C. formicarius larvae in-
fected in the first three trials were removed from the tissue culture plates and placed
in individual, modified White traps (White 1924) and incubated in the dark at 252C.
Cadavers were inspected daily for nematode emergence. The time required for infec-
tive juveniles to emerge from each cadaver was recorded. The percentage of cadavers
that produced infective juveniles was recorded in both trials with G. mellonella, but
in only the first and third trials with C. formicarius. Once emergence began, infective
juveniles were removed from the outer well of each White trap once per week and
counted. The time that emergence ceased was also recorded at which time cadavers
were dissected and the numbers of infective juveniles within each cadaver were re-
corded. In the first two trials with C. formicarius, the biomass of each larva was re-
corded before each trial to determine if progeny production was related to host
biomass.
Data Analysis
Most data were analyzed by least squares analysis of variance or regression tech-
niques, accordingly (Zar 1984). Percentage infectivity was compared among nema-
todes within hosts by chi-square analysis (Conover 1980). Mean incubation periods
and numbers of progeny produced were compared among nematodes by the Waller-
Duncan K-ratio t test (Waller & Duncan 1984). Numbers of progeny produced per C.
formicarius cadaver in the first two trials were pooled and regressed on biomass to de-
termine if reproduction was dependent upon host biomass.
RESULTS
Infectivity
Infectivity to G. mellonella larvae did not differ (X = 1.71, df = 2, P > 0.05) among
nematodes. Adjusted percentages of infective juveniles that invaded larvae were 15.4,
16.0, and 26.7% for Heterorhabditis sp. Bacardis, Heterorhabditis sp. FL2122, and H.
bacteriophora HP88, respectively.
Infectivity to C. formicarius also did not differ (X = 1.9, df = 2, P > 0.05) among
nematodes. Adjusted percentages of infective juveniles that invaded C. formicarius
larvae were 30.3, 25.9, and 20.4% for Heterorhabditis sp. Bacardis, Heterorhabditis
sp. FL2122, and H. bacteriophora HP88, respectively.
Infectivity of the Heterorhabditis sp. Bacardis nematode was affected by the insect
host (X = 4.9, df = 1, P < 0.05). A higher percentage of infective juveniles invaded C.
formicarius larvae than G. mellonella larvae. Infectivity of Heterorhabditis sp.
FL2122 and H. bacteriophora HP88 did not differ between hosts (X2 2.3, df = 1, P >
0.05).
September, 1996
Jansson: Infectivity and Reproduction of Nematodes
Incubation Time
The incubation times of nematodes within the two hosts also did not differ (F < 1.1,
df = 2,4, P > 0.05) among nematodes. In G. mellonella larvae, Heterorhabditis sp. Bac-
ardis, Heterorhabditis sp. FL2122, and H. bacteriophora HP88 required 11.90.1,
12.10.1, and 12.00.1 d, respectively, to emerge from cadavers. In C. formicarius lar-
vae, these nematodes required only 8.70.2, 7.80.2, and 8.10.2 d, respectively, to
emerge from hosts. Incubation periods for each nematode were significantly shorter in
C. formicarius than in G. mellonella, which is probably related to differences in host
size.
Reproduction
In the first trial, percentages of G. mellonella cadavers that produced progeny dif-
fered among nematodes (X = 10.7, df = 2, P < 0.01). Higher percentages of G. mel-
lonella infected with Heterorhabditis sp. Bacardis (66.7%; n = 6) and FL2122 (71.4%;
n = 7) produced progeny than those infected with H. bacteriophora HP88 (38.5%; n =
13). In the second trial, percentages of cadavers that produced progeny did not differ
among nematodes (X = 3.6, df = 2, P> 0.05). All nematode-infected cadavers produced
progeny in high percentages and were 75% (n = 8), 100% (n = 8), and 91% (n = 11) for
Bacardis, FL2122, and HP88, respectively.
Patterns of progeny production were consistent for each nematode in G. mellonella
larvae (Fig. 1). Most progeny emerged from cadavers infected with Heterorhabditis sp.
Bacardis (67.7-79.3%), Heterorhabditis sp. FL2122 (69.0-81.3%) and H. bacteriophora
HP88 (85.6-96.2%) within 23 d after inoculation. Emergence of progeny declined 23 d
after inoculation for all nematodes and lasted for up to 48 d after inoculation.
Total progeny production in G. mellonella cadavers did not differ (F = 2.3, df = 2,12,
P > 0.05) among nematodes in the first trial, but did differ (F = 4.4, df = 2,12, P < 0.05)
among nematodes in the second trial (Table 1). Trends in the data were consistent be-
tween the two trials. More progeny were consistently produced by Heterorhabditis sp.
Bacardis followed in decreasing order by Heterorhabditis sp. FL2122 and H. bacterio-
phora HP88. Considerably more progeny were produced in the second trial than in the
first trial, and the reasons for this are unclear.
Percentages of C. formicarius cadavers that produced progeny differed among
nematodes in the two trials (X > 15.3, df = 2, P < 0.001). In the first trial, the highest
percentages of C. formicarius cadavers produced progeny when infected with Heter-
orhabditis sp. Bacardis (90%; n = 20) followed in decreasing order by those infected
with Heterorhabditis sp. FL2122 (41.7%; n = 12) and H. bacteriophora HP88 (35.3%;
n = 17). In the third trial, cadavers infected with Heterorhabditis sp. FL2122 had the
highest percentage of nematode production (90%, n = 20) followed by Heterorhabditis
sp. Bacardis (58.8%; n = 17) and H. bacteriophora HP88 (46.7%; n = 15).
No consistent patterns of emergence of infective juveniles from C. formicarius ca-
davers were found in the three trials (Fig. 2). In the first trial, emergence of Heter-
orhabditis sp. Bacardis peaked 19 d after inoculation, whereas those of H.
bacteriophora HP88 and Heterorhabditis sp. FL2122 peaked 28 d after inoculation. In
trial 2, emergence of H. bacteriophora HP88 peaked 12 d after inoculation, whereas
those of Heterorhabditis sp. FL2122 and Bacardis peaked 19 d after inoculation. Few
progeny emerged after 19 d. In the third trial, emergence of all three nematodes
peaked 16 d after inoculation and few progeny emerged after 16 d.
Total progeny production in C. formicarius cadavers did not differ (F < 1.3, df =
2,24, P > 0.05) among nematodes in the first two trials, but did differ (F = 11.2, df =
2,32, P < 0.001) in the third trial (Table 1). As found in G. mellonella, more progeny
Florida Entomologist 79(3)
Mean No. of IJs/Larva (Thousands)
16 23 30 38 47
Days After Inoculation
16 23 29 35 42 48
Days After Inoculation
Fig. 1. Emergence patterns of infective juveniles from G. mellonella cadavers in-
fected with three different heterorhabditid nematodes: Heterorhabditis sp. Bacardis
strain, Heterorhabditis sp. FL2122 strain, and H. bacteriophora HP88 strain. Num-
bers counted on the last sample date of each trial are the sum of those that emerged
plus any remaining infective juveniles within cadavers.
September, 1996
Jansson: Infectivity and Reproduction of Nematodes
TABLE 1. PROGENY PRODUCTION OF THREE HETERORHABDITID NEMATODES IN TWO IN-
SECT HOSTS, CYLAS FORMICARIUS AND GALLERIA MELONELLA.
Trial
Host x Nematode 1 2 3
C. formicarius
Heterorhabditis sp. 6,522.7(1,844.0)a' 7,919.5(1,061.0)a 8,625.6(1,162.6)a
Bacardis
Heterorhabditis sp. 5,610.6(1,086.5)a 6,896.8(1,167.1)a 4,834.6(676.7)b
FL2122
H. bacteriophora 1,880.8(469.0)a 3,387.0(259.0)a 2,071.6(388.7)c
HP88
G. mellonella
Heterorhabditis sp. 194,222.5(54,228.0)a 238,692.2(45,442.2)a
Bacardis
Heterorhabditis sp. 122,580.2(34,579.4)a 229,903.4(30,825.3)a
FL2122
H. bacteriophora 74,992.5(30,940.0)a 137,228.8(13,072.1)b
HP88
'Data are means + sem. Means within a column followed by the same letter do not differ by the Waller-Dun-
can K-ratio t test (Waller & Duncan,1969).
were consistently produced (although not consistently significant) by Heterorhabditis
sp. Bacardis followed in decreasing order by Heterorhabditis sp. FL2122 and H. bac-
teriophora HP88.
The nematode host influenced progeny production. Approximately 28.1-, 30.5-,
and 43.4-fold more progeny were produced in G. mellonella than in C. formicarius for
Heterorhabditis sp. Bacardis, Heterorhabditis sp. FL2122, and H. bacteriophora
HP88, respectively.
Progeny production in C. formicarius was not related to host biomass for any of the
three nematodes (Bacardis: Y = 1054 + 646713X, F = 1.3, df = 1,31, P > 0.05, r = 0.04;
FL2122: Y = 91 + 687518X, F= 1.2, df = 1,13, P > 0.05, r2 = 0.09; HP88: Y = -1901 +
510346X, F = 2.4, df = 1,6, P > 0.05, r2 = 0.29).
DISCUSSION
Few differences in infectivity were found among nematodes within insect hosts;
however, infectivity differed between hosts for Heterorhabditis sp. Bacardis strain.
This nematode was more infective against the weevil, C. formicarius, than against the
lepidopteran, G. mellonella, despite the fact that these nematodes were all reared in
vivo in G. mellonella before experiments. Bedding et al. (1983) showed that H. bacte-
riophora was least infective against the host from which it was isolated, Heliothis
punctigera (Wallengren). They also found that Steinernema feltiae (Filipjev) (= Neoa-
plectana bibionis) isolated from Lucilia cuprina (Wiedemann) and Otiorhynchus sul-
catus (F.) was least infective to these insects. Kaya (1987) suggested that the host
from which a nematode is isolated in soil probably has little, if any, affect on the suit-
ability of the nematode/host encounter. Entomopathogenic nematodes attack a broad
Florida Entomologist 79(3)
WI, n NW. vt fTmoI Real mIau t'rlswmu>)
Trial 1 --I "- Trial 2
1- 7 I Ii
hMn HOl U -t.v. 4Tha fl
Tria3
Days AlMr Inaouaon
Fig. 2. Emergence patterns of infective juveniles from C. formicarius cadavers in-
fected with three different heterorhabditid nematodes: Heterorhabditis sp. Bacardis
strain, Heterorhabditis sp. FL2122 strain, and H. bacteriophora HP88 strain. Num-
bers counted on the last sample date of each trial are the sum of those that emerged
plus any remaining infective juveniles within cadavers.
spectrum of insects; thus, isolation from soil is, in part, due to a chance encounter be-
tween the isolation host and the nematode. Kaya (1987) also suggested that, in cer-
tain cases, continued association with the same insect species may reduce virulence
rather than enhance it.
Incubation times also did not differ among nematodes within each species; how-
ever, incubation times were considerably shorter for all nematodes in the weevil, C.
formicarius, than in the lepidopteran, G. mellonella. It is well known that emergence
of infective juveniles is related to depletion of food reserves and crowding within the
host cadaver (Kaya 1985, 1987).
Patterns in total reproduction of nematodes differed among the three nematodes
in both hosts. Heterorhabditis sp. Bacardis consistently produced more (although not
consistently significant) progeny than the other two nematodes in both hosts. These
data concur with previous studies (Jansson & Lecrone 1994, Jansson et al. 1993)
which showed that Heterorhabditis sp. Bacardis produced more progeny per cadaver
than H. bacteriophora HP88 in G. mellonella larvae. Progeny production of Bacardis
was higher in the previous study (range, 272,576-396,598 per cadaver) than in the
present study (range, 137,229-238,692 per cadaver). However, a dose of 20 infective
juveniles per larva was used in the previous study compared with only 1 per larva in
the present study. Progeny production of H. bacteriophora HP88 was comparable be-
tween the two studies despite large differences in dose (range of previous study,
76,260-219,181 per cadaver; range of present study, 74,992-194,222 per cadaver).
September, 1996
Jansson: Infectivity and Reproduction of Nematodes
Patterns of emergence from cadavers were consistent in G. mellonella, but not in
C. formicarius. In G. mellonella, most infective juveniles emerged within 23 d after in-
oculation. In C. formicarius, emergence patterns varied among trials. As noted earlier,
emergence of infective juveniles is related to depletion of food reserves and crowding
(Kaya 1985, 1987). These factors may have been less apparent to emerging infective
juveniles from C. formicarius due to the smaller size of this host compared with G.
mellonella.
Total progeny production of heterorhabditid nematodes in C. formicarius con-
curred with a previous report (Mannion & Jansson 1992), although progeny produc-
tion was higher for FL2122 and lower for HP88 in the present study. The dose used
might have affected these data. Mannion & Jansson (1992) used a dose of 25 infective
juveniles per larva compared with 1 per larva in the present study.
It is recognized that a laboratory bioassay that predicts performance of ento-
mopathogenic nematodes in the field is needed to facilitate selection of nematodes in
biological control programs (Hominick 1990, Mannion 1992). Mannion (1992) con-
ducted Petri dish, sand, soil, and simulated field bioassays to select suitable ento-
mopathogenic nematodes for the biological control of C. formicarius and consistently
found that heterorhabditids were superior to steinernematids in all bioassay systems
tested. The present bioassay system may also have potential for selecting suitable en-
tomopathogenic nematodes, especially heterorhabditid nematodes, for C. formicarius.
Morris et al. (1990) noted that both infectivity and reproduction within hosts were
important attributes of nematodes capable of reinfesting new hosts in the field. The
present study demonstrated that Heterorhabditis sp. Bacardis was most infective and
reproductive against C. formicarius, and, for this reason, it should be a suitable nem-
atode for use in biological control of this weevil. Recent results from field studies con-
firm this belief. Heterorhabditis sp. Bacardis was shown to be very efficacious at
controlling weevil damage to storage roots and persisted at levels higher than those of
all other nematodes tested, including H. bacteriophora HP88, S. carpocapsae (Weiser)
All and S20, and S. feltiae N27 (Jansson et al. 1993). Collectively, these data suggest
that the use of this single nematode/single host bioassay may be an important tool for
identifying potential candidate heterorhabditid nematodes in biological control pro-
grams of target insect pests. More work is needed, however, to confirm this belief.
These data also suggest that certain Neotropical and subtropical nematodes, Het-
erorhabditis sp. Bacardis and FL2122, may be more suitable as biological control
agents against the pantropical sweetpotato weevil, C. formicarius, than the temper-
ate nematode, H. bacteriophora HP88 strain. Recent studies by Lawrence (1994) also
suggested that certain Neotropical nematodes may be more suitable as biological con-
trol agents of this weevil; however, not all tropical and subtropical isolates were supe-
rior to all temperate isolates.
ACKNOWLEDGMENTS
I thank K. Ericsson for assistance with data collection and R. Miller (Biosys Inc.,
Columbia, MD), R. Gaugler (Rutgers University, New Brunswick, NJ), and C. M.
Mannion and R. J. Zimmerman (University of Florida, I.F.A.S., T.R.E.C., Homestead)
for critically reviewing the manuscript. Thanks are also extended to R. Miller for tech-
nical assistance. I thank G. C. Smart, Jr. (University of Florida, Gainesville) for pro-
viding Heterorhabditis sp. FL2122. Heterorhabditis sp. Bacardis was collected near
San Juan, Puerto Rico by W. Figueroa (University of Puerto Rico, Rio Piedras) as part
of a study [supported by the U.S. Department of Agriculture under CSRS Special
Grant No. 91-34135-6134 (to R.K.J.)] to survey the Caribbean for entomopathogenic
Florida Entomologist 79(3)
nematodes. These nematodes are currently being held at the University of Florida,
I.F.A.S., Entomology and Nematology Department, Gainesville, FL and at the Depart-
ment of Entomology, Rutgers University. The present research was supported by the
U.S. Department of Agriculture under CSRS Special Grant No. 91-34135-6134 (to
R.K.J.) managed by the Caribbean Basin Administrative Group (CBAG). Research
was conducted at the University of Florida, I.F.A.S., T.R.E.C., Homestead. The manu-
script was prepared, in part, at Merck & Co., Inc.
REFERENCES
BEDDING, R. A., A. S. MOLYNEUX, AND R. J. AKHURST. 1983. Heterorhabditis spp.,
Neoaplectana spp., and Steinernema kraussei: interspecific and intraspecific
differences in the infectivity for insect hosts. Exp. Parasitol. 55: 249-257.
CONOVER, W. J. 1980. Practical nonparametric statistics, 2nd ed. J. Wiley, New York.
DUTKY, S. R., J. V. THOMPSON, AND G. E. CANTWEL. 1964. A technique for the mass
propagation of the DD-136 nematode. J. Insect Pathol. 6: 417-422.
HOMINICK, W. M., AND A. P. REID. 1990. Perspectives in entomopathogenic nematol-
ogy, pp. 327-345 in R. Gaugler and H. K. Kaya [eds.], Entomopathogenic nem-
atodes in biological control. CRC Press, Boca Raton, FL.
JANSSON, R. K. 1991. Biological control of Cylas spp., pp. 169-201 in R. K. Jansson,
and K. V. Raman [eds.], Sweet potato pest management: a global perspective.
Westview Press, Boulder and London.
JANSSON, R. K., AND S. H. LECRONE. 1994. Application methods for entomopathogenic
nematodes (Rhabditida: Heterorhabditidae): aqueous suspensions versus in-
fected cadavers. Florida Entomol. 77: 281-284.
JANSSON, R. K., S. H. LECRONE, AND R. GAUGLER 1991. Comparison of single and
multiple releases of Heterorhabditis bacteriophora Poinar (Nematoda: Heter-
orhabditidae) for control of Cylas formicarius (F.) (Coleoptera: Apionidae). Biol.
Control. 1: 320-328.
JANSSON, R. K., S. H. LECRONE, AND R. GAUGLER. 1993. Field efficacy and persistence
of entomopathogenic nematodes (Nematoda: Steinernematidae, Heterorhab-
ditidae) for control of sweetpotato weevil (Coleoptera: Apionidae) in southern
Florida. J. Econ. Entomol. 86: 1055-1063.
JANSSON, R. K, S. H. LECRONE, R. R. GAUGLER, AND G. C. SMART, JR. 1990. Potential
of entomopathogenic nematodes as biological control agents of sweetpotato
weevil (Coleoptera: Curculionidae). J. Econ. Entomol. 83: 1818-1826.
KAYA, H. K. 1985. Entomogenous nematodes for insect control in IPM systems, pp.
283-302 in M. A. Hoy and D. C. Herzog [eds.], Biological control in agricultural
IPM systems. Academic, New York.
KAYA, H. K. 1987. Diseases caused by nematodes, pp. 453-470 in J. R. Fuxa and Y.
Tanada [eds.], Epizootiology of insect diseases. J. Wiley, New York.
LAWRENCE, J. L. 1994. Potential of Neotropical and temperate Heterorhabditid nem-
atodes as biological control agents for the sweetpotato weevil, Cylas formicar-
ius (Fabricius) (Coleoptera: Apionidae). M.S. Thesis, University of Florida,
Gainesville, FL.
MANNION, C. M. 1992. Selection of suitable entomopathogenic nematodes for biologi-
cal control of Cylas formicarius (Coleoptera: Apionidae). Ph.D. Diss., Univer-
sity of Florida, Gainesville, FL.
MANNION, C. M., AND R. K. JANSSON. 1992. Comparison of ten entomopathogenic
nematodes for biological control of the sweetpotato weevil (Coleoptera: Apion-
idae). J. Econ. Entomol. 85: 1642-1650.
MILLER, R. W. 1989. Novel pathogenicity assessment technique for Steinernema and
Heterorhabditis entomopathogenic nematodes. J. Nematol. 21: 574 (abstract).
MOLYNEUX, A. S., R. A. BEDDING, AND R. J. AKHURST. 1983. Susceptibility of larvae of
the sheep blowfly Lucilia cuprina to various Heterorhabditis spp., Neoaplec-
tana spp., and an undescribed steinernematid (Nematoda). J. Invertebr.
Pathol. 42: 1-7.
September, 1996
Jansson: Infectivity and Reproduction of Nematodes 373
MORRIS, 0. N., V. CONVERSE, AND J. HARDING. 1990. Virulence of entompathogenic
nematode-bacteria complexes for larvae of noctuids, a geometrid, and a pyralid.
Canadian Entomol. 122: 309-319.
POINAR, G. 0., JR. 1990. Taxonomy and biology of Steinernematidae and Heterorhab-
ditidae, pp. 23-61 in R. Gaugler and H. K. Kaya [eds.], Entomopathogenic nem-
atodes in biological control. CRC Press, Boca Raton.
TANADA, Y., AND J. R. FUXA, J. R. 1987. The pathogen population, pp. 113-157 in J. R.
Fuxa and Y. Tanada [eds.], Epizootiology of insect diseases. J. Wiley, New York.
WALLER, R. A., AND D. B. DUNCAN. 1969. A Bayes rule for the symmetric multiple
comparison problem. J. American Stat. Assoc. 64: 1484-1489.
WHITE, G. F. 1924. A method for obtaining infective nematode larvae from cultures.
Science 66: 302-303.
ZAR, J. H. 1984. Biostatistical analysis. Prentice-Hall, Englewood Cliffs, NJ.
6666666666666666666666666666666666666666666666666666
Morse and Lindegren: Suppression of Fuller Rose Beetle
SUPPRESSION OF FULLER ROSE BEETLE (COLEOPTERA:
CURCULIONIDAE) ON CITRUS WITH STEINERNEMA
CARPOCAPSAE (RHABDITIDA: STEINERNEMATIDAE)
JOSEPH G. MORSE' AND JAMES E. LINDEGREN2'3
'Department of Entomology
University of California
Riverside, CA 92521
2USDA ARS Rt.
Horticultural Crops Research Laboratory
Commodity Protection & Quarantine Insect Research Unit
2021 South Peach Avenue
Fresno, CA 93727
"Current address: Kapow Consulting, 2826 E. Los Altos, Fresno, CA 93710
ABSTRACT
A laboratory bioassay with larvae and adults of the Fuller rose beetle, Asynony-
chus godmani Crotch, and the Kapow strain of Steinernema carpocapsae (Weiser), re-
sulted in 12, 44, and 67% mortality with rates of 50, 150, and 500 infective juveniles
per 3-wk old larva, respectively; 100% mortality with 150 infective juveniles per 3-mo
old larva; and 24, 48, and 83% mortality with adults. A field trial was conducted on Va-
lencia orange trees harboring high levels of Fuller rose beetle late instar larvae. A sin-
gle application of either the Kapow or All strain of S. carpocapsae each applied at 3
rates (50, 150, and 500 infective juveniles per cm2) reduced the number of emerging
adult Fuller rose beetles a combined 55 and 38% compared with the water control the
year following treatment and 79 and 82%, respectively, the 2nd year. Because of high
variability between treatments, however, it was difficult to choose between the two
nematode strains or the 3 rates of each strain. Infested fruit was reduced by a com-
bined mean level of 62% one year after treatment. Based on nematode recovery at 6
months and the further reduction of Fuller rose beetle emergence in the second year
Florida Entomologist 79(3)
after application, we suspect that nematodes persisted and recycled in the soil and
provided added control in the second year of the trial.
In order to conduct the laboratory bioassay, a method of rearing Fuller rose beetle
from egg to adult was developed. A corn rootworm artificial diet was used but resulted
in high larval mortality. After 6 months on the diet under laboratory conditions, 12
parthenogenetic females resulted from approximately 1,000 eggs and 8 of these fe-
males produced viable eggs masses.
Key Words: Asynonychus godmani, parasitic nematodes, Kapow strain, All strain,
BiovectorT, biological control
RESUME
Un bioensayo con larvas y adults de Asynonychus godmani Crotch, y la cepa Ka-
pow de Steinernema carpocapsae (Weiser), dio como resultado 12, 44, y 67% de mor-
talidad con dosis de 50, 150 y 500 juveniles infectivos por larva de 3 dias de edad,
repectivamente; 100% de mortalidad con 150 juveniles infectivos por larvas de 3 me-
ses de edad, y 24, 48 y 83% de mortalidad con adults. Un esayo de campo fue condu-
cido en arboles de nanranjo Valencia, con altos niveles de instares tardios de A.
godmani. Una sola aplicaci6n de la cepas Kapow o All de S. carpocapsae aplicadas
cada una a 3 dosis (50, 150, y 500 juveniles infectivos por cm2) redujo el numero deA.
godmani emergidos en un 55 y 38%, comparado con un testigo con agua al afio si-
guiente al tratamiento, y 79 y 82%, respectivamente, en el segundo afio. Sin embargo,
debido a la alta variabilidad entire los tratamientos, fue dificil de escoger entire las dos
cepas del nematodo o las tres dosis de cada cepa. Los frutos infestados fueron reduci-
dos por un nivel de media combinada del 62% un aio despues del tratamiento. Basas-
dos en la recuperaci6n de nematodos en el segundo afo despues de la aplicaci6n,
sospechamos que los nematodos persistieron, se reciclaron en el suelo, y aportaron
control adicional en el segundo aio del ensayo.
Para conducir los bioensayos de laboratorio, fue desarrollado un m6todo de cria
para A. godmani desde el huevo hasta el adulto. Fue usada una dieta artificial para
gusanos de la raiz del maiz pero produjo una alta mortalidad larval. Luego de 6 meses
en la dieta bajo condiciones de laboratorio, 12 hembras partenogeneticas se obtuvie-
ron de aproximadamente 1,000 huevos y 8 de esas hembras produjeron masas viables
de huevos.
The Fuller rose beetle (FRB),Asynonychus godmani Crotch (Pantomorus cervinus
(Boheman)), is a flightless, parthenogenetic beetle which was reported in California
as early as 1879 (Chadwick 1965). Until recently, however, it was considered to be a
relatively unimportant pest of citrus in the state. In 1985, viable egg masses were
found under the calyx of fruit shipped to Japan. Japan lists FRB as a quarantine pest
and loads of fruit found to contain a single fruit infested with a viable egg mass are
fumigated with methyl bromide. Fumigation damages the fruit and is costly to the
grower.
The FRB is thought to have one generation per year in California with peak emer-
gence of adults from the soil occurring August to October although a few adults
emerge each month of the year (Morse et al. 1987). After emerging from a pupal cham-
ber in the soil, adults feed on the leaves of citrus for several wks prior to laying egg
masses in cracks and crevices in the tree and under the calyx of fruit. Eggs hatch after
3-4 wks, larvae drop to the ground, enter the soil, and feed on citrus roots for most of
a year before building a pupation chamber several centimeters below the soil surface.
September, 1996
Morse and Lindegren: Suppression of Fuller Rose Beetle 375
Methods of reducing FRB infestation of citrus shipped to Japan include: (1) moni-
toring citrus groves or fruit lots in the packinghouse and prioritizing fruit with low
levels of viable FRB egg masses for shipment to Japan, (2) skirt-pruning and trunk
treatments to exclude adult beetles from ovipositing on fruit in the tree, (3) foliar
chemical treatments to kill adults in the tree, and (4) holding lemons in cold storage
long enough to allow eggs to hatch (Morse et al. 1987, 1988, Haney & Morse 1988,
Lakin & Morse 1989). None of these options has yet proven to be totally satisfactory
due to a combination of logistic and economic constraints consistent with the need for
almost 100% control required by quarantine protocols (Morse et al. 1987). This re-
search investigates an alternative control strategy-use of the entomopathogenic
nematode, Steinernema carpocapsae (Weiser) (= S. feltiae Filipjev), to reduce FRB lar-
val populations in the soil thereby reducing subsequent adult emergence and oviposi-
tion on the fruit.
MATERIALS AND METHODS
FRB Laboratory Rearing
Adult FRB for larval rearing were collected from the Fairview citrus grove in
Hemet, CA, and approximately 25 were isolated in each of 4 ventilated plastic sweater
boxes (11 cm x 26 cm x 36 cm) with Valencia orange leaves, Citrus sinensis (L.) Os-
beck, provided as a food source, and 10 tightly folded pieces of wax paper as an ovipo-
sition substrate. Eggs were collected, allowed to hatch, and larvae were reared for 3
wks or 3 mo at room temperature (24.4 + 1.7'C) on a corn rootworm artificial diet (Diet
Premix No. 1675, Bio-Serv. Inc., Frenchtown, NJ) as suggested by Dr. W. J. Schroeder,
USDA-ARS, Orlando, FL. The diet mixture was prepared by mixing 25 liters cold wa-
ter, 750 g agar, 60 ml formaldehyde, and 10 kg diet premix. The mixture was heated
to 93"C in a mixing machine, poured while still hot into small 25-ml plastic diet cups,
dried for 5 d, and refrigerated prior to use.
Laboratory Nematode Bioassay
Preliminary laboratory bioassays with the Kapow selection of S. carpocapsae were
conducted on five dates with adult FRB, and on a single date each with 3-week old (1-
2 mm in length) FRB larvae, and a small number of 3-month old larvae (about 5 mm
in length). Adult FRB bioassays were conducted using beetles collected from a com-
mercial citrus grove in Hemet, CA. Larvae used in the bioassays were reared from
eggs collected in the laboratory.
Kapow selection infective juveniles (IJ), selected since 1984 from the Mexican
strain for increased production and pathogenicity (Agudelo-Silva et al. 1987, Linde-
gren 1990), were produced at the USDA-ARS, Horticultural Crops Research Labora-
tory, Fresno, California using the in vivo method described by Lindegren et al. (1993).
The laboratory bioassays were conducted with recently harvested Kapow infective ju-
veniles (IJ) counted by pipetting 0.02-ml drops of IJ-water suspension onto the bottom
of a 10-cm diam, 1.5-cm high, plastic petri dish. The drops were overlaid with 9-cm
diam filter paper, and deionized water was added to give a total volume of 1 ml per
dish. The dish was covered with a lid and allowed to equilibrate for 1 h, then 10 FRB
adults or larvae were added to each petri dish. Each test consisted of 10 replications
of 0, 5, 15, and 50 nematodes per FRB (i.e., 0, 50, 150, and 500 IJ per dish). Adult FRB
were bioassayed on 5 dates, 3-wk old larvae were bioassayed once, and because only
twenty 3-mo old larvae were available, a single replicate was bioassayed using a wa-
Florida Entomologist 79(3)
September, 1996
ter control and 150 IJ/ larva. The dishes were incubated at 25'C and 45% RH. Mor-
tality was evaluated after 72 h, and data were corrected for control mortality using
Abbott's formula (1925). All larval and adult mortalities were verified by dissection.
Field Trial
A mature Valencia orange grove (Block 8, Fairview Ranch) in Hemet, CA was cho-
sen for the field trial based on high levels of FRB adults observed the previous year.
Pre-treatment counts of fruit calyx infestation with FRB egg masses (both hatched
and unhatched eggs) were taken 7 June 1989 by removing and examining the calyx of
20 fruit randomly picked from the exterior canopy of each of 160 trees. Seventy trees
with the highest infestation levels were chosen and seven single tree replicates were
assigned to each of 10 blocks based on the percent of fruit infested (there were 10 rep-
licates per treatment). One tree in each block was assigned randomly to each of the
following seven treatments: water control, Kapow strain at 50, 150, and 500 nema-
todes per cm2, and All strain (BiovectorTM, Biosys, Columbia, MD) at 50, 150, and 500
per cm2 applied to the soil surface beneath the drip line of the tree. Data trees were
separated by a minimum of one untreated buffer tree in each direction down the row
and were spread uniformly over an area of 13 rows by 47 trees (611 trees total). Tree
spacing was 6.1 m both down and across the row, trees were approximately 5.5 m
high, and the tree canopy extended approximately 5.5 m in diam. Tree skirts were
pruned to 1 m above ground level and, if necessary, between data trees and adjacent
trees in a row, so that beetles could access data trees only via the trunk. Treatments
were applied 26 June 1989 using a 189-liter diaphragm pump sprayer (Lindegren et
al. 1987) to apply the nematodes in 1.9 liters of water, uniformly under the drip line
of each tree. The grove was irrigated for 9 h before and for 15 h after nematode appli-
cations were made. The irrigation system consisted of fanjet J2 minisprinklers (Tee-
jet, Wheaton, IL) positioned mid-way between each trunk down the row emitting 60.0
liters of water per h.
FRB adult emergence from the soil was monitored every two wks July to December
for two years following treatment using three 58.4-cm square emergence boxes placed
under each tree. Emergence box frames were constructed of wood, covered with
screen, and were pushed one cm into the soil to prevent beetle escape. Counted adults
were released under the sampled data tree. Fruit FRB egg mass contamination was
determined again 21 June 1990 by examining 100 fruit selected randomly from the
exterior of each data tree.
Persistence of nematodes in the soil was bioassayed on 0-, 0+ (several h before and
after treatment, respectively), 9, 15, 23, and 193 d post-treatment using soil cores
(5.0-cm diam, 7.0-cm deep) collected from 5 of the 10 replicates from each treatment
using a circular aluminum cylinder with a volume of 137 cm3. On each date, 3 soil
cores were collected 1.5 m from the trunk of each of the same 5 replicate trees (ini-
tially chosen randomly) and were shipped by 2-d mail to Fresno where bioassays were
conducted one d later using wax moth larvae, Galleria mellonella (L.). The three sam-
ples from each tree were mixed and an aliquot of 88.4 cm3 of soil was spread evenly in
individual 15-cm diam by 2.5-cm high, plastic petri dishes with 10 larvae. Mortality
was evaluated after 72 h incubation at 25C. Soil samples from +15 and +23 d were
quite dry and 10 ml of water was mixed with the soil prior to bioassays being con-
ducted.
Statistical analysis was conducted using PROC GLM in SAS (SAS Institute 1985)
and mean separation was determined using the REGWQ option.
Morse and Lindegren: Suppression of Fuller Rose Beetle 377
RESULTS
FRB Laboratory Rearing
The corn rootworm diet appears suitable for rearing FRB larvae although mortal-
ity rates were quite high especially during early larval instars. Diet moisture content
may be important and some manipulation of this might lead to increased survivor-
ship. Under laboratory conditions, 12 adults were obtained from approximately 1,000
neonate larvae placed on the diet with the larval and pupal stages lasting five and one
months, respectively. Of the 12 adults obtained, 8 laid egg masses within 4 wks of pu-
pal emergence. Egg hatch was comparable to that observed with field collected adults
and neonate larvae appeared to be normal.
Laboratory Nematode Bioassay
The laboratory bioassay with 3-week old FRB larvae and the Kapow strain of S.
carpocapsae resulted in 7% control mortality and 12, 44, and 67% corrected mortality
with rates of 50, 150, and 500 IJ per larva, respectively (Table 1). A limited number
of large FRB larvae (3 mo old, about 5 mm in length) were available and in the single
bioassay conducted, 10% control mortality was observed and 100% mortality with 150
IJ per larva.
Five laboratory bioassays were conducted with adult FRB and results were vari-
able (Table 1). On average, 7.6% control mortality was observed and 24.1, 47.6, and
83.3% corrected mortality with 50, 150, and 500 IJ per larva, respectively. Based on
the above data, a field trial was determined to be worthwhile.
Field Trial
The Valencia orange grove in Hemet chosen for the field trial had as high a level
of FRB infestation for the two years prior to the trial as we have seen in California
since 1985. As a result of choosing the most heavily infested trees for the trial and
blocking treatments based on fruit egg mass infestation levels, pre-trial levels were
similar among treatments and varied from 24.0 to 25.0% of the outside fruit infested
just prior to harvest in June, 1989 (Table 2). Petal-fall for these fruit was May, 1988
and thus, they were exposed to FRB oviposition for 13 mo prior to harvest with the
highest levels of oviposition expected approximately August to December, 1988
(Morse et al. 1987).
Nematodes were applied 26 June 1989. Morse et al. (1987) monitored FRB soil
emergence in four groves in Riverside and San Bernardino counties. One of these
groves was the same Block 8 used in this S. carpocapsae trial and a second was a
nearby grapefruit grove. Based on data from all four groves, monthly adult emergence
was 2.7, 6.8, 32.0, 34.9, 16.0, 6.0, and 1.5% of the yearly emergence during the months
of June through December, respectively, and only 0.1% of the FRB adults emerged in
the months of January through May. Thus, at the time of nematode application, we as-
sume that the majority of the FRB were in the late larval or pupal instars.
Soil laboratory wax moth bioassays indicated that nematodes initially persisted in
the soil for at least 23 d and the All strain was recovered in the persistence bioassay
193 d later (Table 2). Soil samples from the +9 d sample overheated in transport and
low bioassay mortalities for this date were suspect. Although not statistically differ-
ent than laboratory wax moth mortality seen with soil from the water control, numer-
ically higher mortality was observed with soil from the 150 and 500 nematodes per
|
RSS
