Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00017
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 2005
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00017
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

Liendo et al: Olfactory and Electroantennographic Responses of S. breve


'Universidad Sim6n Bolivar (USB), Dpto. de Quimica, Valle de Sartenejas
Apdo. 89000,Caracas 1080A-Venezuela

2Universidad Sim6n Bolivar (USB), Dpto. de Biologia de Organismos
Laboratorio de Comportamiento Valle de Sartenejas. Apdo. 89000, Caracas 1080A-Venezuela

3Instituto Nacional de Investigaciones Agricolas (INIA), Estaci6n Experimental del Estado Miranda
Caucagua, Edo. Miranda, Venezuela. Zona Postal 1246-Venezuela


With the aim of studying the olfactory behavior of one of the main pests in neotropical cocoa
plantations, the cocoa beetle Steirastoma breve (Sulzer) (Coleoptera: Cerambycidae), we
studied behavioral and antennal responses towards different odor sources in a two-choice ol-
factometer and an electroantennographic system, respectively. Odor sources tested as stim-
uli in olfactometric experiments were chopped pieces of cocoa branches, adult males, adult
females of S. breve, and combinations of these. Extracts of female and male body parts in n-
hexane were tested in electroantennnographic experiments. Statistically significant attrac-
tion responses in the olfactometer were observed only when S. breve individuals were stim-
ulated with odors from pieces of cocoa branches. Both sexes showed active EAG responses to
odors of cocoa branches, and females showed active EAG responses to adult male odors.
These results suggest that olfactory behavior of S. breve is mediated by volatiles derived
from cocoa trees and from adult male insects.

Key Words: behavior, cocoa, Theobroma cacao, pheromone, kairomone, ethological control,


Una de las principles plagas del cultivo de cacao en el neotr6pico, es la comunmente cono-
cida "Gota del Cacao", Steirastoma breve (Sulzer) (Coleoptera: Cerambycidae). A fin de cono-
cer el comportamiento olfativo de S. breve, se evalu6 la respuesta del insecto hacia diferentes
fuentes de aroma. Para ello se us6 un dispositivo olfatom6trico de dos vias de selecci6n y se
evaluaron las respuestas de antenas de machos y hembras en un electroanten6grafo. En el
olfat6metro se evaluaron tres fuentes de aroma: trozos de ramas cacao, machos adults,
hembras adults y sus combinaciones; y en el electroanten6grafo adicionalmente se midi-
eron las respuestas de las antenas hacia extractos de parties del cuerpo de machos y hem-
bras. Se observe que las inicas fuentes de aroma que demostraron un efecto significativo de
atracci6n en el comportamiento olfativo y respuestas electroantenograficas de S. breve fu-
eron: los trozos de ramas de cacao que resultaron atractivos para hembras y machos en el ol-
fat6metro, y que ademas estimularon la antena de individuos de ambos sexos; y los volatiles
producidos por el macho resultaron ser atractivos para las hembras y produjeron una fuerte
respuesta electroantenografica en la antena de la hembra. Los resultados sugieren que el
comportamiento olfativo de S. breve se encuentra modulado por cairomonas provenientes de
plants de cacao y por una feromona secretada por el macho de este insecto en su estado

Translation provided by the authors.

Various species of Coleoptera such as Sco- Cosse & Bartelt 2000). Steirastoma breve (Sulzer)
lytidae, Curculionidae, Dermestidae, Nitidulidae, (Coleoptera: Cerambycidae), known as the "cocoa
Scarabeidae, Dynastidae, and Cerambycidae beetle", is one of the main pests in cocoa (Theo-
have been reported as pests (Rochat el al. 1991, broma cacao, Sterculiaceae) plantations in many
2000a,b; Finnegan & Chambers 1993; Jaffe et al. countries, and its presence has been reported in
1993; Fettdther et al. 1995; Malosse et al. 1995; South America from Argentina to Venezuela, in

Florida Entomologist 88(2)

addition to some Caribbean islands such as Trin-
idad, Grenada, Martinique, Puerto Rico, and Ja-
maica among others (Entwistle 1972; Sanchez &
Capriles 1979). According to the literature on
S. breve, the egg phase lasts an average of 4.2
days, the larval phase 54.9 days, the pupal phase
10.9 days, and the adult phase 35 days for males
and 69 days for females (Mendes & Garcia 1984).
Adults feed on epidermal tissues of the main
stem and branches of young plants, producing a
characteristic gnawed area, which produces dam-
age to the floral clusters due to the fact that flow-
ers grow in tight clusters on the stem and
branches. Moreover, female adults lay eggs inside
slits cut into the bark with their mandibles. Usu-
ally, both sexes attack young trees aged from six
months to four years (Entwistle 1972; Sanchez &
Capriles 1979).
Larvae cause severe damage in stem and
branches. After eclosion, larvae bore into the bark
where they feed on the cambial tissues and the
bark itself. First, larvae make a round chamber,
progressively enlarging and elongating it until it
forms a tunnel or irregular spiral-like galleries.
This results in a ringed stem or ringed branch.
Depending on the age and location of the damage,
these events can kill the apical area. If the main
stem is attacked, they can quickly kill the entire
plant (Entwistle 1972; Sanchez & Capriles 1979).
Because the use of insecticides has progres-
sively decreased, it is very important to search for
alternative pest control methods that are safe for
the environment and highly efficient in the man-
agement of S. breve populations. In most cases,
ethological control has been applied successfully
(Nakamuta et al. 1997; Howse et al. 1998; Sey-
bold et al. 2000). This entails previous study of
olfactory behavior, as well as the identification
and evaluation of the chemical compounds in-
volved in insect communication (Hernandez et al.
1992; Howse et al. 1998). An olfactometer has
been used to study the olfactory behavior of many
coleopteran pests (Rochat et al. 1991; Jaffe et al.
1993; Cerda et al. 1996, 1999). Electroantennog-
raphy (EAG) is a technique that has been used for
the analysis of biologically active compounds
(Marion-Poll & Thiery 1996). In Coleoptera, EAG
and gas chromatography coupled with electroan-
tennographic detection (GC-EAD) have been
used to identify the chemical structure and stere-
ochemistry of the sex pheromone of another cer-
ambicidAnaglyptus subfaciatus Nakamuta et al.
(1994). Since nothing is known about the olfac-
tory behavior of S. breve towards odors derived
from its host plant or from odors emitted by
adult insects, the objectives of this study are
evaluation of olfactory behavior and electroan-
tennographic responses of S. breve when stimu-
lated by odors derived from cocoa plant tissues
and by volatiles produced by males, females, and
n-hexane extracts of insect body parts.


Collection of Insects and Cocoa Plant Branches

Insects and cocoa plant branches (CPB) were
collected at experimental cocoa plantations of the
Institute Nacional de Investigaciones Agricolas
(INIA), located in Campo Central and Padron, in
Municipio Acevedo, Miranda State, Venezuela.
Adult insects used in all experiments were col-
lected directly from the field and then placed indi-
vidually in plastic containers (7 cm high x 11 cm
0), transported to the laboratory in a thermal con-
tainer, and kept under controlled laboratory con-
ditions. Insects were fed with pieces of CPB for 12
to 16 h before olfactometric and electroantenno-
graphic experiments were performed. The pieces
of CPB were collected from 2- to 4-month old
EEM-003 and Ocumare 61 cocoa plant cultivars.

Olfactometric Experiments

Olfactometric bioassays were carried out at
Laboratory of Entomology INIA-Miranda, Cauca-
gua, in a two-choice olfactometer (Cerda et al
1996) with some modifications. Males and fe-
males were evaluated alternately with a mini-
mum of 25 individuals of each sex. Odor sources
used to stimulate the insects were adult females,
adult males, apical chopped pieces of cocoa
branches (2 to 4 months old) from EEM-003 selec-
tion, and combinations of the above totaling seven
different odor sources. These were: (1) males,
(2) females, (3) cocoa plant branches (CPB), (4)
females + males, (5) females + CPB, (6) males +
CPB, and (7) air. Each individual was placed in
a chamber located in the olfactometer for 3 min
before the evaluation of insect responses, in or-
der to get them used to the system. After that pe-
riod, the chamber door was opened and a small
fan with a 200-ml/min flow located at the back of
the olfactometer chamber was immediately
switched on in order to disperse odors inside the
system. Insect behavior was observed for 15 min
with (1) quantification of individuals reaching
either odor source, and (2) quantification of in-
active or undecided individuals recorded. Each
tested individual was considered as a replicate.
At the end of each experiment, all remaining cue
odors were removed by applying 70% v/v ethanol,
and then circulating a hot air stream with a
hair dryer through the device for approximately
10 min. All bioassays were performed from 8:00
to 12:00 and 13:00 to 15:00, and mean values of
temperature and relative humidity were re-
corded during each experiment. Experimental
data were analyzed by the binomial test
(Wiedenh6fer 1993) after comparison of the fre-
quencies of individuals selecting one odor source
vs. the other.

June 2005

Liendo et al: Olfactory and Electroantennographic Responses of S. breve

Insect Extracts

Extracts of body parts from 25 males and 25 fe-
males of S. breve were prepared by placing each
individual for 5 min in a refrigerator at 5C in or-
der to diminish insect activity. Then, each was dis-
sected in a Petri dish without any fluid into three
parts: head, prothorax, and pterothorax + abdo-
men. Each group of body parts was immediately
placed in a 5-ml clean glass vial, containing 3 ml
of n-hexane (HPLC grade), and extracted during a
48-h period. Then, the supernatant was removed
with a Pasteur pipette and placed in a 4-ml clean
glass vial. The extracts were concentrated to ap-
proximately 50 1l by a gentle nitrogen stream,
and kept at -5C until electroantennographic ex-
periments were conducted.

Electroantennographic Experiments

EAG experiments were performed in the Labo-
ratorio de Comportamiento, Universidad Simon
Bolivar. EAG responses of S. breve males and fe-
males were evaluated while being stimulated by
(1) pieces of CPB of Ocumare 61 clone (four pieces
5 cm long x 1 cm 0), (2) males (4 individuals), (3)
females (4 individuals), (4) 1 pl of n-hexane extract
of body parts (head, prothorax, pterothorax + abdo-
men), or (5) n-hexane. In experiments 1-3, the odor
source was introduced into a glass chamber where
the stimulus source was placed; a system to pro-
duce a wet air current, calibrated to 300 ml/min
flow, and a stimulus controller (Syntech model
CS-05, Hilversum, The Netherlands), was ad-
justed to a time pulse of 0.5 s and 500 ml/min flow.
In experiments 4 and 5 (n-hexane extracts and
control), stimuli were released from Pasteur pi-
pettes containing a piece of filter paper previously
impregnated with 1 pl of each extract or solvent
after the solvent had been allowed to evaporate.
The puff was delivered into the continuous air
stream, after placing the pipette tip into the hole
of the tube carrying the air stream. The antennal
responses were amplified and recorded with a Syn-
tech data acquisition controller and software.
Male or female antennae were excised and fixed
between silver-gold electrodes with two droplets of
an electrically conductive gel (Spectra 360 elec-
trode gel, Parker, Orange, NJ) applied to the elec-
trodes. Each stimulus was tested in 10 replicate
experiments in which the antenna received three
stimulus pulses at 3-min intervals. For quantifica-
tion of EAG amplitudes (mV) we only considered
the first pulse applied in each replicate. Antennal
responses (mV) from males and females were com-
pared by means of the Mann-Whitney U-Test.


The responses of S. breve in olfactometric bio-
assays when stimulated with various odor

sources are shown in Table 1. Only three sets of
experiments showed statistically significant dif-
ferences. For example, when CPB vs. male vola-
tiles were tested as odor sources, 55% of the
insects showed a statistically significant orienta-
tion towards CPB while 29% responded to male
odors, with 16% remaining undecided or inactive.
There was a clear preference of S. breve towards
CPB (67%, P < 0.05 Binomial test) versus a clean
air stimulus.
When results from orientation of both sexes to-
wards female vs. male odors are compared, we ob-
served that 67% of the evaluated females oriented
toward male odors (Binomial test, P < 0.05). Male
to male, male to female, and female to female ori-
entations were not statistically different. The
treatments with other odor source combinations
(Y + CPB, S + CPB, S + 2) did not show signifi-
cant differences (Binomial test,P > 0.05) (Table 1).
Table 2 shows the results of EAG experiments
when male and female antennae were stimulated
with female odors, male odors, CPB, and clean
wet air. Female antennae produced significantly
higher amplitudes (Mann-Whitney U-test, P <
0.05) than males, and gave stronger responses
when stimulated with male odors. Male and fe-
male electric potential average responses were
similar when exposed to CPB odors. Table 2
shows that male and female antennae stimulated
with clean wet air showed slight variations in
electrical potentials (mV). EAG responses of fe-
male antennae towards male + CPB odor pro-
duced a larger antennal depolarization that was
significant compared to male antennae (Mann-
Whitney U-test, P < 0.05) (Table 2).
Additional EAG results obtained from stimu-
lating male and female antennae with n-hexane
extracts of body parts are shown in Table 3. Fe-
male antennal responses were stronger than
males, especially when male prothorax extract
was applied as stimulus. Female antennal re-
sponses were higher (Mann-Whitney U-test, P <
0.05) when they were exposed to male extracts.
On the other hand, these results also show that
female head extract produced a stronger signal
from male antennae, even though they are both of
low intensity, and that prothorax extract gener-
ated a significant, slightly higher electroantenno-
graphic response from female antennae.


The results from olfatometric and EAG exper-
iments suggest that S. breve olfactory behavior is
highly influenced by odors emitted by cocoa
plants (kairomones) and also quite possibly by a
sex pheromone. The fact that during olfactomet-
ric tests S. breve female and male individuals
were attracted by CPB volatiles, and that addi-
tionally these plant odors produced a significant
electrical potential deflection (mV) in antennae of

Florida Entomologist 88(2)

June 2005



Odor sources S $ Total

Air vs. Air Air 9 6 15
Air 8 5 13
Not decided 5 4 9
Total 22 15 37
CPB vs. Air CPB 17 13 30*
Air 6 5 11
Not decided 2 1 3
Total 25 20 45
CPB vs. S CPB 17 15 32*
j 10 7 17
Not decided 3 6 9
Total 30 28 58
CPB vs. Y CPB 10 13 23
y 16 4 20
Not decided 6 11 17
Total 32 28 60
6 vs. $ 20 20* 40
$ 15 6 21
Not decided 6 4 10
Total 41 30 71
6 vs. 6 + 2 12 9 21
S + 14 9 23
Not decided 6 10 16
Total 32 28 60
S + 2 vs. $ + 2 7 12 19
y 9 12 21
Not decided 9 1 10
Total 25 25 50

*Statistically significant differences, Binomial test, P < 0.05.

both sexes of S. breve, suggests that volatile com-
pounds emitted by cocoa plant tissues are used by
the insects as cues to locate cocoa plants as their

hosts. After the insect arrives, the host could be a
site for feeding, mating, and oviposition, as has
been reported for other coleopteran pests of the


EAG responses from males (mV) EAG responses from females (mV)
Volatile compounds emitted by: Average SD Average SD

$ 0.7 0.1 1.2 +0.3*
S 0.2 0.1 7.5 +0.4*
CPB 1.4 0.1 1.6 0.3
Y + CPB 0.6 0.1 0.5 0.1
j + CPB 1.5 0.2 3.4 0.3
S+ 2 1.3 0.2 1.3 0.2
Air 0.1. 0.0 0.1 +0.0

*Statistically significant differences (Mann-Whitney U-Test, P < 0.05) when column values are compared.

Liendo et al: Olfactory and Electroantennographic Responses of S. breve


EAG response from males (mV) EAG response from females (mV)
n-hexane extracts of: Average SD Average SD

Head 6 0.4 0.1 6.0 0.8*
Prothorax 6 0.5 0.2 15.5 3.0*
Pterothorax + abdomen of 6 1.3 0.3 6.5 1.0*
Head Y 1.0 0.2 0.5 0.1*
Prothorax Y 1.2 0.3 2.1 0.3*
Pterothorax + abdomen of Y 0.6 0.1 0.8 0.5
n-hexane 0.2 0.3 0.1 0.0

*Statistically significant differences (Mann-Whitney U-test, P < 0.05) when the columns are compared.

Curculionidae and Cerambycidae families (Jaffe
et al. 1993; Hanks 1999). The presence of a char-
acteristic superficially gnawed area over the at-
tacked cocoa plant cortex is evidence that S. breve
use plants as a substrate for feeding.
In another Cerambycidae, Anaglyptus subfas-
ciatus, the existence of a male-produced sex pher-
omone was first suspected by the fact that
females were attracted to males in wind tunnel
bioassays (Nakamuta et al. 1994). This was later
confirmed by Leal et al. (1995), who identified the
pheromone components, but Nakamuta et al.
(1997) showed that a blend of host plant volatiles
and the male sex pheromone used as baits in yel-
low water traps were more attractive than sex
pheromone or host attractant alone. In some
other Cerambycidae species, the existence of sex
pheromones in males and also in females is con-
firmed, along with their host plant relationships
(Schrdreder et al.1994; Fettkdther et al. 1995;
Bento et al. 1993; Fukaya et al. 1996).
The presence of a male sex pheromone in S.
breve is suggested by olfactometric test results
which indicate that females are significantly
more attracted to males. In addition, female an-
tennae showed a very strong deflection in EAG
experiments only when stimulated by male odors.
Female antennae generated strong amplitude
signals with all n-hexane male extracts. Male pro-
thorax extract was the source that caused the
strongest response.
Sources for pheromone production in Coleop-
tera vary according to the kind of tissues involved.
In Carpophilus freeman (Nitidulidae) the phero-
mone gland has been located in the abdomen
(Dowd & Bartelt 1993). However, in other co-
leopterans the gland is located in the prothorax.
In the case of Anthonomus grandis Boheman
(Curculionidae), the aggregation pheromone is
produced in the fat bodies associated with the di-
gestive tract (Wiygul et al. 1982), and in Rhyn-
chophorus palmarum L. (Curculionidae), the
pheromone glands are located in the male protho-
rax (Sanchez et al. 1996). In the cerambycid

Hylotrupes bajulus the pheromone is produced in
a gland situated in the male prothorax (Fett-
kither et al. 1995). According to the strong evi-
dence provided by the olfactometric and
electroantennographic experiments of the present
study, it is highly probable that the S. breve pher-
omone production system is also located in the
prothorax. That each of the n-hexane body ex-
tracts elicited EAG responses implies that the S.
breve digestive system could be associated with
pheromone dispersion, as has been reported for
other coleopteran insects. Therefore, once male in-
sects arrive at a plant having been attracted by
volatile host compounds, it is very probable that
males start releasing a pheromone in order to en-
hance female searching behavior for mating.
In conclusion, the results of the present study
when combined with the findings reported in the
literature indicate that the chemical communica-
tion system and olfactory behavior of S. breve is
probably similar to that described for the ceram-
bycid A. subfasciatus. However, it is necessary to
continue research, currently in progress, in order
to identify the chemical compounds involved in
the S. breve communication system. This may en-
able their use as safe tools for the control of this
important neotropical pest.


The authors thank Professor Alexander Natera for
valuable help in English translation and Professor
Daniel Bailey for language corrections. We thank Jos6
Antonio Arias for help in Laboratorio de Compor-
tamiento, and INIA-Miranda and DID-USB GID-12 for
financial support.


FRIGHETTO. 1993. Male response to natural sex
pheromone of Migdolus fryanus Westwood (Coleop-
tera: Cerambycidae) females as affected by daily cli-
matic factors. J. Chem. Ecol. 19(10): 2347-2351.

Florida Entomologist 88(2)

1999. Olfactory attraction of the sugar cane weevil
(Coleoptera: Curculionidae) to host plant odors and
its aggregation pheromone. Florida Entomol. 82(1):
JAFFE. 1996. Atracci6n olfativa de Cosmopolites sor-
didus Germar (1824) (Coleoptera: Curculionidae) es-
timulado por volatiles originados en musiceas de
distintas edades y variedades gen6micas. Agron.
Trop. 46(4): 413-429.
COSSE, A., AND R. J. BARTELT. 2000. Male-produced ag-
gregation pheromone of Colopterus truncatus: struc-
ture, electrophysiological and behavioral activity. J.
Chem. Ecol. 26: 1735-1748.
DOWD, P. F., AND R. J. BARTELT. 1993. Aggregation pher-
omone glands of Carpophilus freeman (Coleoptera:
Nitidulidae) and gland distribution among other sap
beetles. Ann. Entomol. Soc. Am. 86: 464-469.
ECHANDI, E., AND C. E. FERNANDEZ. 1962. Relaci6n en-
tre el contenido de acido clorog6nico y la resistencia a
la llaga macana o cancer de los cafetos causada por
Ceratocystis fimbriata. Turrialba. 12: 2-5.
ENTWISTLE, P. 1972. Pest of cocoa. Ed. Longman, Lon-
don. 779 pp.
W. FRANCKE, AND U. NOLDT. 1995. The male phero-
mone of the old house borer Hylotrupes bajulus (L.)
(Coleoptera: Cerambycidae): Identification and re-
sponse. Experientia. 51: 270-277
FINNEGAN, D., AND J. CHAMBERS. 1993. Identification of
the sex pheromone of the Guernsey carpet beetle,
Anthrenus sarnicus Mroczkowski (Coleoptera: Der-
mestidae). J. Chem. Ecol. 19 (3): 971-983.
1996. Reproductive biology of the yellow-spotted lon-
gicorn beetle, Psacothea hilaris (Pascoe) (Coleoptera
Cerambycidae). III. Identification of contact sex
pheromone on female body surface. J. Chem. Ecol.
22(2): 259-270.
HANKS, L. M. 1999. Influence of the larval host plant on
reproductive strategies of cerambycid beetles. Annu.
Rev. Entomol. 44: 483-505.
SANCHEZ. 1992. Localizaci6n hospedera, actividad
diaria y optimizaci6n de la capture del picudo del co-
cotero Rhynchophorus palmarum L., mediante
trampas in6cuas. Agron. Trop. 42(3-4): 211-226.
HOWE, H. F., AND L. C. WESTLEY. 1997. Ecological Rela-
tionships of Plants and Animals. Oxford University
Press, New York. 278 pp.
HOWSE, P., I. STEVENS, AND O. JONES. 1998. Insect
Pheromones and Their Use in Pest Management.
Chapman and Hall, Great Britain. 321 pp.
HUGHES, J. C., AND T. SWAIN. 1960. Scopolin production
in potato tubers infected with Phytophthora in-
festans. Phytopathology 50: 391-398.
J. V. HERNANDEZ. 1993. Chemical ecology of the
palm weevil Rhynchophorus palmarum (L.) (Coleop-
tera: Curculionidae): Attraction to host plants and to
a male-produced aggregation pheromone. J. Chem.
Ecol. 19: 1703-1720.
MEINWALD. 1995. Structure, stereochemistry, and
thermal isomerization of the male sex pheromone of
the longhorn beetle Anaglyptus subfasciatus. Proc.
Nat. Acad. Sci. USA. 92: 1038-1042.

MORIN. 1995. Solid-phase microextraction, an alterna-
tive method for the study of airborne insect phero-
mones (Metamasius hemipterus, Coleoptera, Curculi-
onidae). J. High Resol. Chromatogr. 18: 669-670.
MARiON-POLL, F., AND D. THIERY. 1996. Dynamics of
EAG responses to host-plant volatiles delivered by a
gas chromatograph. Entomol. exp. Appl. 80: 120-123.
MENDES, A., AND J. GARCIA. 1984. Biologia do besouro
do cacao Steirastoma breve (Sulzer) (Coleoptera: Ce-
rambycidae). Rev. Theobroma. 14(1): 61-68
havioral and morphological evidence for to male-pro-
duced sex pheromone in the Cryptomeria twig borer,
Anaglyptus subfasciatus Pic (Coleoptera: Ceramby-
cidae). Japanese J. Entomol. 62: 371-376.
M. ONO, AND M. NAKANISHI. 1997. Increase of trap
catches by a combination of male sex pheromones
and floral attractant in longhorn beetle, Anaglyptus
subfasciatus. J. Chem. Ecol. 23(6): 1635-1640.
produced aggregation pheromone of the American
palm weevil, Rhynchophorus palmarum (L.) (Co-
leoptera, Curculionidae): Collection, identification,
electrophysiological activity and laboratory bioas-
say J. Chem. Ecol. 17(11): 2127-2141.
AND C. DESCOINS. 2000a. Identification of phero-
mone synergists in American palm weevil, Rhyn-
chophorus palmarum, and attraction of related
Dynamis borassi. J. Chem. Ecol. 26(1): 155-187.
DANA, T. KAKUL, AND J. MORiN. 2000b. Role of solid-
phase microextraction in the identification of highly
volatile pheromones of the two rhinoceros beetles Scap-
anes australis and Strategus aloeus (Coleoptera, Scara-
caeidae, Dynastinae). J. Chromatog. A 885:433-444.
asociados al cultivo del cacao en Venezuela. Mara-
cay: Fondo Nacional de Investigaciones Agropecuar-
ias. Boletin T6cnico 11. 56 pp.
MATERAN, AND F. SANCHEZ. 1996. Secretory mecha-
nisms for the male produced aggregation pheromone
of the palm weevil Rhynchophorus palmarum L. (Co-
leoptera: Curculionidae). J. Insect Physiol. 42(11-
12): 1113-1119.
AND W. FRANCKE. 1994. Synthesis of (3R)-3-hydroxy-
2-hexanone, (2R,3R)-2,3-hexanediol and (2S,3R)-
2,3-hexanediol, the male sex pheromone of Hylo-
trupes bajulus and Pyrrhidium sanguineum (Ceram-
bycidae). Liebigs Ann. Chem. 1211-1218.
Biosynthesis of coniferophagous bark beetle phero-
mones and conifer isoprenoids: Evolutionary perspec-
tive and synthesis. Can. Entomol. 132: 697-753.
WIEDENHOFER, H. 1993. Pruebas no param6tricas para
las ciencias agropecuarias. Muestras pequeias. Ma-
racay: Fondo Nacional de Investigaciones Agropec-
uarias. Serie A. 132 pp.
J. E. WRIGHT. 1982. Localization of pheromone in
male boll weevils Anthonomus grandis. Entomol.
exp. Appl. 31: 330-331.

June 2005

Pluke et al.: Coccinellids and D. citri in Puerto Rico


'Southwest Florida Research and Education Center, Institute of Food and Agricultural Science
University of Florida, 2686 St. Rd. 29 N., Immokalee, FL 34142

2H. Eslava-18, 8.-Burlada, Navarra, Spain-31600

3Department of Entomology, Kansas State University, Agricultural Research Center-Hays
1232 240th Ave., Hays, KS 67601


The first detections of Asian citrus psyllid, Diaphorina citri Kuwayama, in North America oc-
curred almost simultaneously in Florida and in the Caribbean (Guadeloupe). Damaging pop-
ulations on citrus have been reported in Florida but not in Puerto Rico where the psyllid was
first detected in 2001, having probably arrived with its parasitoid, Tamarixia radiata Water-
ston. In an effort to identify additional sources of mortality, the relative abundance of coccinel-
lid species was estimated on 180 citrus trees from early April to early July 2003. The most
abundant species were Coelophora inaequalis F. (38.8%) and Cycloneda sanguine limbifer L.
(31.3%), and the least abundant were Cladis nitidula F. (5.9%), Coleomegilla innonata Mul-
sant (4.8%), Chilocorus cacti L. (2.1%), Scymnus sp. (5.9%), Hippodamia convergens Guerin
(2.4%), and Cryptolaemus montrouzieri Mulsant (8.8%). These eight species were collected
from citrus production areas in Puerto Rico for a laboratory study of feeding behavior. All eight
consumed D. citri nymphs, with C. innonata consuming psyllids at a greater rate than C. cacti
or Scymnus sp. Choice tests showed that C. inaequalis and C. s. limbifer preferred brown citrus
aphid (BCA) Toxoptera citricida to D. citri as prey, whereas C. nitidula and C. cacti (both Chilo-
chorini) preferred D. citri. Our results suggest that coccinellid species could play an important
role as predators of the psyllid in Puerto Rico and contribute to its natural control.

Key Words: Asian citrus psyllid, Coccinellidae, predator, relative abundance, feeding behav-
ior, natural control


La primera identificaci6n en Norte Am6rica del psilido asidtico del citrico, Diaphorina citri
Kuwayama, se produjo casi simultaneamente en Florida y Guadeolupe. Se habian detectado
daios en citrico en Florida pero no en Puerto Rico, donde el psilido fue descubierto por pri-
mera vez en el 2001, llegando posiblemente junto con su parasitoide Tamarixia radiata Wa-
terston. En un intent por encontrar posibles fuentes adicionales de mortalidad para este
insecto plaga, se estim6 la abundancia relative de species de cocinlidos en 180 arboles de
citrico desde principios de Abril hasta principios de Julio del 2003. Las species mas abun-
dantes fueron Coelophora inaequalis F. (38.8%) y Cycloneda sanguine limbifer L. (31.3%),
y las menos abundantes fueron Cladis nitidula F. (5.9%), Coleomegilla innonata Mulsant
(4.8%), Chiloccorus cacti L. (2.1%), Scymnus sp. (5.9%), Hippodamia convergens Guerin
(2.4%) y Cryptolaemus montrouzieri Mulsant (8.8%). Estas ocho species fueron capturadas
de areas de producci6n citricola de Puerto Rico para el studio en laboratorio de su compor-
tamiento alimenticio. Las ocho species consumieron ninfas D. citri, pero C. innonata con-
sumieron un mayor numero de ninfas por hora comparado a C. cacti o Scymnus sp. Los
ensayos de elecci6n de presa mostraron que C. inaequalis y C. s. limbifer prefirieron como
presa al afido negro del citrico (BCA) Toxoptera citricida en lugar de D. citri, mientras que
C. coerulus y C. cacti amboss Chilocorini) prefirieron como presa a D. citri. Nuestros resulta-
dos sugieren que estas species de cocinlidos podrian jugar un papel important como
depredadores del psilido en Puerto Rico y contribuir a su control natural.

Translation provided by the authors.

The Asian citrus psyllid (ACP), Diaphorina until September 2001 (French et al. 2001). In the
citri Kuwayama, was first described from Taiwan Caribbean, the psyllid was first reported from
in 1907. It has been present in Brazil since the Guadeloupe (Etienne et al. 1998) and subse-
1940s (Lima 1942) but was not detected in Flor- quently detected in Puerto Rico in May, 2001
ida until 1998 (Halbert et al. 1998) and in Texas (Halbert & Nuiez 2004; P Stansly, unpublished

Florida Entomologist 88(2)

data). Diaphorina citri spread rapidly throughout
commercial citrus areas in Florida and heavy
populations were observed, especially in the
south. In contrast, infestations in Puerto Rico and
Texas are typically light (unpublished data and
V. French, pers. comm.).
Diaphorina citri feeds and reproduces on citrus
and additional hosts belonging to the subfamily
Aurantioidea of the family Rutacea including jas-
mine orange, Murraya paniculata L., which is
widely grown in Florida, Puerto Rico, and else-
where as an ornamental plant. Feeding damage on
citrus by large psyllid populations causes shoot dis-
tortion and abscission of the growing terminals.
The psyllid also can vector the bacterium Candida-
tus Liberibacter spp., causal agent of citrus green-
ing or 'huanglongbing' (Xu et al. 1988a). Citrus
greening is considered to be the most serious dis-
ease of citrus in Asia; it has recently been reported
from Brazil (Coletta-Filho et al. 2004), and poses a
serious threat to citrus in Florida and Puerto Rico.
Two parasitoids have been imported and re-
leased in North America for control ofD. citri, the
endoparasitic encyrtidDiaphorencyrtus aligarhen-
sis (Shafee, Alam & Agarwal) and the ectoparasitic
eulophid Tamarixia radiata (Waterson). Tama-
rixia radiata established itself in Florida and ap-
parently arrived spontaneously in Puerto Rico
(unpublished data) and Texas (French et al. 2001).
Predators in the family Coccinellidae (Coleop-
tera) have been shown to be responsible for a con-
siderable degree of natural control in Florida
(Michaud 2004). Michaud (2001) also presented
field data indicating that Olla v-nigrum Mulsant
and Harmonia axyridis (Pallas), coccinellid spe-
cies capable of feeding and reproducing on
D. citri, increased in relative abundance in psyl-
lid-infested citrus groves. The following study
was undertaken to provide baseline data on rela-
tive abundance of coccinellid species in Puerto
Rican citrus groves and to evaluate the ability of
the most abundant species to feed on D. citri.


Relative Abundance of Coccinellids on Citrus

Observations were carried out from early April
to early July 2003, in citrus groves at the Agricul-
tural Experimental Station in Adjuntas (18'10"N,
66'29"W. 457 m). The citrus groves of Adjuntas
are well established and cover a significant pro-
portion of both sides of the valley in which the ex-
perimental station is located. Because of the
altitude and cooler weather, new growth in citrus
is largely restricted to the main flush in the first
half of the year. This is the time that population
outbreaks of herbivorous insects such as aphids
occur, with the subsequent increase of lady beetle
numbers. The sampling period was designed to
encompass this period of greater lady beetle activ-

ity. Thirty citrus trees were randomly selected
and inspected between 8.00 a.m. and 12.00 a.m.
every other week for a total of 180 trees, based on
direct counts of all coccinellid adults, larvae, or
pupae encountered. The relative abundance of
each species was calculated as a proportion of the
total coccinellids counted.

Prey Acceptability and Prey Preference

Psyllid nymphs used in the experiments were
collected from a greenhouse colony established
from field-collected individuals and reared on
Murraya panniculata at the Rio Piedras Agricul-
tural Experiment Station. Nymphs of brown cit-
rus aphid (BCA) Toxoptera citricida (Kirkaldy)
were collected from infested citrus flushes at the
Adjuntas Agricultural Experimental Station and
used directly in the experiments. Coccinellid
adults of the various species were collected from
citrus groves at the Adjuntas Agricultural Exper-
imental Station in Puerto Rico during the period
from April to June, 2003 and maintained on Ephe-
stia kuehniella Keller (Lepidoptera: Pyralidae)
eggs at Adjuntas until needed. Chilocorus cacti,
Cladis nitidula, Cryptolaemus montrouzieri, and
Scymnus sp. could not be reared on E. kuehniella
eggs, and so were collected directly from the field
when needed for experiments.
The following eight species of cocinellids were
tested for prey acceptance and prey preference in
choice and no-choice tests: Cycloneda s. limbifer
L., Coelophora inaequalis F., Cladis nitidula F.,
Chilocorus cacti L., Coleomegilla innonata Mul-
sant, Cryptolaemus montrouzieri Mulsant, Hip-
podamia convergens Guerin, and Scymnus sp.
Individual adult coccinellids were starved for 24 h
and then confined in Petri dishes (5 cm diam.)
with one of the following prey configurations:10
psyllid nymphs only, 10 aphid nymphs only, or a
combination of 5 psyllid nymphs and 5 aphid
nymphs. The life stages of the two prey species
were chosen to be of similar size (usually third-
instars). Petri dishes were lined with white paper
to assist with the visual assessment of feeding
events. An experiment was judged to have been
completed in the choice tests once the adult coc-
cinellid had consumed all of one or the other prey
species in the Petri dish. In the no-choice tests,
the experiments were terminated after 7 h. Ten
replicates of the tests were carried out and each
adult coccinellid was tested only once. All experi-
ments were carried out under controlled condi-
tions at 25 1C and 75 10% RH.


For determining the differences in coccinellid
abundance, the square root transformation was
used and then Tukey's multiple comparison test
applied in Proc GLM in SAS (1999). For the feed-

June 2005

Pluke et al.: Coccinellids and D. citri in Puerto Rico

ing data comparisons in the combined host exper-
iments, normality was examined by the Shapiro-
Wilk test and plot functions of Proc Capability
(SAS 1999). The paired analysis was conducted
based on the signed rank test in the same SAS
procedure. For the consumption rate comparisons
in the single host experiments, normality was de-
termined by Proc Capability. Tukey's multiple
comparison test in Proc GLM was then used to
compare coccinellid consumption.


Relative Abundance of Coccinellids

A total of eight species of coccinellids were
found in citrus groves of the Adjuntas experimen-
tal research station during the course of the sur-
vey; C. inaequalis, C. s. limbifer, C. innonata, C. cacti,
C. nitidula, Scymnus sp., H. convergens, and
C. montrouzieri. The number of coccinellids ob-
served was relatively constant over the 3-month
study period (Fig. 1). Coelophora inaequalis and
Cycloneda s. limbifer were the most common spe-
cies found over the entire study period with the
remaining species found on a regular basis, but in
much lower numbers (Fig. 2).
Additionally, two species of Hymenoptera par-
asitized the coccinellids. Homalotylus sp. near ter-
minalis Say (Hymenoptera: Encyrtidae) emerged
from a C. s. limbifer pupa, while Oomyzus sp. near
scaposus Thomson (Hymenoptera: Eulophidae)
emerged from C. montrouzieri (larva) and C. s.
limbifer (pupa) (determinations by M. W. Gates,
pers. comm.).



Z 10 I

Prey Acceptability

All eight coccinellid species in the no-choice
tests fed on both host species with no rejection of
any offered prey type, although variation in the
quantity and rate of prey consumption was ob-
served. During the allotted interval there were
few differences in relative amounts of prey con-
sumed in no-choice tests, with the possible excep-
tion of C. nitidula, which ate twice as many
psyllids as aphids (Table 1). Hippodamia conver-
gens demonstrated the highest rate of aphid con-
sumption whereas C. cacti, C. nitidula, and
Scymnus sp. showed the lowest rate (Table 2).
The highest rate of psyllid consumption was ob-
served with C. innonata, although not signifi-
cantly different from the rates of all others except
C. cacti and Scymnus sp. (Table 3).

Prey Preference

Two of the coccinellid species examined, C. s.
limbifer and C. inaequalis, showed a strong pref-
erence for the brown citrus aphid, while C. nitid-
ula, C. cacti, C. innonata, and C. montrouzieri
showed significant preference for the Asian citrus
psyllid (Table 4). The remaining two species,
H. convergens and Scymnus sp. showed no prefer-
ence for either prey species.


Of the seven species of Coccinellidae listed by
Michaud (2004) as common in Florida citrus, only
the exotic Coelophora inaequalis also was found

April, 22 May. 6 May. 20 June. 3 June, 24 July,3

C. inaequalis C. sanguinea limbifer a C. montrouzieri
O C. nitidula m Scymnus sp. C. innonata
SH. convergens O C. cacti

Fig. 1. Abundance of 8 coccinellid species during the 3-month sampling period in 2003 at the Adjuntas Agricul-
tural Experimental Station in Puerto Rico.

Florida Entomologist 88(2)

a a


0 0.4-


> 0.2-

i 0.1 -

0 -

C. inaequalis I C. sanguinea limbifer 0 C. montrouzieri
0[ C. nitidula U Scymnus sp. 1 C. innonata
H. convergens 0 C. cacti

Fig. 2. Relative abundance of the coccinellid species caught at the Adjuntas Agricultural Experimental Station
in Puerto Rico.

in this survey. However, C. s. limbifer was consid-
ered by Gordon (1985) as a subspecies of C. san-
guinea, the most commonly encountered lady-
beetle in Florida citrus groves prior to the inva-
sion of Harmonia axyridis Pallas (Muma 1953;
Michaud 2001; Michaud 2002). Cycloneda limbi-
fer and C. inaequalis, both aphid-feeders, were
the two most abundant species encountered in
this study and also the most abundant reported
from Puerto Rican citrus groves by Michaud and
Browning (1999). The predominance of aphid
feeders followed by scale feeders are characteris-
tics shared by the coccinellid fauna in citrus
groves of Florida and Puerto Rico.


Host species

Species BCA % eaten ACP % eaten

H. convergens 94 7% 80 23%
C. innonata 84 12% 83 16%
C. s. limbifer 60 32% 60 13%
C. inaequalis 58 26% 54 21%
C. nitidula 30 13% 65 25%
C. montrouzieri 46 21% 47 21%
C. cacti 36 9% 47 8%
Scymnus sp. 16 5% 16 9%

Our results showed that all ladybeetle species,
with the exception of C. nitidula, consumed simi-
lar quantities of either host species presented
separately. Cladis nitidula consumed more psyl-
lids and also was the species showing greatest
preference for D. citri when offered a choice be-
tween prey. Thus, choice and no-choice experi-
ments were consistent in indicating C. nitidula's
preference for D. citri over T citricida. A prefer-
ence for T citricida over D. citri was strongly ex-
pressed by Cycloneda s. limbifer, and to a lesser
extent by Coelophora inaequalis, the two most
abundant ladybeetles in this study, although both
species consumed equal amounts of both prey
when given no choice. Michaud & Olsen (2004)
found that Cycloneda sanguine also fed on
D. citri as both larva and adult, but that larval de-
velopment time was almost doubled compared to
a diet of E. kuehniella and that female C. san-
guinea stopped ovipositing following transferal to
the D. citri diet. Further studies would be neces-
sary see whether aD. citri diet negatively impacts
the larval and reproductive performance of Cy-
cloneda s. limbifer as it does C. sanguine.
Prey suitability for Coccinellidae is a complex
issue that goes beyond the scope of prey accep-
tance and prey preference studied here, and gen-
eralizations such as "aphid feeders" and scale
feeders" are overly simplistic (Hodek 1996). Nev-
ertheless, the species demonstrating the greatest
degree of acceptance and preference for BCA,

b b b b


d d d

Coccinellid species

June 2005

Pluke et al.: Coccinellids and D. citri in Puerto Rico


Species Mean # of BCA consumed/h Standard deviation Tukey's comparison

H. convergens 1.54 0.64 a
C. innonata 1.09 0.42 a b
C. s. limbifer 0.93 0.43 b c
C. inaequalis 0.78 0.40 b c d
C. montrouzieri 0.53 0.39 b c d
C. cacti 0.37 0.27 c d
C. nitidula 0.30 0.26 d
Scymnus sp. 0.23 0.08 d

Note: different letters denote a significant difference at P = 0.05.

such as C. s. limbifer, C. inaequalis, and H. conver-
gens were not surprisingly those considered to be
aphid feeders. Those that preferred and/or most
readily accepted ACP or showed no clear tenden-
cies were either known to feed principally on
other prey such as Diaspididae (C. cacti), and
Pseudococcidae (C. montrouzieri), or had feeding
habits that were largely undocumented (C. nitid-
ula, Scymnus sp.).
Diaphorina citri was rarely encountered in Ad-
juntas during the course of this study, and ACP
populations have remained generally low in
Puerto Rico on both orange jasmine and citrus.
Coccinellids typically respond to dense prey popu-

nations whereas parasitoids with narrow host
ranges such as T radiata are expected to track
their host population at low densities. It is difficult
to ascertain, in retrospect, why ACP never
achieved high infestations in Puerto Rico as it has
Florida; however, preliminary evidence suggests
that T radiata may be responsible for holding the
psyllid in check in Puerto Rico (R. Pluke, unpub-
lished data). In any case, the coccinellid guild
present in Puerto Rican citrus, with its demon-
strated ability to consume ACP and its similar mix
of species in regard to feeding habits to the coc-
cinellid guild in Florida, would likely respond pos-
itively to any future increase in psyllid numbers.


Species Mean # ofACP consumed/h Standard deviation Tukey's comparison

C. innonata 1.21 0.18 a
H. convergens 1.14 0.33 a b
C. nitidula 1.14 0.74 a b
C. s. limbifer 0.88 0.55 a b
C. inaequalis 0.76 0.60 a b c
C. montrouzieri 0.67 0.30 a b c
C. cacti 0.50 0.32 b c
Scymnus sp. 0.16 0.15 c

Note: different letters denote a significant difference at P = 0.05.


Prey species

Species BCA % eaten ACP % eaten Preference P value

C. s. limbifer 64 4 Aphid** 0.004
C. inaequalis 78 28 Aphid** 0.016
C. nitidula 8 76 Psyllid** 0.002
C. cacti 30 80 Psyllid** 0.031
C. innonata 46 72 Psyllid** 0.031
C. montrouzieri 32 68 Psyllid* 0.055
H. convergens 60 58 N/A 0.711
Scymnus sp. 31 31 N/A 1.000

Note: *significant at P < 0.05, *P significant at P < 0.1.

Florida Entomologist 88(2)


We thank the personnel of the Rio Piedras and Adjun-
tas agricultural experiment stations for invaluable sup-
port and technical assistance, in particular, W. Gonzalez,
O. Cintr6n, E. M. P6rez, F. Ortiz, F. Sherwood, and Dr. K.
Nelo. We are grateful to M. Thomas of the Florida State
Collection of Arthropods for identification of ladybeetles
and M.W. Gates of USNM in Washington DC for identifi-
cation of parasitoids. Research funded under grant num-
ber 51099 provided by the USDA_CSREES Tropical/
Subtropical Agricultural Research Program. Florida Ex-
periment Station Number R-10588.


ANONYMOUS. 2004. 2002 census on agriculture. USDA
National Agricultural Statistics Service. http://
www.nass.usda.gov/census/. Viewed 05/28/2004.
BOVE, J. M., AND M. GARNIER 1984. Greening and
psylla vectors of the disease in the Arabian Penin-
sula, pp. 109-114 In S. M. Garnsey, L. W. Timmer,
and J. A. Dodds [eds.], Proceedings 9th Conference of
International Organization of Citrus Virology
(IOCV), University of California, Riverside.
CHEN, C. C., AND Y. I. CHU. 1996. Biological control of
citrus psyllid, Diaphorina citri, in Taiwan, pp. 93-
105 In Biological Pest Control in Systems of Inte-
grated Pest Management. Proceedings of the Inter-
national Symposium on "The Use of Biological
Control Agents under Integrated Pest Manage-
ment". Food and Fertilizer Technology Center for the
Asian and Pacific Region, Taipei, Taiwan.
A. M. Do AMARAL, AND G. W. MULLER. 2004. First
Report of the Causal Agent of Huanglongbing ("Can-
didatus Liberibacter asiaticus") in Brazil. Plant Dis-
ease 88:1382
First Record of the Asian Citrus Psylla, Diaphorina
citri Kuwayama (Homoptera:Psyllidae), in Texas.
Subtropical Plant Science 53: 14-15.
HALBERT, S., L. BROWN, AND W. DIXON. 1998. Asian cit-
rus psyllid update: Fla. Dep. Agric. Consumer. Serv.
Div. Plant Industry 18 Nov.
HALBERT, S. W., AND C. A. NUNEZ. 2004. Distribution of
the Asian citrus psyllid Diaphorina citri Kuwayama
(Rhynchota: Psyllidae) in the Caribbean Basin. Fla.
Entomol. 87: 401-402.
HODEK, I. 1996. Food relationships, pp. 143-234 In
I. Hodek and A. Honek [eds.], Ecology of Coccinel-

lidae. Kluwer Academic Publishers, Dordrect, Bos-
ton and London.
LIMA, A. M. DA C. 1942. Insectos do Brazil: Homopteros
8, Rio de Janeiro (Imprenso Nacional), p. 101.
MARTINEZ, A. L., AND J. M. WALLACE. 1967. Citrus leaf-
mottle-yellows disease in the Philippines and trans-
mission of the causal virus by a psyllid: Diaphorina
citri. Plant Dis. Rep. 51: 692-695.
MICHAUD, J. P., AND H. W. BROWNING. 1999. Seasonal
abundance of the brown citrus aphid, Toxoptera
citricida (Homoptera: Aphididae) and its natural
enemies in Puerto Rico. Florida Entomol. 82: 424-
MICHAUD, J. P. 2001. Numerical response of Olla v-ni-
grum (Coleoptera: Coccinellidae) to infestations of
Asian citrus psyllid (Hemiptera: Psyllidae), in Flor-
ida. Florida Entomol: 84: 608-612.
MICHAUD, J. P. 2002. Biological control of Asian citrus
psyllid, Diaphorina citri (Homoptera: Psyllidae), in
Florida: a preliminary report: Entomol. News 113:
MICHAUD, J. P. 2002. Invasion of the Florida Citrus Eco-
system by Harmonia axyridis (Coleoptera: Coccinel-
lidae) and Asymmetric Competition with a Native
Species, Cycloneda sanguine. Environ. Entomol.
MICHAUD, J. P. 2004. Natural mortality of Asian citrus
psyllid, Diaphorina citri (Homoptera: Psyllidae) in
central Florida. Biological Control 29: 260-269.
MICHAUD, J. P., AND L. E. OLSEN. 2004. Suitability of
Asian citrus psyllid, Diaphorina citri, as prey for
ladybeetles (Coleoptera: Coccinellidae). BioControl
49(4): 417-431.
MOLL, J. N., AND S. P. VAN VUUREN. 1977. Greening dis-
ease in Africa. Proc. Int. Soc. Citricult. 3: 903-912.
MUMA, M. H. 1953. Ladybeetle predators of citrus
aphids. Citrus Mag. 15(8): 32-33.
SAS INSTITUTE, INC. 1999. SAS/STAT user's guide. Ver-
sion 8.01. SAS Institute, Inc., Cary, NC
TSAI J. H., AND Y. H. LIU. 2000. Biology of Diaphorina
citri (Homoptera: Psyllidae) on four host pants. J.
Econ. Entomol. 93: 19219-1725.
TSAI, J. H., Z. Y. CHEN, C. Y. SHEN, AND K. X. JIN. 1988.
Mycoplasmas and fastidious vascular prokaryotes
associated with tree diseases in China, pp. 69-240,
In C. Hiruki [ed.], Tree Mycoplasmas and Myco-
plasma Disease, The University of Alberta Press,
Edmonton, AB, Can.
TSAI, J., H., J.-J., WANG, AND Y.-H. LIU. 2002. Seasonal
abundance of the Asian citrus psyllid, Diaphorina
citri (Homoptera: Psyllidae) in southern Florida.
Florida Entomol. 85: 446-451.

June 2005

Goolsby et al.: Interaction of Climate and Floracarus


'United States Department of Agriculture, Agricultural Research Service
Australian Biological Control Laboratory, Indooroopilly, Queensland, Australia
Present address: USDA-ARS, 2413 Hwy 83, Weslaco, TX 78501, USA

2Department of Zoology, Madras Christian College, Tambaram, Chennai, India 600 059

3ATER University Paul Sabatier, IRD, Noum6a Cedex, Nouvelle Cal6donie

4CSIRO Entomology, Long Pocket Laboratories, 120 Meiers Rd., Indooroopilly, QLD, Australia 4068

5United States Department of Agriculture, Agricultural Research Service, Invasive Plant Research Laboratory
Ft. Lauderdale, FL 33314-7700, USA


The Old World climbing fern, Lygodium microphyllum, is an invasive weed in the Florida
Everglades and the leaf roll galling mite, Floracarus perrepae, is a proposed biological con-
trol agent. Field studies were conducted for one to two years at sites in its native range in
Australia, New Caledonia, and India to evaluate the effect of climate on F perrepae. Monthly
counts of the proportion of L. microphyllum subpinnae (leaflets) with leaf roll galls were
used to measure the incidence of damage caused by F perrepae. Between sites the most sig-
nificant weather variable was rainfall 14 to 28 days prior to sampling, with higher levels
having a depressive effect on the incidence of leaf rolls. Within sites the mean maximum
temperature was the only significant weather variable, showing a decrease in the incidence
of leaf rolls above 27 C, and it was predicted that no leaf rolls would form above 35 C. The
weather parameters in Homestead, Florida for 2002 were within the range of those evalu-
ated in the eight native range field sites. Thus, we do not predict that climate will prevent
the establishment of this biological control agent for L. microphyllum in southern Florida.

Key Words: climate, biological control of weeds, ferns, Florida Everglades, New Caledonia,


El helecho trepador del viejo mundo, Lygodium microphyllum, es una maleza invasora en los
Everglades de Florida y el acaro que causa la agalla de enrollamiento de las hojas, Flora-
carus perrepae, es el agent de control biol6gico propuesto. Estudios de campo fueron lleva-
dos a cabo de uno a dos anos en sitios localizados en su area native en Australia, Nuevo
Caledonia, y India para evaluar el efecto del clima sobre F perrepae. Se usaron los conteos
mensuales de la proporci6n de subpinnae (hojuelas) de L. microphyllum con agallas de hojas
enrolladas para medir la incidencia del daho causado por F perrepae. Entre los sitios, la va-
riable mas significativa del clima fue la lluvia 14 a 28 dias antes del muestreo, con los niveles
mas altos causando un efecto negative sobre la incidencia de las hojas enrolladas. En el
mismo sitio el promedio de la temperature maxima fue la unica variable del clima significa-
tiva, demostrando una baja de la incidencia del enrollamiento de las hojas a las temperatu-
ras mayor de 27 C, y se predijo que ninguna hoja enrollada se formard a temperatures
mayores de 35 C. Los parametros del clima en Homestead, Florida para el aho 2002 estaban
en el rango de los parametros evaluados en el campo de los ocho sitios estudiados en el area
native del acaro. Por eso, nosotros no estimamos que el clima pueda prevenir el estableci-
miento de este agent de control biol6gico para L. microphyllum en el sur de Florida.

Lygodium microphyllum (Cav.) R. Br. (Lygodi- tats in southern Florida with the potential to
aceae, Pteridophyta), Old World climbing fern, is spread into Central and South America (Pember-
native to the wet tropics and subtropics of Africa, ton & Ferriter 1998; Goolsby 2004). A biological
Australasia, Asia, and Oceania (Pemberton 1998), control program was initiated in 1997 and surveys
and is an aggressive invasive weed of moist habi- for potential agents were conducted in Australia

Florida Entomologist 88(2)

and South Asia (Goolsby et al. 2003). The erio-
phyid mite, Floracarus perrepae Knihinicki and
Boczek, was the most widely distributed with sev-
eral geographically specific genotypes identified
(Goolsby et al. 2003; Goolsby et al. 2004a).
Throughout its native distribution in Australia
and Asia, F perrepae causes significant damage to
L. microphyllum. Feeding by the adults and imma-
tures causes formation of leaf roll galls, leading to
necrosis and premature defoliation of L. micro-
phyllum pinnae, and the gradual debilitation of
the plant (Goolsby et al. 2003; Freeman et al. 2005;
Ozman & Goolsby 2005). Based on its narrow host-
range (Goolsby et al. in press), and significant im-
pact on L. microphyllum (Goolsby et al. 2004b),
F perrepae was prioritized for evaluation as a bio-
logical control agent (Goolsby & Pemberton 2005).
Ozman & Goolsby (2005) documented the biol-
ogy and seasonal phenology of F perrepae in
Southeast Queensland, Australia, and found that
it was active year round, with populations peak-
ing when temperatures were cool and soil mois-
ture levels were highest. In these studies, the
proportion of mite infested, or incidence of rolled
subpinnae, was measured monthly at four loca-
tions over a two-year period. The incidence of leaf
rolls was used as an indicator of the mite's impact
on the fern. Because the effect of weather was the
most significant factor in the phenology of the
mite in Australia, we replicated this study in In-
dia and New Caledonia to determine if the geno-
types from these locations were affected in the
same manner. The Indian and New Caledonian
F perrepae genotypes were known to have unique
biological differences, in that they were co-
adapted to the local genotype of the fern (Goolsby
et al. unpublished data), but little was known
about their possible adaptations to climate.
Therefore, we used the incidence of the proportion
of mite induced leaf rolls (or plant damage) to
evaluate the seasonal phenology of F perrepae in
other parts of its native range that were climati-
cally different to Southeast Queensland.


Study Sites

The initial baseline studies were conducted in
Southeast Queensland (QLD) from February 2001
to March 2003 on Bribie Island; at Gallagher's
Point (2701.17'S, 15306.53'E) and McMahon Rd.
(2704.33'S, 153010.55'E) and at Logan; Carbrook
Creek (2741.30'S, 15316.23'E) and Lagoon Rd.
(2740.01'S, 15316.03'E). Samples were collected
monthly from February 2001 to February 2003.
These sites are seasonally inundated with fresh
water and L. microphyllum grows as a climbing
vine on paper bark trees, Melaleuca quinquen-
ervia (Cav.) St. Blake. The site at Carbrook Creek
is different in that L. microphyllum grows under

a closed canopy ofM. quinquenervia in deep shade
as compared to the other sites with open canopies.
The climate in southeast Queensland is subtropi-
cal. Rain falls mainly in the summer months,
which are hot and humid. Winters are cool and
dry with an average yearly temperature of 20.6C
and average rainfall of 1393 and 1256 mm at Bri-
bie Island and Logan, respectively.
Sites in India were located in southern Tamil
Nadu at Thomaiyarpuram, Nagercoil (819.00'N,
77025.70'E) and in Kerala near Ithikkara River,
Quilon (8051.95'N, 76041.79'E). Samples were col-
lected monthly from July 2002 to August 2003. At
the Nagercoil site L. microphyllum was growing
under a canopy of coconut palms in deep shade.
The Quilon site was exposed to full sun, with the
fern growing in a ditch up a roadside embank-
ment. The climate in southern India is monsoonal
with heavy rainfall from April to August and hot,
humid summers. The average yearly temperature
and rainfall for Nagercoil and Quilon are 30.0 and
27.50C, and 905 and 2932 mm, respectively Sites
in New Caledonia were located in Province Sud
near Noumea at la Coulee (2214.09'S,
166034.75'E) and Yate (2206.36'S, 166056.08'E).
Samples were collected monthly from May 2002
to November 2003. At both sites L. microphyllum
grows in partial shade with M. quinquenervia.
The climate in New Caledonia is subtropical with
rain evenly spaced throughout the year with an
average yearly rainfall of 1106 mm. The average
yearly temperature is 23.40C.
We obtained the weather data from the closest
available weather station for each location, daily
maximum and minimum temperature, RH at 9
a.m. and 3 p.m. and rainfall. From these the daily
soil moisture index (between 0 and 1 at soil satu-
ration) was calculated based on a simple model
developed by Fitzpatrick and Nix (1969). Yearly
rainfall averages for India and New Caledonia
were obtained from CLIMEX 2 (Sutherst et al.
2004) and the Queensland Department of Natu-
ral Resources for sites in Australia (Queensland
2004). For comparison of the native range to the
proposed area of introduction for F perrepae, we
used weather data from 2002 for the Florida Au-
tomated Weather Network (Florida 2004), for
Homestead, Florida, USA, which is located near
Everglades National Park.

Sampling Methods

Lygodium microphyllum grows as a twining
vine, and each shoot or rachis is a true leaf, con-
sisting of pinnae (leaflets) and subpinnae (sub-
leaflets). Vines were typically found climbing up
the trunks of trees, reaching up to 10 m with
many yellowish-green, fertile or sterile pinnae
branching off the main stem. Each pinna consists
of 6-12 paired subpinnae, which are the smallest
leaf unit. Fertile subpinnae are fringed with lobes

June 2005

Goolsby et al.: Interaction of Climate and Floracarus

of sporangia, with sterile subpinnae having a
smooth outer margin. Floracarus perrepae feed
on and cause leaf roll galls on the fertile subpin-
nae, but they did not prefer this leaf form. There-
fore, the field phenological studies were based on
counts of sterile subpinnae. Each month 50 newly
expanded sterile pinnae were collected at each
site along a transect and returned to the labora-
tory for counting. At each site the numbers of in-
fested and uninfested subpinnae were counted for
each pinna. This count provided a measure of the
incidence of infested subpinnae (leaf rolls), or to-
tal mite damage, at each location.

Statistical Analyses

Based on previous field research, it was deter-
mined that leaf rolls were the result ofF. perrepae
feeding activity from between 28 and 14 days be-
fore field collection (Goolsby et al. 2004; Ozman &
Goolsby, in press). Therefore mean values of
weather variables for this two-week period were
used for the analyses. An analysis of variance was
performed on the proportional incidence of leaf
rolls to determine which of the weather variables
accounted for the most variation between and
within sites. Sites were assigned a value of 1 if the
L. microphyllum grew under a canopy of full
shade, with a value of 0 for sites with partial
shade or full sun.


The incidence of leaf rolls was observed
monthly for periods between one and two years at
eight sites, two in India, two in New Caledonia,
and four in Australia. The mean value of the pro-
portional incidence of leaf rolls at each site is
shown in the Table 1, together with mean

weather variables daily maximum and minimum
temperature, rainfall, soil moisture, and shade.
An analysis of variance was performed on the
proportion of leaf rolls to determine which of the
abiotic variables accounted for the most variation
between and within sites. The rainfall of the pre-
vious two-week period, averaged over each site,
fitted as a quadratic equation explained 90% of
the variation between the 8 sites (P < 0.001, lin-
ear regression coefficient (bl) = -0.00323, SE
0.00031 and quadratic regression coefficient (b2)
= 0.0000485, SE = 0.000009, F2, = 22.71) (Fig. 1).
Shade explained a further 5% of the between site
variation but only at (P < 0.07) with leaf rolls from
shaded locations (0.048 SE 0.019, F14 = 3.22)
higher than unshaded. The shape of the quadratic
for rainfall indicated that rainfall decreased the
incidence of leaf rolls steadily as it increased.
However, after 72 mm of rainfall for a two-week
period, the predicted proportional incidence of
leaf rolls remained fairly constant as rainfall in-
creased. Temperature, relative humidity, and soil
moisture were not significant in the analysis be-
tween sites.
Within sites (after adjusting for mean site dif-
ferences) the mean maximum temperature fitted
as a quadratic explained the most variation in the
incidence of leaf rolls (P < 0.001, bl = 0.1598, SE
0.0532, b2 = -0.00319 SE = 0.00099, F2,13 = 7.95)
(Fig. 2). From the shape of the fitted curve, maxi-
mum temperature appeared to decrease the inci-
dence of leaf rolls as the temperature rose above
27C, but below the prediction was fairly con-
stant. At mean maximum of above 35C no curls
were predicted. There was no interaction between
the variables and sites; hence, all sites had the
same response to the weather variables. Relative
humidity, rainfall, and soil moisture were not sig-
nificant within sites.


No. Mean proportion Meanb max. Mean min. Mean" Soild
Site obs. leaf rolls + SE temp. C temp. C rain mm. moisture Shade"

Quilon, India 12 0.222 0.051 ab 32.4 23.5 60.1 0.63 0
Nagercoil, India 13 0.346 0.065 bcd 32.7 23.8 37.1 0.59 1
La Coulee, New Caledonia 20 0.157 + 0.036 a 28.1 17.3 70.0 0.74 0
Yate, New Caledonia 20 0.170 + 0.028 a 26.9 19.6 117.0 0.85 0
Carbrook Creek, Australia 24 0.358 0.022 cd 26.0 15.4 34.0 0.61 1
Lagoon Rd., Australia 24 0.289 0.035 bc 26.0 15.4 33.5 0.66 0
Gallagher's Pt., Australia 24 0.329 0.029 bcd 25.7 16.3 30.0 0.65 0
McMahon Rd., Australia 12 0.432 0.053 d 25.5 16.0 18.5 0.65 0
Homestead, Florida 12 n/a 29.1 18.4 89.6 0.80 n/a

aMeans in the column followed by the same letter are not significantly different (P < 0.05) from transformed data analysis.
bMean of the means for daily maximum/minimum temperatures 28-14 days prior to each sampling date.
'Mean of the sum of rainfall for 28-14 days prior to each sampling date.
dA value of 1 indicates soil saturation.
Zero indicates full sun, with 1 being partial or full shade at the site.

Florida Entomologist 88(2)



o 0.3

o 0.2
o 0.1

20 30 40 50 60 70 80 90 100

Rainfall (mm)

Fig. 1. Estimated effect of rainfall on annual proportion of uninfested subpinnae and mite induced leaf rolls (be-
tween sites). A higher proportion indicates a higher level of plant damage.

The weather data from Homestead, FL was
calculated for 12 hypothetical monthly sample
dates in 2002. The mean rainfall of the 12 two-
week totals was 90 mm with the mean maximum
temperatures of the two-week sampling periods
in a range between 23.1 and 32.4C (Table 1).


Floracarus perrepae is widely distributed in
Australia and Asia across a range of tropical and
subtropical climates (Goolsby et al. 2003). Within
this distribution are several location specific gen-
otypes that are adapted to their corresponding
genotypes of their host, L. microphyllum (Goolsby
et al. 2004a; Goolsby et al., unpublished data).







22 24 26 28 30

From these studies, it appears that the specific
genotypes represented in southern Australia,
New Caledonia, and India responded similarly to
the weather variables in their environments,
since there was no interaction with sites in the
within site analysis.
The most significant weather variable between
sites was rainfall. Jeppson et al. (1975) reported
similar effects from high rainfall during the rainy
season of Asiatic monsoon climates, which caused
major reductions in mite populations such as the
tea spider mite, Oligonychus coffeae (Nietner). In
our studies, high amounts of rainfall 28-14 days
months prior to the sampling date had a negative
impact on the incidence of leaf rolls, which is an
indicator of the local population density. Persis-

32 34 36

Temperature C

Fig. 2. Estimated effect of mean maximum temperature on proportion uninfested subpinnae and mite induced
leaf rolls (within sites). A higher proportion indicates a higher level of plant damage.

June 2005

Goolsby et al.: Interaction of Climate and Floracarus

tent and heavy rainfall during this time period
most likely dislodged the dispersing F perrepae
as they attempted to settle and induce leaf rolls.
This may explain why the sites in Quilon, India
and New Caledonia had the lowest mean propor-
tional incidence of leaf rolls. Quilon receives a
double rainy period during the monsoon, and
heavy persistent rain is common. In addition, the
Quilon site has high rainfall, is in the open, and
exposed to the direct effect of rain. In contrast,
the site in Nagercoil received less rain, the
L. microphyllum stand was under a sheltering
canopy of palms, and the incidence of leaf rolls
galls was higher, within the range of values for
the Australian sites. In New Caledonia, high rain-
fall during 2002-03, including that associated
with cyclone Erica, (March 2003) appears to have
been responsible for the lower incidence of leaf
rolls. Additionally, both sites in New Caledonia
are partially exposed with L. microphyllum grow-
ing under a broken canopy of M. quinquenervia,
thus allowing for the full impact of rainfall.
The higher average proportional incidence of
leaf rolls at the Australian field sites also may be
due to the drought conditions, which have affected
eastern Australia since 2001 (Queensland 2004).
Floracarus perrepae at these field sites have been
exposed to fewer episodes of high rainfall, and
mortality of dispersing females may have been
lower. Thus, the higher proportional incidence of
leaf rolls. The coconut mite, Aceria guerreronis
Keifer (Acari: Eriophyidae), appears to be affected
similarly across its known distribution where it
reaches higher densities in areas with drier cli-
mates (Moore & Howard 1996). The study by
Ozman & Goolsby (2005) found populations of F
perrepae peaked when temperatures were cool
and soil moisture levels were highest, which ap-
pears to contradict the findings of this study. How-
ever, the results from the continental comparisons
of climatic effects further clarify the differences
between the effects of soil moisture and rainfall
on F perrepae populations. Our results suggest
that F perrepae performs best when soil moisture
is near saturation (promoting growth of L. micro-
phyllum) and there are few periods of high inten-
sity rainfall, which interfere with dispersal.
The mean rainfall for Homestead, Florida in
2002 was higher than that for any station mea-
sured in Australia. One could predict that popula-
tions of F perrepae in the latter localities, would
be similar to those in New Caledonia, where the
mite caused considerable damage to L. micro-
phyllum. The rainfall effect appears to have a
limit as shown in Figure 1, in which the model
predicts that effect will remain constant above 72
mm for a two-week period. This prediction is cor-
roborated by the data from New Caledonia and
India, which showed that high rainfall reduced
the incidence of leaf rolls but the populations re-
mained stable.

High temperatures also had a significant nega-
tive impact on F perrepae populations within sites.
The model predicts that the proportional incidence
of leaf rolls will decrease as the mean maximum
temperature rises above 27C with no leaf roll for-
mation above 35C. None of the sites in the native
range experienced mean maximum temperatures
for a two-week period above 35C. Similarly, in
Homestead, Florida during 2002 the highest
means for the summer months ranged between 31
and 32C, respectively. High temperatures are im-
portant to consider for rearing and release of
F perrepae. Summer greenhouse temperatures can
often reach mean maximum temperatures above
35C. Optimum rearing environments should be
maintained at a mean maximum temperature of
27C. Field colonization of F perrepae in Florida
would be the most difficult in the summer months
when temperatures and rainfall are the highest.
In conclusion, the genotypes of F perrepae
evaluated in this study from its native range in
Australia, New Caledonia, and India were all
negatively impacted by high rainfall and temper-
atures during leaf roll formation. Conversely, pop-
ulations of F perrepae reach their highest levels
when temperatures are cool and episodes of high
intensity rainfall are few. The climate of southern
Florida falls within the parameters experienced
in the native range of F perrepae. Therefore, we
do not predict that climate will prevent the estab-
lishment of this biological control agent for
L. microphyllum.


The authors thank USDA-ARS, Office of Interna-
tional Research Programs, the Florida Department of
Environmental Protection, and the South Florida Water
Management District for financial support; Dr. Dave
Walter (University of Queensland), Drs. Richard
Greene and Ernest Delfosse (USDA-ARS, National Pro-
gram Staff) for support and encouragement; Ryan Zon-
neveld (CSIRO Entomology) for collecting and
processing the field samples, and John Lydon (USDA-
ARS) and Marc Coombs (CSIRO Entomology) for re-
viewing the manuscript.


FLORIDA. 2004. Florida Automated Weather Network.
University of Florida, IFAS. http://fawn.ifas.ufl.edu/
FITZPATRICK, E. A., AND H. A. NIX. 1969. A model for
simulating soil water regime in alternating fallow-
crop systems. Agri. Meteorol. 6: 303-319.
NELSON. 2005. An ultrastructural study of the rela-
tionship between the mite Floracarus perrepae Kni-
hinicki & Boczek (Acariformes: Eriophyidae) and the
fern Lygodium microphyllum (Cav.) R. Br. (Lygodi-
aceae). Aust. J. Entomol. 44: 57-61.
2003. Exploratory Surveys in Australia and Asia for

Florida Entomologist 88(2)

Natural Enemies of Old World Climbing Fern, Lygo-
dium microphyllum: Lygodiaceae. Biol. Control 28:
ZONNEVELD, AND A. D. WRIGHT. 2004a. Pre-release
evaluation and host range testing of Floracarus per-
repae (Eriophyidae) genotypes for biological control
of Old World climbing fern, pp. 113-116 In J. M.
Cullen, D. T. Briese, D. J. Kriticos, W. M. Lonsdale,
L. Morin, and J. K. Scott [eds.], Proceedings of the XI
International Symposium on Biological Control of
Weeds CSIRO Entomology, Canberra, Australia.
GOOLSBY, J. A., AND R. W. PEMBERTON. 2004. Proposed
field release of the Australia Floracarus perrepae
Knihinicki & Boczek (Acariformes: Eriophyidae), a
leaf roll gallery of the Old World climbing fern, Lygo-
dium microphyllum (Cav.) R. Br. (Pteridophyta:
Lygodiaceae) in Florida. On file at the U.S. Depart-
ment of Agriculture, Agricultural Research Service,
Biological Control Documentation Center, National
Agricultural Library, Beltsville, Maryland, USA (pe-
tition to release a biological control agent).
Pre-release assessment of impact on biomass pro-
duction of an invasive weed, Lygodium microphyl-
lum (Lygodiaceae: Pteridophyta), by a potential
biological control agent, Floracarus perrepae (Acari:
Eriophyidae). Environ. Entomol. 33: 997-1002
GOOLSBY, J. A. 2004. Potential distribution of the inva-
sive Old World climbing fern, Lygodium microphyl-
lum in North and South America. Nat. Areas J. 24:
W. PEMBERTON. Host-range and cold temperature

tolerance of Floracarus perrepae (Acariformes: Erio-
phyidae) a potential biological control agent of Lygo-
dium microphyllum (Pteridophyta: Lygodiaceae).
Aust. J. Entomol. (In Press).
Population ecology; mite responses to weather vari-
ations, pp. 28-29 In Mites Injurious to Economic
Plants. University of California Press, Berkeley.
MOORE, D., AND F. W. HOWARD. 1996. Ecology of coco-
nut mites, pp. 561-570 In E. E. Lindquist, M. W. Sa-
belis, and J. Bruin [eds.], Eriophyoid Mites-Their
Biology, Natural Enemies and Control, Elsevier Sci-
ence, Amsterdam.
OZMAN S. K., AND J. A. GOOLSBY. 2005. Biology and phe-
nology of Floracarus perrepae, on its native host in
Australia, Old World climbing fern, Lygodium mi-
crophyllum. Exp. Appl. Acarol. 35: 197-213.
PEMBERTON, R. W. 1998. The potential of biological con-
trol to manage Old World climbing fern (Lygodium
microphyllum), an invasive weed in Florida. Am.
Fern J. 88: 176-182.
PEMBERTON, R. W. 2003. Northward extension of the in-
vasive weed Lygodium microphyllum in Florida.
Sida 20: 1759-1761.
World climbing fern (Lygodium microphyllum), a
dangerous invasive weed in Florida. Am. Fern J. 88:
QUEENSLAND. 2004. Bureau of Meteorology, Queen-
sland rainfall maps. http://www.bom.gov.au/climate/
A. BOURNE. 2004. CLIMEX v2. Hearne Scientific
Software Pty. Ltd., Melbourne, Australia.

June 2005

Boesi et al.: Trap-nesting Ancistrocerus sikhimensis


'Dipartimento di Biologia Evolutiva, Universita degli Studi di Siena-Via A. Moro, 4, 35100, Siena, Italy

2Dipartimento di Biologia, Sezione di Zoologia e Citologia, Universita degli Studi di Milano-Via Celoria
26, 20133, Milan, Italy

3Unidad de Zoologia, Facultad de Biologia, Universidad de Salamanca-37071, Salamanca, Spain


The contents of 21 trap-nests located in Sagarmatha National Park, Nepal, in 2002 and 2003
revealed interesting aspects of the biology of Ancistrocerus sikhimensis Bingham (Hy-
menoptera: Eumenidae). The nests included 2-7 brood cells separated by mud partitions.
The dimensions of these structures seem to increase from the first cell to the last one con-
structed by the wasp. Females always used all the available space of the trap-nests, and the
variability in the number of cells per nest essentially depended on their different dimen-
sions. All the emerged adults were females, and we suspect that this species is bivoltine,
with a highly shifted sex ratio between the two generations. In 2002, the parasite Chrysis sp.
aff. coelestina Klug, recorded for the first time on this host, was responsible for a rate of par-
asitism per nest of 0%-100%, with an average of 41.65%. A second cuckoo wasp, Chrysis vi-
olenta ultramonticola Linsenmaier, emerged from one nest in 2003. Most A. sikhimensis
females housed, mainly on the abdomen, hypopi of the mite Vespacarus sp., which is known
to be involved in other wasp-mite associations. Unlike other mite-symbiotic eumenid wasps,
A. sikhimensis does not present an acarinarium on its body to house the mites.

Key Words: trap-nest, solitary wasp, parasitism, mutualism, Himalaya, Hymenoptera, Eu-
menidae, Ancistrocerus sikhimensis


El studio de los contenidos de 21 nidos trampa, localizados en el Parque Nacional de Sagar-
matha, Nepal, ha dado a conocer aspects interesantes de la biologia deAncistrocerus sikhi-
mensis Bingham (Hymenoptera: Eumenidae). Los nidos contenian entire 2 y 7 celdas de cria,
separadas por tabiques de barro, cuyas dimensiones-tanto de las celdas como de los
tabiques-aumentaban segun iban siendo construidas. En todos los casos las hembras con-
struyeron celdas en todo el espacio disponible del nido, condicionando las dimensions de
6ste el numero de celdas por nido. Solo hembras emergieron de los nidos trampa, lo cual nos
lleva a sospechar que esta es una especie bivoltina, con un sex-ratio diferente entire las dos
generaciones. En el ano 2002 el parasito Chrysis sp. aff. coelestina Klug, colectado por prim-
era vez de este hospedador, fue el responsible de un alto parasitismo (x (por nido) =
41.65%). En el ano 2003 otro crisidido, Chrysis violent ultramonticola Linsenmaier,
emergi6 de un nido. Muchas hembras de A. sikhimensis albergaban, principalmente en el
gaster, acaros del g6nero Vespacarus en el estado de hypopus, revelando una asociaci6n en-
tre estos artr6podos. A diferencia de otros eum6nidos simbiontes con acaros, A. sikimensis no
present un acarinarium donde albergar a los acaros.

Translation provided by the authors.

Most vespoid wasps of the family Eumenidae chewed leaves, dug into the soil, or built in pre-ex-
nest individually. Females provision their brood isting cavities (Iwata 1976), and include several
with paralyzed larvae of Lepidoptera or Coleop- cells. When the wasp has collected enough food for
tera, and no overlapping of generations exist one larva, she seals the brood cell and starts to
(Iwata 1976; Krombein 1978; Bohart et al. 1982; work on a new one (Krombein 1967; Cowan 1991).
Cowan 1991). They are mass-provisioners and prey Females of the widely distributed genusAncis-
are rapidly brought to the nest after the oviposi- trocerus Wesmael nest in pre-existing cavities,
tion. Nests are built with materials such as mud or modifying them with mud (Berland 1928; Nielsen

Florida Entomologist 88(2)

1932; Bliithgen 1943, 1961; Bonelli 1969; Rich-
ards 1980), however Ancistrocerus oviventris
Wesmael typically builds aerial mud nests (Ber-
land 1928). The cavity adopted by the wasp may
be a tube in the hollowed pith of a twig, an old in-
sect bore-hole in rotten wood, or many other
kinds of holes (Bliuthgen 1943; Kurzenko 1981).
Most species separate the cells with simple mud
cell partitions, but in a few species, such asAncis-
trocerus antilope Panzer, A. parietinus L., and A.
ichneumonideus Ratzeburg, cells (located in the
cavity) are entirely made of mud (Bliuthgen
1943). This opportunistic nesting habit has al-
lowed researchers to use trap-nests in order to
study many aspects of the biology of these wasps
(Krombein 1967; Jacob-Remacle 1976; Wearing &
Harris 1999). Ancistrocerus females provision
their nests with caterpillars of several families of
Lepidoptera (Iwata 1976; Harris 1994). Some
species show considerable seasonal variations in
the sex-ratio (typically male-biased in winter and
female-biased in summer) (Fye 1965; Longair
This paper offers basic information about the
nest structure and the brood sex ratio ofAncistro-
cerus sikhimensis Bingham. Data come from a
study carried out by placing a number of wood
trap-nests in Sagarmatha National Park, Nepal.
We also report data on three organisms associ-
ated with A. sikhimensis nests, two cuckoo wasps
(Chrysididae), and a saproglyphid mite. Cuckoo
wasps are very common parasites of eumenids
(Alfken 1914; Nielsen 1932; Bonelli 1969). Associ-
ations between saproglyphid mites and solitary
wasps are known for species belonging to differ-
ent families, but in particular for Eumenidae
(Cooreman & CrBvecoeur 1948; Krombein 1961,
1967; Mostafa 1970; Pekkarinen & Hulden 1991).
This mutualistic relationship is complex and of-
ten includes venereal transmission of the mites
through wasp genital chambers (Cooper 1955), or
sometimes alternating parasitic and sapropha-
gous phases of the mite on the host body (Okabe &
Makino 2003).


Study Area

The study was carried out in Sagarmatha Na-
tional Park (2745'-2807'N, 8628'-87007'E; Solu-
Khumbu district, Nepal). The average tempera-
tures, measured during the study periods, were
12.9C in 2002 and 13.4C in 2003. In the study
area, precipitation is concentrated in the mon-
soon season, lasting from the end of May to the
end of September. In 2002 rain occurred, mostly
in June, during 32% of the whole observation pe-
riod (43 days) and in 2003 it rained every day
(100%, 22 days).


Trap-nests of different sizes were made using
wood ofAbies alba (according to Krombein, 1967).
The blocks (2 x 2 cm square section) were perfo-
rated to provide a suitable space where wasps
could establish their nests. Seven different hole
sizes were provided, with diameter ranging from
3 to 10 mm and tunnel length from 55 to 90 mm.
After removal, the trap nest holes closed with
mud were protected with a net or an adhesive
tape, to avoid flights from each trap.
The trap-nests were later reopened by sawing
along the longitudinal side and the collected ma-
terial was preserved in 70% ethanol for determi-
nation. Head width of all specimens of A. sikhi-
mensis and its chrysidid parasite collected in
2002 were measured to the nearest 0.1 mm.

Collecting Sites

Seventy-one artificial nests were placed in
2002 and forty-six in 2003. In 2002 the nests, di-
vided into 9 groups, were placed at two different
sites from May 9th to June 20th and reopened on
September 10th. Site 1 was a garden in Namche
Bazar, 3450 m, on a south-facing slope. Site 2 was
an open area outside of Kangyuma, 3700 m, on a
north-facing slope. Two groups were placed at site
1, composed of 9 (la) and 8 (Ib) blocks. Block la
was placed at 1 m" under the roof of a house, while
block lb was placed at 2 m" between stones in a
wall in the proximity of a garden in bloom. At site
2, there were seven groups (2a-g). Block 2a was on
a Rhododendron arboreum tree in bloom, and
blocks 2b-c were in a stone wall in an open area
near a wood of birches (Betula utilis) and rhodo-
dendron trees. Blocks 2d-g were under the roof of
an isolated house. Block 2a was set at 1.5 m high,
2b-c at 1 m, and blocks 2d-g at 2 m above the
ground. The nests that were not covered by roofs
were protected from the rain with transparent
plastic sheets.
In 2003, nests were placed (June 22nd) only at
site 2, in the same location of groups d-g of the
previous year, i.e., under the roof of a house, 2 m
above the ground. On July 14th nests were re-
moved and reopened the following spring (2 April


Nest Structure

In 2002, seven holes from site 2 were colonized
(Table 1). We found two nesting species: A. sikhi-
mensis and Ancistrocerus sp. (gr. parietinus) (Hy-
menoptera: Eumenidae). Six out of 7 nests were
colonized by A. sikhimensis. Since they were all
closed by a final plug of mud, they should be con-
sidered as completed nests. The number of cells

June 2005

Boesi et al.: Trap-nesting Ancistrocerus sikhimensis

IN 2002.

0 hole Tunnel length Rate of parasitism by
Nest (mm) (mm) N cells N Y /nest N parasites Chrysis sp. aff. Coelestina

894 4 68 2 1* 1 50.0
829 10 90 5 0 2 100.0
879 6 80 4 4 0 0.0
762 6 80 7 7 0 0.0
750 6 80 6 4 2 33.3
844 6 80 3 1 2** 66.6
775 6 80 1 0 1 100.0

*Larva (undetermined sex); **1 cocoon. Nest 775 was colonized byAncistrocerus sp., the other nests byA sikhimensis. Determi-
nation was based upon new born emergence, but female owners of nests 894 and 775 were captured in the field.

per nest was very variable (x = 4.14 1.87; range
= 2-7), but this value does not seem to be related
to the length of the cavity (Table 1). Apart from
the vestibular cells (sensu Krombein, 1967), in all
the other cells we found a wasp or a parasite at
the stage of adult, pupa, or larva. All the eumenid
individuals collected were females (n = 16), such
that for each nest (n = 4), the sex-ratio was 0 6 6:
1 Y Y (Table 1). Head width of females ranged
from 2.55 mm to 3.00 mm (x = 2.76 + 0.14; n = 16).
In two nests no vestibular cells were observed. No
empty cells (sensu Krombein 1967) were found.
The 4 (out of 5) empty cells found in nest 829 were
probably used by the females as brood cells, but
some eggs or the larvae at very immature stages
were already dead from unknown causes. One ad-
ditional nest (775) was colonized by another spe-
cies ofAncistrocerus.
In all trap-nests the wasps used all the avail-
able space and the differences in the number of
cells, for nests with equal burrow length, de-
pended on the variable dimensions of the cells
and/or the cell mud partitions. On average, both
the cell lengths and the cell partition thicknesses
were correlated with their order of construction
(Table 2) (Spearman rank correlation; re0n1 = 0.92,
rei partition = 0.98; n = 8; P < 0.05): vestibular cells
and closing plugs were on average larger than the
subsequent cell/cell partition, and the first cell/
cell partitions (built by the wasp at the bottom of
the nest) were the smallest ones.

In 2003, 14 active nests ofA. sikhimensis were
observed in the field. Only two of them were
closed by the wasps, and they contained only lar-
vae. In 8 nests we found adults ofA. sikhimensis,
sometimes associated with larvae. All the adults
were females. In 6 other nests (included the two
sealed ones) we found only dead larvae. In 4 nests
we found also dry Lepidoptera larvae. In 2003,
the observation of the activity ofA. sikhimensis
over six days, for a total of 10 h, gave the following
results: (a) the wasps left the nest without any
orientation flight (100%; n = 36); (b) a collected
prey was a larva of Lepidoptera (Glyphipterigi-
dae); (c) the duration of provisioning flights (x =
486 sec 524) was not different from that of non-
provisioning ones (x = 704 sec + 729; Mann-Whit-
ney Test; nl = 8; n2npf = 20; n.s.); d) the time spent
in the nest after a provisioning flight (x = 347 sec
880) was not different from that spent inside the
nest after a non-provisioning flight (x = 295 sec
405; Mann-Whitney Test; nlpf = 11; n2npf = 21; n.s.).


Nest associates include Hymenoptera Chrysidi-
dae, and Acarina Saproglyphidae. Cuckoo wasps
belong to two species, Chrysis sp. aff. coelestina
Klug (found in 2002) and Chrysis violent ultra-
monticola Linsenmaier (3 specimens found in
2003 in one nest containing even one specimen of
A. sikhimensis). The rate of parasitism due to


Cp Vc 7P 7C 6P 6C 5P 5C 4P 4C 3P 3C 2P 2C 1IP 1C

x 2.8 26.0 2.3 15.7 2.0 12.5 2.0 10.4 1.5 9.0 1.3 10.0 1.0 10.0 1 8
SD 1.3 14.5 1.5 7.1 0.9 4.9 0.7 3.2 0.6 2.9 0.6 2.0 0.0 0.0 -
n 5 5 4 6 6 6 5 5 4 4 3 3 2 2 1 1
Range 1-4 15-24 1-4 6-26 1-3 5-19 1-3 6-15 1-2 5-12 1-2 8-12 -

Cp = Closing plug; Vc = Vestibular cell; P = cell partition; C = brood cell.

Florida Entomologist 88(2)

Chrysis sp. aff. coelestina, calculated for A. sikhi-
mensis, ranged from 0% to 100%, and in general
was high (x = 41.65%; n = 6) (Table 1). All speci-
mens were females. Head width ranged from 2.00
mm to 2.65 mm (x = 2.35 + 0.20; n = 7). On 15 out
of 17 A. sikhimensis females collected from the
2002 nests we found individuals of the mite
Vespacarus sp. (Acarina: Saproglyphidae). The
mites were located mainly on the dorsal side of
the gaster of the wasps, between tergites I and II.
Six individuals had a few mites on other parts of
the body (eyes, wings, legs, neck, and thorax). All
the mites were at the hypopi stage, the typical
resting stage that these arachnids have evolved to
face adverse conditions and as an efficient dis-
persal phase (Walter & Proctor 1999). The num-
ber of mites per wasp varied from 2 to 97, but
these values should be considered underesti-
mated because of the difficulty involved in count-
ing all the specimens deeper in the abdominal
tergites. We did not observe mites on the females
emerging from the 2003 nests.


Nest Structure

As many other Ancistrocerus species, including
the closely related Ancistrocerus parietum L.
(Nielsen 1932; Krombein 1967), A. sikhimensis
nests in pre-existing cavities. Although we did not
find empty cells (sensu Krombein 1967), they have
been reported in some species of Ancistrocerus,
and they are probably constructed to better defend
the brood from parasites (e.g. Krombein 1967). A
shifted sex-ratio, female-biased in summer, has
been recorded for other species of Ancistrocerus,
probably to face the mortality due to cold winter
climatic conditions, as assumed by other authors
(Fye 1965; Longair 1981). In fact we do not know
the influence of monsoon in sex allocation in this
wasp. Moreover, considering a partial bivoltinism
model (Seger 1983), we could expect that males
live longer than females (males mate with females
of their own and next generation), resulting in a
female-bias. More data should be obtained to clar-
ify sex-ratio dynamics in this species.


Bonelli (1969) recorded Chrysis ruddii Shuck-
ard for A. oviventris; and Alfken (1914), Van Lith
(1953), Nielsen (1932) and Micheli (1930) re-
corded Chrysis ignita L. for A. parietinus,A. anti-
lope, Ancistrocerus nigricornis Curtis, and A.
oviventris. Chrysis coerulans Fabricius and Chry-
sis nitidula Fabricius were recorded as parasites
of A. antilope and Ancistrocerus catskill catskill
(Saussure) (Cooper 1953; Hobbs et al. 1961;
Medler 1964; Fye 1965; Krombein 1967). Chrysis
inflata Aaron was found as a parasite ofAncistro-

cerus durangoensis Cameron and Ancistrocerus
tuberculiceps tuberculiceps (Saussure) (Krombein
1967). Chrysis sp. aff. coelestina and Chrysis vio-
lenta ultramonticola are the first parasites re-
corded forA. sikhimensis.
Mutualistic associations between saprogly-
phid mites and solitary wasps are known for dif-
ferent species: e.g., Vidia concellaria Cooreman
and Cerceris arenaria L. (Crabronidae), Vidia coore-
mani Thomas and Ectemnius sp. (Crabronidae),
several Vidia Oudemans, Macroharpa Mostafa,
Zethacarus Mostafa, Caluolia Oudemans species
and Zethus spp. (Eumenidae) (Cooreman &
Crevecoeur 1948; Baker 1964; Mostafa 1970). Ok-
abe & Makino (2003) reported an association be-
tween Kurosaia jiju Okabe & O'Conner and
Anterhynchium flavomarginatum mikado (Smith),
where mites display an alternation of parasitic
and saprophagous phases during their life cycle
on the host. In many cases the transmission of
mites from one individual to another is known to
be venereal, because hypopi enter the genital
chamber of the female host during copulation of
the wasps (Cooper 1955; Okabe & Makino 2003).
For the genus Ancistrocerus, associations are
known between some species with Kennethiella
trisetosa (Cooreman) and Ensliniella trisetosa
Vitz. (Cooper 1955; Baker & Cunliffe 1960; Krom-
bein 1961, 1967; Cowan 1984). The genus Vespa-
carus Baker, as far as we know, is associated only
with Ancistrocerus catskill catskill and Ancistro-
cerus tigris tigris (Saussure) (now Ancistrocerus
adiabatus (Saussure)) (Krombein 1967: Vespac-
arus tigris Baker and Cunliffe). Contrary to what
we observed in our wasp-mite association, in most
Parancistrocerus species the hypopi of Vespacarus
are located in an acarinarium between tergites I
and II of the abdomen (Krombein 1967).Ancistro-
cerus adiabatus does not have a true acarinarium
and hypopi simply cluster in transversal series
under some posterior abdominal tergites (Krom-
bein 1967). This species seems to provide an inter-
mediate step toward species with a true
acarinarium. This structure evolved in some hy-
menopteran species that have mutualistic rela-
tionships with saproglyphid mites (Bequaert
1918; Giordani Soika 1985), possibly to increase
the number of host mites or, maybe, to keep them
in a fixed position of the body. However, the max-
imun (underestimated) number of 97 mites that
we have obtained from a single wasp is close to
the maximum load of mites (118) found by Krom-
bein (1967) in the acarinarium of Stenodynerus
(Parancistrocerus) f fulvipes (Saussure). Clusters
of Kennethiella trisetosa have been found on dif-
ferent parts of the body ofA. antilope, such as the
right side of the propodeum, the thorax, or on the
back of the propodeum (Cooper 1955; Cowan
1984). Mites have been also found exclusively on
the ventral surface of the thoracic segments of the
host, although rarely on the head or the lateral

June 2005

Boesi et al.: Trap-nesting Ancistrocerus sikhimensis

sides of the thorax (Okabe & Makino 2003). This
would confirm the notion that different species of
mite choose a specific part of the host body. On the
other hand it seems that acarinaria of different
wasp species, whenever they exist as in Parancis-
trocerus, are slightly differently shaped, possibly
related to the specific mites they must house
(Krombein, 1967). We collected specimens of
Vespacarus sp. on females, but most studies have
reported the presence of mites typically only on
males (Cooper 1955; Krombein 1961; Pekkarinem
& Huld6n 1991). It has been proposed (Cooper
1955) that hypopi may be unable to infest the fe-
males' bodies because the female larvae eat the
adult mites, while the male larvae allow the mites
to live on them. Our observation demonstrate
that this is not the only possible conclusion, and
that probably other factors affect the survival of
the mites in the wasp's nest cells of both sexes. In
any case, owing to the absence of A. sikhimensis
males in our nests, we do not know the extent of
their infestation by part of Vespacarus.


We are indebted to J. Gusenleitner, W. Borsato, M.
Plumari, and M. Pavesi for the identification of the eu-
menid wasps, the mites, and the cuckoo wasps, respec-
tively. Thanks are due also to Pemba Tsering Sherpa for
logistic help in the fieldwork. This study was carried out
within the framework of the Ev-K2-CNR "Scientific and
Technological Research in Himalaya and Karakorum"
Project in collaboration with the Royal Nepal Academy
of Science and Technology (RONAST) as foreseen by the
Memorandum of Understanding between the Govern-
ment of the Kingdom of Nepal and the Government of
the Republic of Italy. The research conducted was also
made possible thanks to contributions from the Italian
National Research Council and the Italian Ministry of
Foreign Affairs. Part of the work was supported by the 3-
year grant FIRB (Fondo per gli Investimenti della
Ricerca di Base) RBAU019H94-001 (2001). The financial
support ofJ. Tormos and J.D. Asis for this study was pro-
vided by the Junta de Castilla y Le6n, project SA 18/96.


ALFKEN, J. D. 1914. Zur Kenntnis der Bienenfauna von
Algerien. M6moires de la Societ6 Entomologique
Belgique 22: 185-237.
BAKER, E. W. 1964. Vidia cooremani, a new species of
Saproglyphidae from a crabronine wasp (Acarina).
Entomological News 75: 43-46.
BAKER, E. W., AND F. CUNLIFFE. 1960. Notes on sapro-
glyphid mites associated with solitary wasps. Proc.
Entomol. Soc. Washington 62: 209-231.
BEQUAERT, J. 1918. A revision of the Vespidae of the
Belgian Congo based on the collection of the Ameri-
can Museum Congo Expedition. Bull. American Mu-
seum Natural History 39: 1-384.
BERLAND, L. 1928. Hym6nopteres vespiformes, II.
Eumenidae, Vespidae, Masaridae, Bethylidae,
Drynidae, Embolemidae. Institute de Zoologie, Fac-
ult6 des Sciences, Strasbourg. 208 pp.

BLUTHGEN, P. 1943. Taxonomische und biologische no-
tizen uber palaarktische faltenwespen. Stettiner
Entomologische Zeitung 104: 149-158.
BLUTHGEN, P. 1961. Die faltenwespen mitteleuropas.
Abhandlungen der Deutschland Akademie der Wis-
senschaften. Berlin (1961) (2): 1-252.
Notes on the biology of Odynerus dilectus (Hym.: Eu-
menidae), apredator of the alfa-alfa weevil, Hypera
postica (Col.: Curculionidae). Entomophaga 27: 23-31.
BONELLI, B. 1969. Osservazioni biologiche sugli imenot-
teri melliferi e predatori della Val di Fiemme. XXX.
Bollettino dell' Istituto di Entomologia dell'Univer-
sita di Bologna 29: 155-163 + lpl.
COOPER, K. W. 1953. Biology of eumenin wasps. I. The
ecology, predation and competition of Ancistrocerus
antilope (Panzer). Trans. American Entomol. Soc. 79:
COOPER, K. W. 1955. Biology of eumenine wasps. II. Ve-
nereal transmission of mites by wasps, and some
evolutionary problems from the remarkable associa-
tion of the Ensliniella trisetosa with the wasp Ancis-
trocerus antilope. Trans. American Entomol. Soc. 80:
COOREMAN, J., AND A. CREVECOEUR. 1948. Le cycle bi-
ologique de Vidia concellaria Cooreman (Acaridiae,
Ensliniellidae) acarien vivant dans les nids de Cerc-
eris arenaria L. (Hymenoptera, Sphecidae). Bull. &
Annales de la Soci6t6 Entomologique de Belgique 84:
COWAN, D. P. 1984. Life history and male dimorphism
in the mite Kennethiella trisetosa (Acarina: Winter-
schmidtiidae), and its symbiotic relationship with
the wasp Ancistrocerus antilope (Hymenoptera: Eu-
menidae). Ann. Entomol. Soc. Am. 77: 725-732.
COWAN, D. P. 1991. The solitary and presocial Vespidae.
In K. C. Ross and R. W. Matthews (eds.). The social
biology of wasps. Comstock Publishing Associates,
Ithaca and London, 678 pp.
COWAN, D. P., AND G. P. WALDBAUER. 1984. Seasonal
occurrence and mating at flowers by Ancistrocerus
antilope (Hymenoptera: Eumenidae). Proc. Entomol.
Soc. Washington 86: 930-934.
FYE, R. E. 1965. The biology of the Vespidae, Pompil-
idae and Sphecidae (Hymenoptera) from trap nests
in northwestern Ontario. Canadian Entomol. 97:
GIORDANI SOIKA, A. 1985. Sulla presenza di acarinari
nei vespidi solitari e descrizione dell'Acarepipona in-
solita n. gen. n. sp., con un acarinario di nuovo tipo.
Bollettino del Museo Civico di Storia Naturale di
Venezia 34: 189-196.
HARRIS, A. C. 1994. Ancistrocerus gazella (Hymenop-
tera: Vespoidea: Eumenidae): a potentially useful
biological control agent for leafrollers Planotortrix
octo, P. excessana, Ctenopseustis obliquana, C. her-
ana, and Epiphyas postvittana (Lepidoptera: Tortri-
cidae) in New Zealand. New Zealand J. Crop and
Hort. Sci. 22: 235-238.
Anthophora occidentalis Cress. and its associates at
a nesting site in southern Alberta. Canadian Ento-
mol. 93: 142-148.
IWATA, K. 1976. Evolution of instinct. Comparative
ethology of Hymenoptera. Amerind Publishing Co.
Pvt. Ltd., New Delhi. 1 portrait, I-IX, 539 pp.
JACOB-REMACLE, A. 1976. Une operation nichoirs artifi-
ciels pour Hym6nopteres dans troisjardins de Liege.

Florida Entomologist 88(2)

Bull. & Annales de la Soci6t6 Entomologique de Bel-
gique 112: 219-242.
KROMBEIN, K. V. 1961. Some symbiotic relations be-
tween saprogyphid mites and solitary vespid wasps.
J. Washington Acad. Sci. 51: 89-93 + lpl.
KROMBEIN, K. V. 1967. Trap-nesting wasps and bees:
life histories, nests, and associates. Smithsonian
Press, Washington, D.C., 1 unnumbered plate, iii-vi
+ 570 pp.
KROMBEIN, K. V. 1978. Biosystematic studies of Cey-
lonese wasps. III. Life history, nest, and associates of
Paraleptomenes mephitis (Cameron) (Hymenoptera:
Eumenidae). J. Kansas Entomol. Soc. 51: 721-739.
KURZENKO, N. V. 1981. Obzer rodov odinochnykh sklad-
chatokrylykh os semejstva Eumenidae fauny SSSR
(revision of genera of Eumenidae of the URSS
fauna). Pereponch. Dal'n. Vstoka (1981): 81-112.
LONGAIR, R. W. 1981. Sex-ratio variations in xylophil-
ous aculeate Hymenoptera. Evolution 35: 597-600.
MEDLER, J. T. 1964. Parasitism of Eumeninae by cuckoo
wasps in trap-nests in Wisconsin. Proc. Entomol.
Soc. Washington 66: 209-215.
MICHELI, L. 1930. Note biologiche e morfologiche sugli
imenotteri contributeo 2). Memorie della Societa En-
tomologica Italiana 9: 46-66.
MOSTAFA, A. I. 1970. Saproglyphid hypopi (Acarina:
Saproglyphidae) associated with wasps of the genus
Zethus Fabricius. Part I. Acarologia 12: 168-192.
NIELSEN, E. T. 1932. Sur les habitudes des
Hym6nopteres aculeates solitaires. II. Vespidae,

Chrysididae, Sapygidae et Mutillidae. Entomolo-
giske Meddelelser 18: 84-174.
OKABE, K., AND S. MAKING. 2003. Life history of Kuro-
saiajiju (Acari: Winterschmidtiidae) symbiotic with
a mason wasp, Anterhynchium flavomarginatum mi-
kado (Hymenoptera: Eumenidae). Ann. Entomol.
Soc. Am. 96: 652-659.
PEKKARINEN, A., AND L. HULDEN. 1991. Distribution
and phenology of the Ancistrocerus and Symmor-
phus species in Eastern Fennoscandia (Hymenop-
tera: Eumenidae). Entomologica Fennica 2: 179-189.
RICHARDS, O. W. 1980. Scolioidea, Vespoidea and Sphe-
coidea. Handbook for Identification of British In-
sects, VI, 3. Royal Entomol. Soc. London. 118 pp.
SEGER, J. 1983. Partial bivoltinism may cause alternat-
ing sex ratio biases that favour eusociality. Nature,
Lond. 301: 59-62.
STEFFAN-DEWENTER, I. 2002. Landscape context affects
trap-nesting bees, wasps, and their natural enemies.
Ecol. Entomol. 27: 631-637.
VAN LITH, J. P. 1953. De Nederlandse metselwespen.
De Levende Natuur 12: 231-233.
WALTER, D. E., AND H. C. PROCTOR. 1999. Mites: Ecol-
ogy, Evolution and Behaviour. University of NSW
Press, Sydney and CABI, Wallingford, 352 pp.
WEARING, C. H., AND A. C. HARRIS. 1999. Evaluation of
the predatory wasp, Ancistrocerus gazella, for biolog-
ical control of leafrollers in otago fruit crops: 1. Prey
composition, nest structure and wasp productivity
from artificial nests. Biocontrol Sci. Tech. 9: 315-325.

June 2005

Deyrup: New Florida Pygmy Mole Cricket


Archbold Biological Station, P.O. Box 2057, Lake Placid, FL 33862, USA


A new species of Tridactylidae, Ellipes eisneri, is described from the Brooksville Ridge in
Florida. This is the second species of flightless tridactylid that is known to inhabit deep, well
drained sand formations in Florida. It is associated with an algal layer occurring a few mm
beneath the surface. When not feeding near the surface, E. eisneri retreats down vertical
burrows. The range of the species is not well known, but it could be restricted to the Brooks-
ville Ridge. The habits of this species are convergent with those of the previously described
Neotridactylus archboldi Deyrup & Eisner. There is a good chance that there are additional
species of upland pygmy mole crickets in other sandy areas of southeastern North America.

Key Words: Ellipes, Neotridactylus, Florida endemics, soil crusts, Brooksville Ridge, pygmy
mole cricket


Se describe una nueva especie de Tridactylidae, Ellipes eisneri, del Brooksville Ridge en el
estado de Florida. Este es la segunda especie conocida de un tridactilido no volante que ha-
bita las formaciones de arena profunda y con amplio drenaje en Florida. Es asociada con la
capa de algas que ocurre unos pocos mm debajo de la superficie. Cuando no esta alimentan-
dose sobre la superficie, E. eisneri regresa por los tuneles verticales. No se conoce bien el
rango geografico de esta especie, pero puede estar restringido al Brooksville Ridge. Los ha-
bitos de esta especie estan convergentes con los de Neotridactylus archboldi Deyrup & Eis-
ner, una especie descrita anteriormente. Hay una buena posibilidad de que existen otras
species de tridactilidos en las areas arenosas del sureste de America del Norte.

The pygmy mole crickets (Tridactylidae) are
small (usually 12 mm long or less) burrowers,
usually found in sandy habitats. They superfi-
cially resemble miniature versions of mole crick-
ets (Gryllotalpidae), but the two groups differ in
many ways, and are not at all closely related.
Pygmy mole crickets are not members of the En-
sifera, the lineage that includes the crickets
(Gryllidae) and katydids (Tettigoniidae), but
rather an early offshoot of the Caelifera, the lin-
eage that includes the grasshoppers (Acrididae)
and pygmy grouse locusts (Tetrigidae) (Blackith
1987). Blackith (1987) suggests that the antiquity
of both the Tridactylidae and the Tetrigidae is cor-
related with their specialization as algal feeders,
since algae are an ancient and conservative group
that appear to present special challenges. Some of
these challenges might be nutritional deficiency,
defensive compounds, the problem of harvesting
sufficient quantities of algae, and the tendency of
algae to dry up and disappear in non-saturated
habitats. The latter problem generally restricts
pygmy mole crickets to wet habitats, such as rain
forests, swamps and marshes, and the edges of
streams, lakes, or standing water. Most species
have well developed wings that allow individuals
to disperse if their habitat becomes too dry. Spe-
cies that live around water usually have enlarged

setae modified as flattened "swimming plates" on
their hind tibiae, or dense rows of swimming
hairs on the enlarged hind tibial spurs.
In the sandy uplands of Florida evolution has
allowed pygmy mole crickets to colonize habitats
that are dry for much of the year. These habitats
are Florida scrub, high pine, and scrubby flat-
woods, all of which were primevally maintained by
fires started by lightning. These fires in the natu-
ral ecosystems appear to have had frequencies of
roughly 5-50 years. For more detail on these fire-
structured habitats, see Myers & Ewel (1990). A
consequence of these relatively frequent fires is
that there are usually patches of bare sand here
and there among the trees, shrubs, and herbaceous
vegetation. Light easily penetrates the upper lay-
ers of the silica sand of these bare patches. At a
depth of 3-4 mm there may be a subsurface soil
crust, composed of algae, cyanobacteria, lichens,
fungi, mosses, and bacteria (Hawkes & Flechter
2002). In upland habitats of the Lake Wales Ridge
of peninsular Florida, this crust sustains a species
of pygmy mole cricket, Neotridactylus archboldi
Deyrup & Eisner (Deyrup & Eisner 1996). The gut
ofN. archboldi has been shown to contain filamen-
tous algae (Deyrup & Eisner 1996). This species is
flightless; its migrations are short and vertical:
down into the sand during dry periods, up to the

Florida Entomologist 88(2)

soil crust after a series of rains. Neotridactylus
archboldi was found on the Lake Wales Ridge, a
fossil sand dune area extending down the center of
Florida through Lake, Polk, and Highlands Coun-
ties. It seemed reasonable to expect that other
sandy uplands would also have pygmy mole crick-
ets. Moreover, there are examples of endemic
flightless insects restricted to ridges in peninsular
Florida (Deyrup 1990, 1996), and it seemed possi-
ble that various isolated upland areas might have
their own species of pygmy mole crickets. My
search for more pygmy mole crickets has not been
very intensive, and has sometimes been thwarted
by a series of droughts that have kept the mole
crickets deep underground. Most of the specimens
that I have seen from sites off the Lake Wales
Ridge resemble N. archboldi, but it is always pos-
sible that these are cryptic species. I did find, how-
ever, a series of specimens that clearly represent
an undescribed species on the Brooksville Ridge, a
large ridge that extends north and south through
Hernando, Pasco, Citrus, and Levy Counties. This
species belongs to the genus Ellipes rather than
A key to the three New World genera of Tridac-
tylidae is provided by Gunther (1975). Ellipes is
characterized by the extreme reduction of the
hind tarsi, which are represented by a minute
flap concealed between the huge hind tibial spurs
(Fig. 2A). Ellipes also lacks the prosternal spur
found in Neotridactylus.

Ellipes eisneri, new species

Description: Holotype male. Measurements in
mm: length of head and body: 3.30; length of
pronotum: 0.75; width of pronotum: 1.02; width of
head across eyes: 0.75; length of hind femur:1.92;
length of hind tibia:1.25; length of forewing: 0.72.
Coloration: background color of a fresh specimen
pale salmon; head blackish brown with narrow
cream line along inner orbits and coronal and
frontal sutures; antennae cream, terminal seg-
ments tinged with brown; pronotum with an ir-
regular median light brown transverse band,
interrupted medially; front legs pale cream, al-
most white; middle legs cream with dark mark-
ings as in Fig. 1; hind femora pale salmon, with a
small basal brown chevron, a median transverse
brown band, a brown dorsal subapical wedge, api-
cal crescents of hind femora reddish brown, hind
tibiae and tarsi cream; wings blackish brown ba-
sally, color more dilute apically, anal area pale;
tergites 5-8 with irregular brown median spots;
basal segment of cerci black, apical segment
white; ventral cercus-like organ (Paraproctfort-
satz of Gunther 1977) white.
Structural character states: antennae 10-seg-
mented; front tibia with 4 teeth (Fig. 2B); swim-
ming plates of hind tibiae absent; hind tibial
spurs with dorsal rows of well-separated fine

hairs, not dense rows of flattened hairs (Fig. 2A);
scraper present on the underside of the forewing,
forewing abbreviated, hind wings absent.
Locality of holotype male (as on specimen la-
bel, except for abbreviations): FLORIDA: Citrus
County; near Inverness; Withlacoochee Forest,
Citrus Area; Forest Road 13, 1.3 mi. south of
State Road 44; sandhill habitat with bare sand;
dug from vertical burrow. 3 April 1995. M. Dey-
rup, collector.
Deposition of holotype male: Florida State Col-
lection of Arthropods, Gainesville, FL. Descrip-
tion: Allotype female. Measurements in mm:
length of head and body: 3.55; length of pronotum:
0.87; width of pronotum: 1.15; width of head
across eyes: 0.87; Length of hind femur: 2.05;
length of hind tibia: 1.47; length of forewing: 0.75.
Coloration: as in holotype, except: pronotal brown
band not interrupted medially; middle femora
with a dark stripe connecting the median and sub-
apical bands, delimiting an irregular pale rectan-
gle; apical crescents of hind femora pale brown,
lighter than femoral maculations; hind tibiae pale
brown, pale at base. Structural characters: simi-
lar to male, including 10 segmented antennae, ex-
cept stridulatory apparatus absent on wing.
Locality, site, date, collector of allotype female
same as holotype male.
Deposition of allotype female: same as for ho-
lotype male.
Paratypes: 4 males, 3 females: same locality,
site, date, collector as holotype; dry pinned speci-
mens. 2 males, 10 females: FLORIDA: Citrus
County; "Pine Oak Estates;" State Road 488, 3.7
mi. south of junction with U.S. 41; sandhill area
with bare sand. 4 April 1996. M. Deyrup, collector.
Specimens individually in vials of alcohol.
Deposition of paratypes: 1 male, 1 female (dry):
Florida State Collection of Arthropods, Gaines-
ville, FL. 1 male, 1 female (dry): Philadelphia
Academy of Sciences, Pennsylvania. Remaining
type material temporarily deposited in the ar-
thropod collection of the Archbold Biological Sta-
tion, Lake Placid, FL.
Diagnosis: Differs from other known Ellipes in
being flightless forewingss abbreviated, hind
wings absent), its pale salmon and brown colora-
tion, and its occurrence in xeric habitats. The an-
tennae have 10 segments in both sexes.
Etymology: This new species is named in honor
of Dr. Thomas Eisner, in gratitude for his many
studies on the natural history of Florida arthro-
pods, including another species of pygmy mole
cricket, Neotridactylus archboldi. The scientific
work of Tom Eisner is remarkable for its scope,
creativity, and sheer volume, but beyond the sci-
ence he has always cheerfully confessed that a
large part of his inspiration comes from aesthet-
ics: the beauty of arthropods, the elegance of their
design. It is hoped that he will find Ellipes eisneri
a beautiful insect.

June 2005

Deyrup: New Florida Pygmy Mole Cricket



, a

Fig. 1. Ellipes eisneri, new species, male, dorsal habitus. Markings matched to live specimen. Extent of macula-
tions on pronotum and tergites highly variable in this species; in some specimens they are lacking.


Florida Entomologist 88(2)

Fig. 2. Ellipes eisneri, new species. A. Dorsal view of
right metatibia; small oval between the apical tibial
spurs is the vestigial metatarsus. B. Lateral external
view of left protibia and protarsus.

In keeping with the long tradition of Blatchley
(1920), Helfer (1953) and Capinera et al. (2001),
I am assigning an informal common name to
Ellipes eisneri, "Eisner's Pygmy Mole Cricket."
Not all orthopterists retain the old name, "pygmy
mole crickets" for the Tridactylidae, because
these insects are not crickets. Some papers (e.g.,
Gunther 1985) use the name "pygmy mole grass-
hoppers." There is, however, no rule that vernacu-
lar names must be phyletically correct, rather
than being based on correlates of habitus or ecol-
ogy, as in the "naked mole rats." Even from a phyl-
etic standpoint, it is a dubious advance to put
"grass" in the name of a lineage whose evolution-
ary and trophic divergence may well antedate the
grasses. There is also the name "false mole crick-
ets" (Naskrecki 2001), but this remains ambigu-
ous (a false mole cricket could still be a cricket),
and is annoyingly negative.
Habitat: All specimens were found in open san-
dhill habitats with a sparse cover of grasses, espe-
cially Aristida beyrichiana Trin. & Rupr., and
various herbs: Pityopsis graminifolia (Michx.)
Nutt., Polygonella robusta (Small) Horton, Paro-
nychia sp., and Balduina angustifolia (Pursh)
Robinson. There were scattered trees of Pinus
palustris Mill., Quercus laevis Walt., Q. incana
Bartr. and small clonal groups of Q. geminata
Small. The openings where pygmy mole crickets
occurred had a dark gray soil crust a few millime-
ters below the surface. The sand was a yellow en-
tisol with no visible horizon in the top 25 cm. The
specimens from the type locality (Withlacoochee
State Forest) were collected from Candler fine
sand, according to maps of the Citrus County soil
survey (Pilny et al. 1988). This is an excessively
drained, nutrient-poor, moderately to strongly
acid, fine sand occurring on sites where the water
table is at least 2 m below the surface throughout
the year; the surface layer is dark grayish brown,
the subsurface layer pale brown or yellowish
brown (Pilny et al. 1988). The paratypes from the
Pine-Oak Estates site were in an area mapped as
Astatula fine sand, whose characteristics are very
similar to Candler fine sand, but the subsurface
layer tends to be yellow or orangish yellow (Pilny

et al. 1988). Attention to soil type, even among
soils that are almost exclusively sand, may be im-
portant for understanding the distribution of bur-
rowing arthropods. Halloran et al. (2000) provide
evidence that Astatula fine sand is less stable
than Satellite sand for burrow construction by
Geolycosa spiders and Myrmeleon antlions.


The discovery of this new species hints that
there may be additional species of upland pygmy
mole crickets still to be found. Ellipes eisneri and
Neotridactylus archboldi are a pair of species that
must have evolved from flying, semiaquatic lin-
eages in their respective genera. This seems
likely because all the genera of Tridactylidae
around the world, not just Ellipes and Neotridac-
tylus, are composed of species that can fly and
occur in wet areas, so these are probably charac-
teristics of early Tridactylidae before the appear-
ance of the various synapomorphies that define
the genera. Now that we know of two lineages
that have made the transition to exploiting soil
crusts of sandy uplands, it seems possible that
there are additional lineages that have made this
same transition, especially since one of the asso-
ciated adaptive character states (loss of flying
ability) tends to isolate lineages.
Several areas in the southeastern U.S. have
ancient sand deposits with commonly occurring
patches of open sand. Any of these areas might
have concealed pygmy mole crickets. The insects
can be found by searching for their trails just af-
ter a rain, as described by Deyrup & Eisner
(1996). Most specimens of E. eisneri were ob-
tained a few days after a rain, based on a more
difficult clue: the small dark tumulus of subsur-
face sand left by the insect as it retreats down a
vertical burrow after the surface sand dries out. If
the sand is dry near the surface, but damp a few
centimeters down, it may be possible to lure
pygmy mole crickets up to the surface by watering
with a watering can. A gallon of water on a square
meter works well for N. archboldi, the subsurface
foraging trails appearing in an hour or so after
watering. Because the main obstacle to finding
upland tridactylids is the need to be at the right
place at the right time, surveys for new species or
new locality records would be good projects for lo-
cal naturalists or students in an ecology class.
Even if no pygmy mole crickets are found, the
time spent searching for them might not be
wasted. There are many other interesting arthro-
pods burrowing just below the surface of the sand,
such as blind, flightless scarabs of the genus
Geopsammodius (formerly placed in the genus
Psammodius) (Deyrup & Woodruff 1991).
The geographic range ofE. eisneri is unknown,
but it has not yet been found co-occurring with
N. archboldi, or specimens that appear to be

June 2005

Deyrup: New Florida Pygmy Mole Cricket

N. archboldi, on the Lake Wales Ridge, the North-
ern Mount Dora Ridge, the Atlantic Coastal
Ridge, and the Trail Ridge. It is possible that
E. eisneri is restricted to the Brooksville Ridge,
which has an endemic grasshopper, Melanoplus
withlacoocheensis Squitier & Deyrup (Squitier et
al. 1998), a genetically distinct population of the
gopher tortoise (Osentoski & Lamb 1995), and a
morphologically distinguishable population of a
species of ant, Dorymyrmex elegans (Trager). It
would be premature to worry about the conserva-
tion status ofE. eisneri, but the fact remains that
there is only one protected site known for the spe-
cies, and that site is only protected as long as
management of the Withlacoochee State Forest
continues to hold to its current enlightened goal
of maintaining natural habitats and natural bio-
logical diversity in the forest. It would be appro-
priate, therefore, to sample more widely on the
Brooksville Ridge during a period of wet weather
and map the range of E. eisneri to determine
whether it is a rare or endangered species.


This work was supported by the Archbold Biological
Station. I thank the managers of the Withlacoochee
State Forest for preserving and maintaining habitat for
this interesting new species.


BLACKITH, R. E. 1987. Primitive Orthoptera and primi-
tive plants, pp. 124-126 In B. M. Baccetti [ed.], Evo-
lutionary Biology of the Orthopteroid Insects, Ellis
Horwood Limited, Chichester, Britain. 612 pp.
BLATCHLEY, W. S. 1920. Orthoptera of Northeastern
America, with Special Reference to the Faunas of In-
diana and Florida. Nature Publishing Co. Indianap-
olis, Indiana: 784 pp.
2001. Grasshoppers of Florida. University Press of
Florida, Gainesville, Florida: xvii + 144 pp.
DEYRUP, M. 1990. Arthropod footprints in the sands of
time. Florida Entomol. 73: 529-538.

DEYRUP, M. 1996. Two new grasshoppers from relict up-
lands of Florida (Orthoptera: Acrididae). Trans.
American Entomol. Soc. 122: 199-211.
DEYRUP, M., AND T. EISNER 1996. Description and nat-
ural history of a new pygmy mole cricket from relict
xeric uplands of Florida (Orthoptera: Tridactylidae).
Mem. Entomol. Soc. Washington 17: 59-67.
DEYRUP, M., AND R. WOODRUFF. 1991. A new flightless
Psammodius from Florida's inland dunes (Coleop-
tera: Scarabaeidae). Coleop. Bull. 45: 75-80.
GUNTHER, K. K. 1975. Das Genus Neotridactylus
Gunther, 1972 (Saltoria, Tridactylidae, Insecta).
Mitt. Zool. Mus. Berlin 51: 305-365.
GUNTHER, K. K. 1977. Revision der Gattung Ellipes
Scudder, 1902 (Saltatoria, Tridactylidae). Deutsche
Entomol. Zeit., N. F. 24: 47-122.
GUNTHER, K. K. 1985. A new pygmy mole grasshopper
from California and Baja California, Mexico (Orthop-
tera: Tridactylidae). Pan-Pac. Entomol. 61: 139-145.
2000. Instability of sandy soil on the Lake Wales
Ridge affects burrowing by wolf spiders (Araneae:
Lycosidae) and antlions (Neuroptera: Myrmeleon-
tidae). Florida Entomol. 83: 48-55.
HAWKES, C. V., AND V. R. FLECHTER 2002. Biological
soil crusts in a xeric Florida shrubland: composition,
abundance, and spatial heterogeneity of crusts with
different disturbance histories. Microb. Ecol. 43: 1-
HELFER, J. R. 1953. How to Know the Grasshoppers,
Cockroaches and Their Allies. Wm. C. Brown Co.,
Dubuque, IA. 353 pp.
MYERS, R. L., AND T. T. EWEL (Eds.). 1990. Ecosystems of
Florida. University of Central Florida Press, Orlando.
NASKRECKI, P. 2001. Grasshoppers and their relatives,
pp. 247-264 In S. A. Levine [ed.], Encyclopedia of
Biodiversity 3. Academic Press, San Diego, CA.
OSENTOSKI, M. F., AND T. LAMB. 1995. Intraspecific phy-
logeography of the gopher tortoise, Gopherus
polyphemus: RFLP analysis of amplified mtDNA
segments. Molec. Ecol. 4: 709-718.
D. L. STANKEY. 1988. Soil Survey of Citrus County,
Florida. U.S. Dept. Agric. Soil Conserv. Serv.
A new species of Melanoplus (Orthoptera: Acrididae)
from an isolated upland in peninsular Florida. Flor-
ida Entomol. 81: 451-460.

Florida Entomologist 88(2)

June 2005


Entomology and Nematology Department, University of Florida, Gainesville, FL 32611-0620


Larra bicolor (F.) (Hymenoptera: Sphecidae) is an introduced biological control agent of pest
Scapteriscus mole crickets (Orthoptera: Gryllotalpidae) in northern Florida. The pests are of
southern South American origin. Larra bicolor is widespread in South America; the im-
ported stock is from Bolivia. Its adults seem to require nectar sources. In South America, Pu-
erto Rico (where it was also introduced from Brazil), and southern Florida (a separate
introduction from Puerto Rico), the neotropical wildflower Spermacoce verticillata L. (Rubi-
aceae) has been observed to be a favored nectar source. In northern Florida (29 N) this wild-
flower is uncommon, freezes to the ground at first winter frost, and does not flower again
until April-May. Nevertheless, where it has been planted in northern Florida, the wasps
feed on it throughout the warmer months. Wasps were observed to feed at nectaries of 10
other plant species in northern Florida. Four of these other plants were compared experi-
mentally with S. verticillata, but all received fewer visits from the wasps. Known disadvan-
tages to the use of S. verticillata to augment L. bicolor are that it is not native to Florida, and
that it grows vigorously in full sun when its roots are not immersed in water. It has been re-
ported as a minor weed in southern Florida. However, it is the best alternative to attract
L. bicolor to places where mole cricket control is needed.

Key Words: Nectar-feeding, nectar source, biocontrol, wasp-gardening, butterfly-gardening,
turfgrass, Larra bicolor, mole crickets


Larra bicolor (F.) (Hymenoptera: Sphecidae) es un agent de control biol6gico de grillotopos
del g6nero Scapteriscus (Orthoptera: Gryllotalpidae) en el norte de la Florida. Esta plaga es
originaria de Sur Am6rica. Larra bicolor se encuentra en varias parties de Sur Am6rica; las
avispas que se encuentran al norte de la Florida se importaron desde Bolivia. En Puerto Rico
(la cual se introdujo desde Brasil), y en el sur de la Florida (introducida desde Puerto Rico),
Spermacoce verticillata L. (Rubiaceae), una plant silvestre neotropical ha sido la principal
fuente de nectar para adults de esta avispa. En el norte de la Florida (29 N) esta flor es
comunmente encontrada, se congela en el invierno con la primera helada y comienza a flo-
recer nuevamente en abril o mayo. No obstante, la avispa se alimenta de esta flor en el norte
de la Florida durante los meses calidos. La avispa fue observada alimentandose de los nec-
tarios de otras 10 species de plants en el norte de la Florida. Cuatro de estas plants fu-
eron experimentalmente comparadas con S. verticillata, pero todas recibieron menos visits
de esta avispa. Algunos arguments en contra del uso de S. verticillata para aumentar las
poblaciones de L. bicolor son: la plant no es native a la Florida, crece vigorosamente bajo
exposici6n total al sol siempre y cuando las races no se encuentren bajo el agua y se ha re-
portado como una maleza menor en el sur de la Florida. Sin embargo, esta plant es la mejor
alternative para atraer L. bicolor a lugares donde el control de grillotopos es necesario.

Translation provided by the authors.

Larra bicolor (F.) is a koinobiont ectoparasi-
toid of Scapteriscus spp. mole crickets in its na-
tive range in South America (Menke 1992). In
1936-1938 stock was imported from Belem, Para,
Brazil, to Puerto Rico, and established as a classi-
cal biological control agent of S. didactylus (La-
treille) (Wolcott 1938, 1941a). In 1981, stock was
imported from Puerto Rico by J. A. Reinert and
established at Ft. Lauderdale, Florida, as a clas-
sical biological control agent of S. abbreviatus
Scudder, S. borellii Giglio-Tos, and S. vicinus

Scudder, all pests of South American origin
(Sailer 1985). Stock of the same species was im-
ported in 1988-89 from Santa Cruz, Bolivia, re-
leased and became established in and near
Gainesville in northern Florida (Frank et al.
1995). The population established at the first Ft.
Lauderdale site spread no more than 3 km, and
attempts to redistribute it failed (Castner 1988).
The stock established at Gainesville spread natu-
rally, and has now been recorded in many coun-
ties in northern Florida, to a distance of >220 km

Ar6valo & Frank: Nectar Sources for Larra bicolor

NW and S (J.H. Frank, unpublished). It was as-
sumed simply that stock obtained from Bolivia
was a more cold-hardy biotype of L. bicolor
(Frank et al. 1995) because Menke (1992) could
not distinguish them at the species level from the
Belem/ Puerto Rico stock by morphological meth-
ods. The possibility of cryptic species has not yet
been investigated. However, Menke (1992) ob-
served and illustrated what he believed to be in-
traspecific variation in punctation of the head
capsule of the adults ofL. bicolor from Belem/Pu-
erto Rico and those from Bolivia. The L. bicolor
established in northern Florida has the puncta-
tion of the latter stock (Frank et al. 1995).
Because of its assumed cold-hardiness, it
seems likely that the Bolivian stock will spread
far more widely in northern Florida. Knowledge of
the nectar sources ofL. bicolor is needed to devise
methods of improving the rate of spread both
state-wide and locally. Encouraging the establish-
ment of plants that serve as nectar sources (wasp-
gardening) could be used to enhance wasp popula-
tions, as has been done to manage other biological
control agents (Jervis 1988; Jervis & Kidd 1996).
Wolcott (1941a) collected L. bicolor adults from
flowers of Spermacoce verticillata L. (Rubiaceae)
in Belem, and imported them into Puerto Rico,
where this plant also grows (Liogier 1980). Wol-
cott (1941b) considered S. spermacoce essential to
the survival of L. bicolor in Puerto Rico. In Flor-
ida, all sites where L. bicolor was released were
prepared in advance with plantings of S. verticil-
lata, which was already widespread in southern
Florida, but very sparse farther north, reported
only in Alachua and St. Johns counties (Wunder-
lin 1979, 1998). Wolcott (1941b) also mentioned
Hyptis atrorubens Poit. (Lamiaceae) as a nectar
source for L. bicolor in Brazil and Puerto Rico.
This plant is not established in Florida (Wunder-
lin 1998). In southern Florida, Castner (1988) ob-
served that S. verticillata outperformed various
native and ornamental plants in attracting adult
L. bicolor of the Belem/Puerto Rico stock.
This paper reports research to explore several
questions regarding nectar sources ofL. bicolor in
northern Florida. Does this plant have weedy char-
acteristics? What other plant species provide use-
ful nectar sources for L. bicolor and might these
further its range expansion throughout Florida?
What other plant species may be used to encourage
local buildup of L. bicolor populations? How does
L. bicolor access nectar from S. verticillata, and
can it do the same from other plant species?
Wunderlin (1998) states that S. verticillata is
not native to Florida, but is native to the Neotro-
pical region, including Cuba, Haiti, Jamaica,
Puerto Rico, and the Bahamas. It was not de-
tected as established in Florida until the 1960s
(B. F. Hansen, pers. comm.).
In southern Florida S. verticillata flowers all
year (Bryan Steinberg, pers. comm.). However, in

northern Florida it does not flower all year; in
Gainesville (29N) it freezes to the ground at the
first frost (typically in early December), and does
not flower again until late April or early May of
the following year (JHF, observations). This limits
its availability as a nectar source in northern Flor-
ida. The limitation is of little consequence for
Larra bicolor, whose pupae diapause underground
in winter, and whose adults have been observed to
be killed by frost (Cabrera-Mireles 2000).
Spermacoce verticillata contains a low level
(0.2%) of alkaloids which would be toxic if present
in higher concentrations, but it serves as a non-
preferred forage plant for cattle (Francis 2002
and references therein).


Movement of Plants and Seeds (Weediness)

In 1990, one of us (JHF) planted a plot of S.
verticillata plants (obtained from roadside waste
land in Miami) in the grounds of the Entomology
& Nematology Department, University of Florida
and, in 1997-1998 planted five other plots (prog-
eny of the first plot) on University of Florida prop-
erty in the Gainesville area. These five were all
planted by the same method; each had 25-26
plants installed in a single line, on 60 cm centers
through a 2.6 x 16 m sheet of black polyethylene,
0.15 mm thick. Plantings were variously de-
stroyed by prolonged flooding, maintenance
crews, or a construction crew, so not all were con-
stantly available. In 2000, a seventh was planted
without mulch and in several rows by USDA col-
laborators at the USDA, Center for Medical, Ag-
ricultural, and Veterinary Entomology (CMAVE)
garden. The seven plots were installed to allow
study of the seasonality of the plant and the wasp
in northern Florida, as well as to harvest wasps
for distribution to distant localities. Collaborators
were recruited to monitor for presence of wasps in
distant counties in 2002 and 2003, and we sup-
plied them with some of the plants. We know of no
easier way of observing and collecting the wasps
than their attraction to S. verticillata flowers, al-
though they can be collected at traps baited with
phenylacetaldehyde (Meagher & Frank 1998).

What other plant species provide useful nectar sources
for L. bicolor and might these further its range expan-
sion throughout Florida?

Three of the Spermacoce verticillata plots in-
stalled in 1997-1998 in the Gainesville, FL area
were used for this 2001 study. They were at the
Beef Cattle Research Unit, the Horse Teaching
Unit, and the Fisheries and Aquatic Sciences De-
partment. A fourth, at the USDA-CMAVE garden
had many more plants, in several rows, without
plastic mulch. The plots constantly (during the

Florida Entomologist 88(2)

warmer daylight hours) had feeding adult L. bi-
color in September-November 2001.
Once every two weeks in September-Novem-
ber 2001, one of us (HAA) walked transects in the
four cardinal and four secondary compass direc-
tions away from those four plots, until he was im-
peded by structures (buildings, fences, roads) or
water bodies. Plants on which adult L. bicolor
were seen feeding were identified and recorded.
No attempt was made to analyze frequency of
wasp-feeding observations because the observa-
tions were not random. The sole purpose was to
compile a list of the plant species other than S.
verticillata on which one or more L. bicolor adults
were observed feeding in areas where S. verticil-
lata maintained a wasp population.

What plant species may be used to encourage local
buildup ofL. bicolor populations?

Sites used were the Beef Cattle Research Unit,
the G. C. Horn Turfgrass Research Laboratory,
the Plant Sciences Unit at Citra, and the USDA-
CMAVE garden. Four of the plant species identi-
fied as providing nectar to L. bicolor (Conoclin-
ium coelestinum (L.) DC, Elephantopus elatus
Bertol., Passiflora coccinea Aubl., and Solidago
fistulosa Mill., see below) were available from lo-
cal nurseries, and 32 of each were purchased in
pots. Plants of S. verticillata were already in cul-
ture. All were planted in a completely randomized
block design with four blocks. Each block was ad-
jacent to one of the existing plots of S. verticillata
to ensure that L. bicolor adults were present. The
plants were removed from pots, planted through
cuts made through a sheet of black polyethylene
in the required block design, and watered daily
for 5 days. Each plant was again watered once
each 15 d with about 0.4 L of a 0.4% N10-P52-K10
fertilizer solution/suspension to promote flower-
ing. Each of the blocks had five treatments (plant
species), with eight plants per treatment, with
each treatment in two lines of four. The separa-
tion between plants was 0.6 m within each treat-
ment, and between treatments was 2.4 m, giving
a block size of 92.8 m2. Observations were made
weekly between late July and early November
2002. Repetitions were at 10, 11, and 12 AM
(GMT-5). Data recorded were the total number of
adult L. bicolor observed, when about 20 s were
spent examining each plant. The routine used
was for one observer (HAA) to move left to right
and clockwise among the treatments, beginning
with the plant in the southwest corner of the
treatment and of the block. This was a repeated
measures experiment with a completely random-
ized block design. Analysis was made by a X2 pair-
wise comparison with the least square means
(LSM) procedure with one degree of freedom. The
data were adjusted to a Poisson distribution for
analysis in the SAS (2000) program.

This experimental design was discussed with
several researchers before it was put into opera-
tion. All recognized that each plant species has a
different floral size and architecture, that the
number of flowers produced by each plant varies in
time, and perhaps nectar production varies within
each plant. However, it was the time spent by L.
bicolor at each plant that was to be compared, so it
was not appropriate to try to control for interplant
species differences that were inherent in the com-
parison-they are not flaws in the design. Our
methods are described so that readers may accept
the results or reject them, or repeat them.

How does L. bicolor access nectar from S. uerticillata
and other plants?

Adult L. bicolor were observed feeding at nec-
taries in the field. For 40 flowers of each of the
four species (C. coelestinum, E. elatus, S. fistulosa,
and S. verticillata) having floral nectaries, the
distance from the rim of the corolla to the necta-
ries was measured in the laboratory under a dis-
secting microscope, as was the length of the
glossa of 20 adult male and 20 female wasps.


Movement of Seeds and Plants (Weediness)

An unrestrained plot of S. verticillata, planted
in 1990 on the grounds of the Entomology/Nema-
tology Department, University of Florida, Gaines-
ville, Florida by the end of 2003 had produced
infrequent seedlings in adjacent, occasionally-
mowed Bahiagrass turf, <1 m to the east, 2 m to
the north, 2 m to the south, and = 25 m to the
southwest. Mowing of adjacent turf was by rotary
mower, which, we suspect (1) prevented nearby
seedlings from flowering and (2) dispersed seeds
around a corner of a building only in a southwest-
erly direction and to a much greater distance. In
other words, a plot of the plant produced seedlings
to a distance of up to 2 m in 13 years, but use of a
rotary mower discharged a few seeds up to = 25 m
in one direction, which was presumably due to the
track of the mowing crew. Later plots were estab-
lished in 1997-1998 through = 2.6 x 16 m sheets of
black polyethylene, whose original purpose was to
allow establishment of the wildflower without
competition from other plants. At the Beef Cattle
Research Unit, after almost six years (fall 2003),
there was only one seedling plant outside (by 5 cm)
the confines of the original plot (outside the edge of
the now-damaged plastic mulch). Mowing was
done by a tractor-drawn reel mower, which may
not have dispersed seeds. At the Horse Teaching
Unit (after six years), there was spread by about 1
m to the south in places, but this probably was due
to partial redistribution of the plot by a bulldozer
moving it from its original line and destroying the

June 2005

Ar6valo & Frank: Nectar Sources for Larra bicolor

plastic mulch. Evidently S. verticillata is not
highly 'invasive.' Seedlings it produces in adjacent
turf may be controlled by occasional mowing.
These are appropriate characteristics for a nectar-
source plant for a beneficial insect: it may spread,
and, once established, it does not demand constant
care for its survival. Furthermore, S. verticillata
plants installed in 2000 at Tifton, Georgia, were
killed outright, by inadvertent application of gly-
phosate (Roundup) (W. G. Hudson, pers. comm.),
suggesting that the plant is easily controlled by
application of this chemical herbicide. In many
places, its vigorous growth is desirable.

What other plant species provide useful nectar sources
for L. bicolor and might these further its range expan-
sion throughout Florida?

Larra adults were observed feeding at nectaries
of 10 species of plants in addition to S. verticillata
(Table 1). The number of Florida counties shown
by Wunderlin & Hansen (2003) to be occupied by
the plant in question is also shown in Table 1.

What other plant species may be used to encourage local
buildup ofL. bicolor populations?

Results of the experiment are shown in Table 2
and Fig. 1. It is clear that L. bicolor adults spent
much more time at S. verticillata plants than at
any of the other four tested. We here assume this
was due to its superiority as a nectar source. There
were significant differences in all pairwise com-
parisons between plant species except between
P coccinea and S. fistulosa (where P = 0.7232).

How does L. bicolor access nectar from S. verticillata
and other plants?

The length of the glossa in relation to the floral
depth is shown in Fig. 2. There was no difference

between length of the glossa of males and females
(F = 0.20; df = 1,19; P = 0.6631). For two of the
plant species, S. fistulosa (F = 80.54; df = 1,39; P <
0.001), C. coelestinum (F = 81.86; df = 1,39; P <
0.0001), the floral depth is less than the length of
the glossa. In P coccinea, the principal nectaries
are extrafloral, and the measurement is irrele-
vant. The floral depth of S. verticillata matches
the length of the glossa (F = 1.46; df = 1, 39; P =
0.2341). The floral depth of E. elatus seems too
great to allow access by the wasp to nectaries (F =
498.02; df = 1,39; P < 0.0001). However, wasps
were observed to extend mandibles, push the head
into the flower, move the head from side to side,
and thus access nectaries with the glossa. The pet-
als are loosened from the corolla and typically fall
as the wasp removes its head or leaves the flower.


Although S. verticillata is not native to Flor-
ida, it is now widely distributed in the south of the
peninsula (Table 1). Its floral nectaries are highly
attractive to adult Larra bicolor wasps. It flowers,
and presumably provides nectar, throughout the
year in southern Florida, and for at least seven
months of the year near Gainesville (29N) in
northern Florida. No other plant has yet been
shown to rival it in Florida or Puerto Rico as an
attractant for these wasps. It has potential for use
in wasp-gardening, in which it is planted in plots
intended to enhance local populations ofLarra bi-
color wasps to help control pest mole crickets. Its
planting in areas not yet occupied by the wasp
will pave the way for arrival, establishment, and
beneficial effects of the wasp. Areas not yet occu-
pied by the wasp are most likely (a) most of south-
ern Florida, in part of which S. verticillata
already is widespread, and (b) most of the Florida
panhandle. Beneficial effects of this wasp may
also be experienced in southern Georgia.

IDA, AREA IN 2001.

Species Family Status Distribution in Florida'

Aralia spinosa L. Araliaceae native 30 counties
Conoclinium coelestinum (L.) DC Asteraceae native 58 counties
Elaphantopus elatus Bertol. Asteraceae native 58 counties
Heliotropum angiospermum Murray Boraginaceae native 18 counties
Heliotropum curassavicum L. Boraginaceae not native 17 counties
Lobularia maritima (L.) Desv. Brassicaceae not native 3 counties
Melilotus albus Medik Brassicaceae not native 40 counties
Passiflora coccinea Aubl.2 Passifloraceae not native 3 counties
Richardia brasiliensis Gomes Rubiaceae not native 51 counties
Solidago fistulosa Mill. Asteraceae native 55 counties
Spermacoce verticillata L. Rubiaceae not native 12 counties

1From Wunderlin & Hansen 2003.
'Observation by Craig Welch, graduate student, Entomology/Nematology Dept., University of Florida. It has extra-floral necta-
ries on which L. bicolor feeds.

Florida Entomologist 88(2)


Plant species 1 Plant species 2 df X2 Pr> X2

C. coelestinum E. elatus 1,326 30.14 <0.0001
C. coelestinum P. coccinea 1,326 13.48 0.0002
C. coelestinum S. fistulosa 1,326 21.57 <0.0001
C. colestinum S. verticillata 1,326 178.13 <0.0001
E. elatus P. coccinea 1,326 8.95 0.0028
E. elatus S. fistulosa 1,326 8.20 0.0042
E. elatus S. verticillata 1,326 98.40 <0.0001
P. coccinea S. fistulosa 1,326 0.13 0.7232
P. coccinea S. verticillata 1,326 122.45 <0.0001
S. fistulosa S. verticillata 1,326 169.96 <0.0001

Vernacular names assigned to S. verticillata in-
clude 'whitehead broom' (Murphy et al. 1998, said
to have been assigned by the Weed Science Society
of America) and 'shrubby false buttonweed' as-
signed by Wunderlin (1998). The Puerto Rican
common name boton blanco (= white button) was
used by Francis (2002). The native Spermacoce as-
surgens Ruiz and Pavon ('bushy buttonweed') and
non-native S. verticillata ('whitehead broom') are
reported as weeds in turfgrass in the southern
USA (Murphy et al. 1998). No golf course superin-
tendent, extension agent, or rancher with whom
we spoke recognized either of these two names
(nor did they recognize the name shrubby false
buttonweed). However, that publication alerts us
to the weedinesss', somewhere, of S. verticillata.
We tried to find a native plant in northern Flor-
ida as attractive as S. verticillata to the wasp.
This was done by searching the vicinity of estab-
lished plots of S. verticillata for evidence of feed-
ing on other plants, and then by experimental
evaluation of relative attractiveness. We did not
test the plants (mostly non-native) on which the
native wasp Larra analis (F.) was reported by

Smith (1935) to feed in Louisiana. That wasp at-
tacks only the native mole cricket Neocurtilla
hexadactyla (Perty). Further tests should be made
of a wider range of plants, including those on
which L. analis has been observed to feed, others
on which L. bicolor has been observed to feed (Ta-
ble 1), and native Florida species of Spermacoce.
Butterfly-gardeners routinely promote some
non-native weedy plants such as Buddleia and Lan-
tana species as nectar sources for butterflies, as well
as others (Asclepias, Aristolochia, etc.) for host
plants to draw interesting butterfly species. The
crops protected are Cynodon dactylon (L.). Pers. and
hybrids with C. transvaalensis Burtt-Davey (Ber-
mudagrass, the major turfgrass in southern Flor-
ida), Paspalum notatum Fluegge (Bahiagrass, the
major pasturegrass in Florida, and also used widely
as a turfgrass), and numerous kinds of vegetable
seedlings, none of which is native to Florida. We
suggest that using a small percentage of the area of
these crop plants to grow S. verticillata, another
non-native plant, is much more sensible than using
broad-spectrum chemical pesticides as the sole
means of control of non-native pest mole crickets.

A Fig. 2

S. fistulosa P coccinea

C. coelestinum S. verticillata


Fig. 1. Mean comparison for plant selection test
made by Larra bicolor in the field. Data represent the
average number of wasps per treatment per sampling
- SE. Bars with different letters are significantly differ-
ent (P < 0.005) according to chi-square pairwise compar-
isons under Poisson distribution.


E eI~ts S -otc-llat.

L distance to the nectary
----- length of glossa of L bicolor



S fistulosa C coelestinum P coccinea

Plant Species

Fig. 2. Mean comparisons of floral depth (distance to
the nectaries) and the length of L. bicolor's glossa. Data
with the same letter do not differ (a = 0.05) according to
Duncan's test. Nectaries ofP. coccinea are extra-floral.


E elatus

June 2005

Ar6valo & Frank: Nectar Sources for Larra bicolor


We thankYongsung Joo (Department of Statistics, In-
stitute of Food and Agricultural Sciences, Univ. Florida)
for statistical advice, B. F. Hansen (Univ. South Florida)
for interpreting conflicting botanical statements on the
nativity of S. verticillata, Bryan Steinberg (Ft. Lauder-
dale Research and Education Center) for comments on
the flowering period ofS. verticillata in southern Florida,
and W. G. Hudson (Univ. Georgia, Tifton) for information
on the effect of glyphosate on S. verticillata. The U.S. Golf
Association Green Section provided partial funding. We
thank F. Slansky, Jr. and R. McSorley (Gainesville), and
two anonymous reviewers for reviews of draft manu-
scripts. This is University of Florida, Agricultural Exper-
iment Stations Journal Series No. R-10205.


CABRERA-MIRELES, H. 2002. Relationship between tem-
perature and development of the ectoparasitoid
Larra bicolor (Hymenoptera: Sphecidae) and the en-
doparasitoid Ormia depleta (Diptera: Tachinidae).
Ph.D. dissertation, University of Florida.
CASTNER, J. L. 1988. Evaluation ofLarra bicolor as a bi-
ological control agent of mole crickets. Ph.D. disser-
tation, University of Florida.
FRANCIS, J. K. 2002. Spermacoce verticillata, In J. K.
Francis [ed.], Wildland Shrubs of the United States
and its Territories: Thermic descriptions. General
Technical Report IITF-WB-1. US Dept. of Agricul-
ture, Forest Service. Also available online at www.fs.fed.us/global/iitf/pdf/shrubs/Spermacoce ver-
FRANK, J. H. 1990. Mole crickets and other arthropod
pests of turf and pastures, pp. 131-139 In D. H. Ha-
beck, F. D. Bennett, and J. H. Frank [eds.], Classical
Biological Control in the Southern United States.
Southern Co-op. Ser. Bull. 355.
FRANK, J.H., AND J. P. PARKMAN. 1999. Integrated pest
management of pest mole crickets with emphasis on
the southern USA. Integr. Pest Manage. Rev. 4: 39-52.
Larra bicolor (Hymenoptera: Sphecidae), a biologi-
cal control agent of Scapteriscus mole crickets (Or-
thoptera: Gryllotalpidae), established in northern
Florida. Florida Entomol. 78: 619-623.
JERVIS, M. A. 1988. Functional and evolutionary as-
pects of mouthpart structures in parasitoid wasps.
Biol. J. Linn. Soc. 63: 462-493.
JERVIS, M. A., AND N. A. C. KIDD. 1996. Phytophagy, pp.
375-394 In M. A. Jervis and N. A. C. Kidd [eds.], In-
sect Natural Enemies. Chapman and Hall, London.
LIOGIER, A. H. 1982. Flora of Puerto Rico and Adjacent
Islands. A Systematic Synopsis. Editorial de la Uni-
versidad de Puerto Rico; Rio Piedras, PR.
MEAGHER, R., AND J. H. FRANK. 1998. Larra bicolor (Hy-
menoptera: Sphecidae: Larrinae) collected in phero-
mone- and phenylacetaldehyde-baited traps. Florida
Entomol. 81: 555-556.
MENKE, A. S. 1992. Mole cricket hunters of the genus
Larra in the New World (Hymenoptera: Sphecidae:
Larrinae). J. Hymenoptera Res. 1: 175-234. [Larra
americana Saussure is a synonym of L. bicolor (F.),
so we use the latter name exclusively no matter
what name was used by other authors cited.]

EST, D. HALL, AND L. B. MCCARTY. 1998. Weeds of
Southern Turfgrasses. University of Florida.
NICKLE, D. A. 1992. Scapteriscus borellii Giglio-Tos. The
correct species name for the southern mole cricket in
southeastern United States. Proc. Entomol. Soc.
Washington 94: 524-526 [S. acletus Rehn & Hebard,
cited by earlier authors in the USA, is a synonym of
S. borellii, so we use the name S. borellii in this pa-
per regardless of the name used by authors cited.]
NICKLE, D. A, AND J. L. CASTNER 1984. Introduced spe-
cies of mole crickets in the United States, Puerto
Rico, and the Virgin Islands (Orthoptera; Gryllotal-
pidae). Ann. Entomol. Soc. America 79: 450-465.
[Scapteriscus vicinus ofWolcott (1938, 1941a,b) and
other 20th century Puerto Rican authors is, in fact, S.
didactylus, so we use the latter name exclusively.]
SAILER, R. I. 1985. Biological control of mole crickets.
Natural enemies, pp. 23-32 In T J. Walker [ed.], Mole
Crickets in Florida. Agric. Exp. Stns, Univ. Florida,
Bull. 846 ('1984').
SAS INSTITUTE. 2000. SAS/STAT user's guide release
8.2. SAS Institute, Cary, NC.
SMITH, C. E. 1935. Larra analis Fabricius, a parasite of
the mole cricket Gryllotalpa hexadactyla Perty. Proc.
Entomol. Soc. Washington 37: 65-87. [Gryllotalpa
hexadactyla is now known as Neocurtilla hexadactyla.]
WALKER, T. J., AND D. A. NICKLE. 1981. Introduction
and spread of pest mole crickets: Scapteriscus vici-
nus and S. acletus reexamined. Ann. Entomol. Soc.
America 74: 158-163. [According to Nickle (1992), S.
acletus Rehn & Hebard is a synonym of S. borellii, so
we use the latter name regardless of the name used
by authors cited.]
WILLIAMS, F. X. 1928. Studies in tropical wasps-their
hosts and associates (with descriptions of new spe-
cies). Hawaiian Sug. Plrs' Assoc. Exp. Stn. Entomol.
Ser. Bull. 19: 1-179.
WOLCOTT, G. N. 1938. The introduction into Puerto Rico
of Larra americana Saussure, a specific parasite of
the changea" or Puerto Rican mole cricket, Scap-
teriscus vicinus Scudder. J. Agric. Univ. Puerto Rico
22: 193-218. [L. americana is a synonym ofL. bicolor
(Menke 1992), and the mole cricket was misidentified
and is in fact S. didactylus (Nickle & Castner 1984).]
WOLCOTT, G. N. 1941a. The establishment in Puerto
Rico ofLarra americana Saussure. J. Econ. Entomol.
34: 53-56. [This name is a synonym of L. bicolor
(Menke 1992).]
WOLCOTT, G. N. 1941b. The dispersion of Larra ameri-
cana Saussure in Puerto Rico. Rev. Agric. Ind. Com.
Puerto Rico 33: 607-609. [This name is a synonym of
L. bicolor (Menke 1992).]
WUNDERLIN, D. 1979. Notes on Spermacoce and Mito-
carpus (Rubiaceae) in southeastern United States.
Phytologia 41: 313-316. [Borreria verticillata is a
synonym of Spermacoce verticillata, which occurs in
Tropical America, Florida, Texas, and West Africa, so
we use the name S. verticillata exclusively, regard-
less of the name used by authors cited above.]
WUNDERLIN, R. P. 1998. Guide to the Vascular Plants of
Florida. University Press of Florida, Gainesville.
WUNDERLIN, R. P., AND B. F. HANSEN. 2003. Atlas of
Florida Vascular Plants. Available online at www.plantatlas.usf.edu>. Institute for Systematic
Botany. University South Florida, Tampa [seen De-
cember 2003].

Florida Entomologist 88(2)

June 2005


'University of Florida, Horticultural Sciences Department, Gainesville, FL 32611

2University of Florida, Gulf Coast Research and Education Center, Wimauma, FL 33698


A well developed management plan is in place for control of pests such as the twospotted spi-
der mite, Tetranychus urticae Koch (Acari: Tetranychidae), in field and greenhouse grown
strawberry, Fragaria ananassa Duchesne; however, an integrated pest management ap-
proach to control the cotton aphid, Aphis gossypii Glover (Homoptera: Aphididae), is not
available. In order to initiate an effective program for the cotton aphid, the population dy-
namics of the aphid and the effectiveness of the pink spotted lady beetle, Coleomegilla mac-
ulata DeGeer (Coleoptera: Coccinellidae), to control aphids were studied on greenhouse
strawberry. Results from this experiment established peaks of aphid infestation throughout
the growing season and location of different cotton aphid life forms on the plant. The great-
est positive response of the pink spotted lady beetle to cotton aphid occurred at high prey
density. This characteristic indicates that the pink spotted lady beetle may be a good candi-
date for augmentative biological control of cotton aphid on strawberry in the greenhouse.
This study provides a basis for developing a biological control of cotton aphid component for
an integrated strawberry pest management program.

Key Words:Aphis gossypii, biological control, Coleomegilla maculata, Fragaria, greenhouse,
integrated pest management, pests, pink spotted lady beetle, strawberry


Existen programs de manejo establecidos para el control de la aranita roja, Tetranychus ur-
ticae Koch (Acari: Tetranychidae) en el cultivo de la fresa, Fragaria ananassa Duchesne, pro-
ducida en el campo e invernadero; sin embargo, un manejo integral para el control del
pulg6n del algod6n, Aphis gossypii Glover (Homoptera: Aphididae), en el mismo cultivo, to-
davia no esta disponible. A fin de establecer la efectividad de programs de control del
pulg6n del algod6n, la dinamica poblacional del pulg6n y la efectividad del predator, Cole-
omegilla maculata DeGeer (Coleoptera: Coccinellidae), fue estudiada en el invernadero. Re-
sultados de este experiment demostraron los picos de infestaci6n durante el desarrollo del
cultivo y la localizaci6n de las diferentes formas del pulg6n en la plant. El predator res-
ponde mejor cuando la poblaci6n de pulgones es mas densa. Esta caracteristica da la indi-
caci6n del potential del predator como agent controlador de pulg6n. Estos studios proven
las bases de control biol6gico que han de implementarse en un program sostenido del
pulg6n en el cultivo de la fresa.

Translation provided by the authors.

The strawberry, Fragaria ananassa Duchesne,
is a high value crop commercially produced in
California, Florida, Michigan, North Carolina,
New York, Ohio, Oregon, Pennsylvania, Washing-
ton, and Wisconsin (Sorensen et al. 1997). Florida
ranks second in harvested area, total yield, and
production after California (USDA-FAD 2001-
2002). The Florida strawberry industry produces
during the months of November through March
in the field and high-quality production during
these months is the key to maintaining profita-
bility. The Florida strawberry industry seeks to
remain competitive during this small window of
opportunity when market prices are high and

the volume from California is low (NASS-USDA
2003). One alternative to increase strawberry
profitability is through greenhouse production.
Growing strawberries under protective culture
has become a viable alternative for strawberry
producers worldwide because of the elimination
of soil fumigation, the reduction of fungal and
bacterial diseases, and the reduction of water us-
age. At present, the area of strawberries grown
under protected cultivation in Florida is less
than 1 ha (NASS-USDA 2003); however, this is
expected to increase as growers look for new
alternatives to enhance early production (Pa-
ranjpe 2004).

Rondon et al.: Seasonal Dynamics of a Strawberry Pest

Mites are the most important pest of field and
greenhouse strawberries, with the twospotted
spider mite, Tetranychus urticae Koch (Acari: Tet-
ranychidae), a potential pest wherever strawber-
ries are produced (Oatman & McMurtry 1966;
Howard et al. 1985; Price & Kring 1991). The
twospotted spider mite has been successfully con-
trolled in the field in some areas of Florida by the
introduction of Phytoseiulus persimilis Athias-
Henriot (Acari: Phitoseiidae) onto the strawberry
crop when about 5-10% of the strawberry leaflets
have been infested with one or more mites (Van
de Vrie & Price 1994). In the greenhouse, the use
of Neoseiulus californicus McGregor to control
twospotted spider mites has been very successful
(unpublished data).
Although aphids are not a major problem on
field strawberries (Mosser & Nesheim 2003), in
greenhouse strawberries, the cotton aphid, Aphis
gossypii Glover (Homoptera: Aphididae), can be a
serious problem (Leclant & Deguine 1994). This
small, soft-bodied insect feeds on the underside of
leaves sucking out plant sap. The cotton aphid
varies in color and size (Watt & Hales 1996);
spring populations can be darker and may be
twice the size of "yellow dwarfs" generally present
in the summer (Nevo & Coll 2001). High popula-
tions of aphids can reduce the vigor of the plant,
making it susceptible to other pests. The honey-
dew that aphids excrete reduces fruit quality be-
cause of the development of a black sooty mold on
the substrate. Moreover, this sooty mold reduces
photosynthate production and otherwise reduces
the quality of the plant causing considerable eco-
nomic injury. Natural enemies are important in
control and regulation of the cotton aphid. Any
factor reducing parasitoids, predators or other
biological control agent could result in economic
damage to the crop (Kaplan & Eubanks 2002).
Natural enemies effective against the cotton
aphid include lady beetles Coccinella septem-
punctata L. and Hippodamia convergens Guerin-
Meneville, the green lacewing C'., .. 1...'. i carnea
Stephens, and wasps Lysiphlebus testaceipes
(Cress) and Aphidius colemani L. (Howard et al.
1985; Van Driesche & Bellows 1996; Kaplan &
Eubanks 2002). The pink spotted lady beetle, Co-
leomegilla maculata DeGeer (Coleoptera: Cocci-
nellidae), also is known to feed on the cotton
aphid (Rondon et al. 2004); however, the role of
the pink spotted lady beetle in the strawberry
ecosystem is relatively unknown. This polypha-
gous predator is abundant in herbaceous crops
such as maize Zea mays L., alfalfa Medicago sa-
tiva L., and potato Solanum tuberosum L. where
the lady beetle feeds on various prey (Gordon
1985; Krafsur & Obrycki 2000).
This 2-year study was initiated to monitor the
population dynamics of natural late-fall and
early-spring infestations of the cotton aphid on
strawberries grown in a greenhouse and to evalu-

ate the effectiveness of pink spotted lady beetle
third instars to control aphids.


Strawberry Production

Strawberry plants were produced at the Uni-
versity of Florida (UF), Horticultural Sciences
Department, Institute of Food and Agriculture
Sciences in Gainesville, FL, following the protocol
of Paranjpe et al. (2003). The following exceptions
were made; after cutting the runners from
mother plants, 'Sweet Charlie' strawberry plugs
were grown in the greenhouse using a low mist
fertigation system (water plus fertilizer). Four
trays of 80 plugs per tray were set on a 1 Peat:1
Perlite mix media. Plugs were exposed to a 2-
week chilling period (4.0-6.0C) in the growth
chamber before transplanting at 25.0 2.0 and
10.0 2.0C day and night temperature, respec-
tively, with a 9-h photoperiod. After the chilling
period, plugs were transplanted to 2-liter plastic
pots in soilless medium (2 Peat:1 Perlite mix).
Forty rows of four pots per row were arranged on
top of an 8-m long metal bench. Monitoring
started when four fully developed leaves ap-
peared. A weekly rotation of Quadris@ 2.08F
(azoxystrobin at 275 g active ingredient/ha) and
Nova@ 40W (myclobutanil at 142 g active ingredi-
ent/ha) sprays were made as necessary for pre-
venting powdery mildew, the main disease in
strawberry greenhouse production.

Aphid Sampling Methods

The seasonal population dynamics of the cot-
ton aphid was monitored in strawberries grown in
a greenhouse at the UF. Aphid populations were
monitored twice weekly from January to May
during 2002 and 2003. The average temperature
in the greenhouse during this experiment was 22
and 16C, day and night, respectively. The exper-
imental design was a randomized complete block
with four replications. Each block consisted of 20
plants from which five plants per replication were
randomly selected. Six rows of strawberries sepa-
rated each block. The total numbers of apterous
and alate adults and nymphs were counted in situ
from the developing bud and from the middle
strawberry leaflet (sampling unit) of one plant
with the aid of 5x and 14x lenses. The 5x lens was
used to locate the aphids, and the 14x to separate
life forms. Nymphs and "dwarf" forms were dis-
criminated from the adults based on the short
cauda plate present at the tip of the abdomen in
the immature stages as compared with a long
cauda present in the adult form (Blackman &
Eastop 2000). No insecticide was used at any time
during the investigation and the reduction of
aphid population was caused by natural "overex-

Florida Entomologist 88(2)

ploitation" of the habitat (number of aphids/leaf-
let). The elimination of old strawberries leaflets
was the only measure used as cultural control.
Aphis forbesi (Weed), the strawberry root aphid,
and other pests were physically removed from
leaves and buds to avoid any effect on the study.

Caged Greenhouse Trial

Means of the proportions of prey consumed by the
pink spotted lady beetle third instar were com-
pared and separated by the least significant dif-
ference (LSD) test (P = 0.05).


Population Dynamics of Aphids

An experiment was conducted to examine the
effectiveness of the pink spotted lady beetle third
instar as a predator of the cotton aphid. The pink
spotted lady beetle was obtained from Entomos
LLC (Gainesville, FL 32608), a local insect sup-
plier during the first year of the experiment. The
second year, the pink spotted lady beetle came
from our own colony. Lady beetles were reared fol-
lowing proprietary Entomos protocol.
In a greenhouse, five clean strawberry plants
(one plant per pot) were placed in a 1-m3 nylon cov-
ered cage. The five strawberry plants were in-
fested with adult aphids on a marked leaf. Three
cages of plants were infested with five aphids per
plant (low infestation), three cages were infested
with ten aphids per plant (medium infestation),
and three cages were infested with 15 aphids per
plant (high infestation). To diminish the possibil-
ity of the dispersal of the aphids, only three stems
per plant were kept. Based on previous observa-
tions, three stems per plant allow the sustainabil-
ity of a strawberry plant. After 1 week, the number
of aphids on the labeled compound leaf per plant
was counted. After counting was completed, one,
three, or five third instars of the pink spotted lady
beetle were released into the cages. The number of
aphids consumed on the marked leaf was counted
weekly for 4 weeks. The experiment was repeated
three times on a split block design in time. Parasit-
ized aphids and other pests were physically re-
moved from the strawberry plants.

Data Analysis

The general linear model (GLM) procedure was
used to construct analysis of variance (ANOVA)
for mean number of nymphs, apterous, and alate
adult aphids each year (SAS Institute 2000).

In 2002, overall mean numbers of nymphs ob-
served on leaves were greater than number of
nymphs observed on emerging buds (F = 26.34; df
= 3, 12; P > 0.001) (Table 1). Nymph densities on
the bud were greatest on 15 March (16.75 5.46)
and then gradually decreased towards the end of
the sampling period (Fig. 1A). Two peak popula-
tions were observed on leaves on 25 February
(15.95 4.33) and on 15 March (14.5 3.81) (Fig.
1A). Overall numbers of apterous adults on the
bud were greater than number of adults on the
leaves (F = 18.34; df = 3, 12; P > 0.001) (Table 1).
Adult density on the bud was greatest on 25 Feb-
ruary (24.55 5.32) and on 15 March (39.65
7.23) and then density gradually decreased to-
wards the end of the sampling period (Fig. 1B).
Two peaks were observed on leaves on 25 Febru-
ary (19.50 7.23) and on 15 March (17.55 4.41)
(Fig. 1B). Overall numbers of alate adults on the
bud were greater than numbers of alate adults on
leaves (F = 14.34; df = 3, 12; P > 0.001) (Table 1).
On the bud, numbers of alate adults were great-
est on 11 March (0.2 0.1) (Fig. 1C). Two peaks
were observed on 11 March (3.00 1.18) and 8
and 15 April (0.5 0.2 and 0.5 0.1, respectively)
(Fig. 1C). Overall combined numbers of aphids
(nymphs, adults and alate adults) on the bud
were greater than combined number of aphids on
the leaves (F = 12.34; df = 3, 12; P > 0.001) (Table
1). On the bud, combined numbers of aphids were
highest on 15 February (24.65 9.87) and on 15
March (56.40 11.35) (Fig. 1D). Two peaks also
were observed on leaves on 25 February (35.50
9.85) and 15 March (32.15 8.15) (Fig. 1D).
In 2003, overall mean numbers of nymphs ob-
served on leaves were greater than number of
aphids observed on the emerging bud (F = 23.18;
df = 3, 12; P > 0.068). Nymph densities on the bud


2002 2003

Life form Leaflet Developing buds Leaflet Developing buds

Nymphs 5.09 1.43 3.94 1.25 4.30 1.35 3.07 1.27
Apterous adult 6.62 1.56 8.29 1.54 6.67 1.13 9.26 2.35
Alate adult 0.03 + 0.01 0.29 + 0.11 0.46 0.23 0.03 0.01
Combined 11.90 3.11 12.26 2.80 11.23 2.89 12.36 3.63
(nymphs and apterous and alate adults)

June 2005

Rondon et al.: Seasonal Dynamics of a Strawberry Pest

q i n ; U. W '- o N Zj iCN 0 M
- C ?JC C 'rI In InM

-- 3.5
, 3 3-
'0 2.5
L_ 2
E b 1.5-
CL 1
i 0.5
0 1"


S In CI C a Cn

1 r *- = F1


LS 30


70 20 st
050 1
, 401
E 30-
S201 V \

i t 4 't z 5 J') Ln

Fig. 1. Population dynamics of nymphs (A), apterous adults (B), alate adults (C), and total (D) of the cotton aphid
on strawberries grown in a greenhouse during 2002 in Gainesville, FL. Legend: - -*- - bud -- leaf.

were greatest on 16 March (8.45 1.46) and then
gradually decreased towards the end of the sam-
pling period (Fig. 2A). Two peaks were observed on
leaves on 9 February (7.85 2.31) and on 16 March
(10.16 3.2) (Fig. 2A). Overall numbers of apter-
ous adults on the bud were greater than on the
leaves (F = 15.14; df= 3, 12; P > 0.069). Adult den-
sities on the bud were greatest on 15 February
(24.18 2.65) and on 1 March (24.96 3.26) and
then gradually decreased towards the end of the
sampling period (Fig. 2B). One peak was observed
on leaves 16 March (21.16 3.86) (Fig. 2B). Over-
all numbers of alate adults on the leaves were
greater than on buds (F = 8.37; df = 3, 12; P >
0.071). In the bud, numbers of alate adults were
low throughout the experiment while in the
leaves, it reached its highest on 9 March (3.68 +
0.9) (Fig. 2C). Overall combined numbers of aphids
on the bud were greater than on the leaves (F =
9.15; df = 3, 12; P > 0.079). On the bud, combined
numbers of aphids were greatest on 15 February
(33.16 2.89) (Fig. 2D). One peak was observed on
leaves on 16 March (37.16 3.46) (Fig. 2D).

Response of the Pink Spotted Lady Beetle Third Instar
to Different Aphid Densities

One week after five aphids were released per
plant, there was an average of 59.9 12.6 aphids
per strawberry leaf. One third instar pink spotted

lady beetle reduced 53.8, 36.2, 27.2, and 20.6% of
the aphid population after 1, 2, 3, and 4 weeks, re-
spectively (F = 13.57; df= 2, 5;P > 0.001) (Fig. 3A).
Three predators reduced 97, 28.6, 15, and 13.2%
of the aphid population 1, 2, 3, and 4 weeks, re-
spectively, after being released (F = 11.14; df = 2,
5;P > 0.078) (Fig. 3A). Five predators reduced 29,
5, 2.2, and 2.2% of the aphid population 1, 2, 3,
and 4 weeks, respectively, after being released (F
= 3.47; df= 2, 5;P > 0.090) (Fig. 3A).
One week after ten aphids were released per
plant, there was an average of 77.7 21.6 aphids
per strawberry leaf. One third instar pink spotted
lady beetle reduced 70.2, 11, 8 and 6.2% of the
aphid population after 1, 2, 3, and 4 weeks, re-
spectively (F = 21.14; df = 2, 5; P > 0.001) (Fig.
3B). Three predators reduced 87.6, 12.6, 11 and
8% of the aphid population 1, 2, 3, and 4 weeks,
respectively, after being released (F = 18.53; df =
2, 5; P > 0.001) (Fig. 3B). Five predators reduced
75.2, 22.8, 8.6, and 4.2% of the aphid population
1, 2, 3, and 4 weeks, respectively, after being re-
leased (F = 15.10; df = 2, 5; P > 0.004) (Fig. 3B).
One week after 15 aphids were released per
plant, there was an average of 167.3 28.4 aphids
per strawberry leaf. One third instar pink spotted
lady beetle reduced 96.3, 46.2, 46.3 and 39.0% of
the aphid population after 1, 2, 3, and 4 weeks, re-
spectively (F = 9.60; df = 2, 5; P > 0.001) (Fig. 3C).
Three predators reduced 98.2, 87.3, 35.4, and

Florida Entomologist 88(2)


0 M 01 C i f 0, c n N 10 e


C Alate


S 30
f| 20


* D All

, T

ti- \
^ *t^4'


S C 0 Dates n
Ni^C-i. ^enee Q 10 iio

Fig. 2. Population dynamics of nymphs (A), apterous adults (B), alate adults (C), and total (D) of the cotton aphid
on strawberries grown in a greenhouse during 2003 in Gainesville, FL. Legend: - -m- - bud ---- leaf.

23.4% of the aphid population 1, 2, 3, and 4 weeks,
respectively after being released (F = 7.34; df = 2,
5; P > 0.001) (Fig. 3C). Five predators reduced
98.2, 96.4, 57.3, and 22.4% of the aphid popula-
tion 1, 2, 3, and 4 weeks, respectively, after being
released (F = 11.75; df = 2, 5; P > 0.011) (Fig. 3C).


This study of the population dynamics of the
cotton aphid on strawberries grown under pro-
tected cultivation established a basis for the de-
velopment of future cotton aphid management.
This study gave us an insight into when to expect
possible pest outbreaks and the best time to apply
control measures. However, multi-year data will
be needed in order to establish a useful extension
of this prediction. The seasonal peaks and distri-
bution of different life forms (stages) of the pest
within plants were successfully established.
In general, the cotton aphid was more abun-
dant from mid-February to late-March, in a
greenhouse located in Gainesville, FL. During
those months, temperature in the greenhouse av-
eraged 22 and 16C, day and night, respectively,
which is favorable for aphid development and re-
production (Leclant & Deguine 1994). Several
studies already have been conducted regarding

the effect of the temperature on development,
survivorship, and reproduction on different aphid
species (Campbell et al. 1974; Aalbersberg 1987;
Wang & Tsai 2000). Moreover, studies by Camp-
bell et al. (1974) have indicated that peaks on
aphid populations were positively correlated with
moderate increases in temperatures.
Determining the location of the pest within the
plant is important in order to establish the most
effective control method. Different cotton aphid
life forms predominated at different plant loca-
tions. In both years, nymphs were more fre-
quently found on leaves than on the developing
buds but apterous adults predominated on the
buds. Alate adults were rare throughout the ex-
periment and their presence in either developing
buds or leaves was inconsistent in both years. The
hypothesis as to why a specific life form prefers a
specific site within the plant would be speculative
at this point. Although in both years we found
similar patterns, sugar content of the plant and
specific nutritional requirements of each stage
should have been taken.
The greatest positive response of the pink
spotted lady beetle to cotton aphid occurred when
the prey was most dense. This is a characteristic
of an efficient predator and indicated that the
pink spotted lady beetle might be a good candi-




S4 2



I- 2
o In

II 0.5

June 2005


Rondon et al.: Seasonal Dynamics of a Strawberry Pest



Fig. 3. Third instar pink spotted lady beetle response
to low (A), medium (B), and high (C) densities of the cotton
aphid. Legend: 1 predator; - -- - 3
predators; A- 5 predators.

date for biological control of cotton aphid on
strawberries. This is not the case for some other
lady beetle species, in which their distribution did
not always correspond with that of other aphid
species (Park & Obrycki 2004). Studies by Ron-
don et al. (2004) have determined the benefits of
using lady beetles to control cotton aphids and
twospotted spider mites. In a series of laboratory
studies, they determined that approximately 80%
of the prey offered was consumed by the pink
spotted lady beetle only after 2 h of being exposed
to the prey. The ability of lady beetles to seek
aphids was evident during the cage greenhouse
trial (personal observation).
These studies have increased the understand-
ing of the relationship among the strawberry, its
cotton aphid pest and an important biological con-
trol agent. They provide a basis for developing a
biological control of cotton aphid component to a
comprehensive integrated program of greenhouse
strawberry pest management.

We extend our sincere gratitude to Alex P. Diaz, Sa-
rah Clark, and Amanda Bisson from the University of
Florida, and Kimberly Gallagher from Entomos, for
assistance during this research. We thank Heather
McAuslane for logistical support. Thanks to Drs. Phil
Stansly and Margaret Smither-Kopperl for editorial
contribution. We appreciate the editorial comments of
Dr. Oscar E. Liburd and anonymous reviewers that im-
proved the quality of this manuscript. This research
was supported by the Florida Agricultural Experimen-
tal Station and approved for publication as Journal Se-
ries No. R-10387.


WESTHUIZEN, AND P. H. HEWITT. 1987. Development
rate, fecundity, and life span of apterae of the Rus-
sian wheat aphid, Diuraphis noxia (Mordvilko) (Ho-
moptera: Aphididae), under controlled conditions.
Bull. Entomol. Res. 77: 629-635.
BLACKMAN, R. L., AND V. F. EASTOP. 2000. Aphids on the
World's Crops: An Identification and Information
Guide. 2nd Edition. Wiley and Sons. New York. 320 pp.
ERREZ, AND M. MACKAUER 1974. Temperature re-
quirements of some aphids and their parasites. J.
Appl. Ecol. 11: 431-438.
GORDON, R. D. 1985. The Coccinellidae (Coleoptera) of
America and north of Mexico. J. N. Y. Entomol. Soc.
93: 1-912.
ALBREGTS. 1985. Diseases, Nematodes, Mites, and In-
sects Affecting Strawberries in Florida. University of
Florida Institute of Food and Agricultural Sciences,
Agricultural Experimental Station. Bulletin 857: 52.
KAPLAN, I., AND. M. D. EUBANKS. 2002. Disruption of
cotton aphid (Homoptera: Aphididae) natural enemy
dynamics by red imported fire ants (Hymenoptera:
Formicidae). Environ. Entomol. 31: 1175-1183.
KRAFSUR, E. S., AND J. J. OBRYCKI. 2000. Coleomegilla
maculata (Coleoptera: Coccinellidae) is a species
complex. Ann. Entomol. Soc. Amer. 93: 1156-1163.
LECLANT, F., AND J. P. DEGUINE. 1994. Aphids, pp. 285-
324 In G. A. Mathews and J. P. Tunstall [eds.], In-
sect Pests of Cotton. CAB, Oxon, UK.
MOSSLER, M., AND O. N. NESHEIM. 2003. Florida Crop/
Pest Management Profile: Strawberries. University
of Florida, IFAS Cooperative Extension Service,
EDIS 129.
NASS-USDA (National Agricultural Statistical Ser-
vice). 2003. Non-citrus fruits and nuts. 2002. Prelim-
inary Summary, pp. 28-31.
NEVO, E., AND M. COLL. 2001. Effect of nitrogen fertili-
zation on Aphis gossypii (Homoptera: Aphididae):
variation in size, color, and reproduction. J. Econ.
Entomol. 94: 27-32.
OATMAN, E. R., AND J. A. MCMURTRY. 1966. Biological
control of the twospotted spider mite on strawberry in
southern California. J. Econ. Entomol. 59: 433-439.
PARANJPE, A. V. 2004. Soilless media growing contain-
ers, plant densities, and cultivars for greenhouse
strawberry production in Florida. University of Flor-
ida, Horticultural Sciences Department. M.S. Thesis
Ms. 260 pp.

Florida Entomologist 88(2)

DASCO. 2003. Trends in fruit yield and quality, suscep-
tibility to powdery mildew (Sphaerotheca macularis)
and aphid (Aphis gossypii) infestation for seven straw-
berry cultivars grown without pesticides in a passively
ventilated greenhouse using pine bark as soilless sub-
strate. Proc. Fla. State Hort. Soc. 116: 740-755.
PARK, Y. L., AND J. J. OBRYCKI. 2004. Spatio-temporal
distribution of corn leaf aphids (Homoptera: Aphid-
idae) and lady beetles (Coleoptera: Coccinellidae) in
Iowa cornfields. Biol. Control. 31: 210-217.
PRICE, J. F., AND J. B. KRING. 1991. Response of
twospotted spider mite, Tetranychus urticae Koch,
and fruit yield to new miticides and their use pat-
terns in strawberry. J. Agric. Entomol. 8: 83-91.
The feeding behavior of the bigeyed bug, minute pirate
bug, and pink spotted lady beetle relative to main
strawberry pests. Environ. Entomol. 33: 1014-1019.
SAS INSTITUTE. 2000. The GLM Procedure. SAS/STAT
User's Guide Version 6, SAS Inst. Cary, NC.

OSTEEN. 1997. Summary: The Importance of Pesti-
cides and Other Pest Management Practices in U.S.
Strawberry Production. North Carolina State Uni-
versity, North Carolina Cooperative Extension Ser-
vice, E97-31588. 7 pp.
USDA-FAD (United States Department of Agriculture,
Florida Agricultural Department). 2001-2002. Flor-
ida Agriculture Statistics. http://www.usda.fas.org.
VAN DE VRIE, M., AND J. F. PRICE. 1994. Manual for Bio-
logical Control ofTwospotted Spider Mites on Straw-
berries in Florida. Univ. Fla. Dover Res. Rept. DOV
1994-1. 10 pp.
VAN DRIESCHE, R. G. V., AND T. S. BELLOWS. 1996. Biol-
ogy and arthropod parasitoids and predators, pp. 309-
335 In Biological Control. Chapman and Hall, NY.
WANG, J. J., AND J. H. TSAI. 2000. Effect of temperature
on the biology of Aphis spiraecola (Homoptera: Aphi-
didae). Ann. Entomol. Soc. Amer. 93: 874-883.
WATT, M., AND D. F. HALES. 1996. Dwarf phenotype of
the cotton aphid, Aphis gossypii Glover (Hemiptera:
Aphididae). Australian J. Entomol. 35: 153-159.

June 2005

Causton et al.: Eradication of W auropunctata


'Department of Terrestrial Invertebrates, Charles Darwin Research Station, Puerto Ayora
Santa Cruz Island, Galapagos

2USDA-ARS, CMAVE, 1600 S.W. 23rd Drive, Gainesville, FL 32608, USA


The development of effective techniques to eradicate populations of invasive ant species is
crucial to the conservation of native biodiversity. An intensive program was initiated in 2001
to eradicate the invasive little fire ant, Wasmannia auropunctata (Roger) from ~21 ha on
Marchena Island in the Galapagos Archipelago. Linear transects, approximately 10 m apart,
were cut through the vegetation of the infested area and a buffer zone of 6 ha. Amdro@ (Hy-
dramethylnon) was applied manually up to three times in the treatment area at three-
month intervals between March and October 2001. To date, five follow-up monitoring sur-
veys have placed sticks painted with peanut butter in a grid 3-4 m apart. Two small popula-
tions (0.1% of the area originally occupied by W. auropunctata) were detected in April and
October 2002 and were subsequently treated with Amdro. No W. auropunctata ants were
found in May 2003 and April 2004. Five nocturnal surveys carried out in the immediate area
of introduction of W. auropunctata did not detect any individuals. Monitoring surveys will
continue for an additional two years to ensure eradication of any remaining populations and
verify the success of this program. This paper discusses the procedures used to kill W. auro-
punctata and monitor the efficacy of the eradication methods, the program's costs, and its
applicability to other island ecosystems.

Key Words: Amdro, ant control, dispersal, costs, invasive ants, monitoring


El desarrollo de t6cnicas efectivas para erradicar poblaciones de hormigas invasoras es es-
encial para la conservaci6n de la biodiversidad native. Un program intensive fue iniciado
en 2001 para erradicar la hormiga colorada invasora Wasmannia auropunctata (Roger), de
un area de ~21 ha en la Isla Marchena, Galapagos. Transectos lineales de aproximadamente
10 m entire cada uno, fueron hechos dentro de la vegetaci6n del area infestada y en una zona
de amortaguimiento de 6 ha. Amdro (Hydramethylnon) fue aplicado manualmente hasta
tres veces en el area de tratamiento a intervalos de tres meses entire marzo y octubre 2001.
Hasta la fecha, se ha realizado cinco monitoreos para evaluar la eficacia del program de er-
radicaci6n colocando palitos pintados con mantequilla de mani en cuadriculas de 3-4 m. En
abril y octubre 2002 se detectaron dos poblaciones pequenas (0.1% del area ocupada origi-
nalmente por W. auropunctata) las cuales fueron tratados con Amdro. No se encontr6 a
W. auropunctata en Mayo 2003 y Abril 2004. Tampoco se encontr6 a la hormiga de fuego en
cinco monitoreos nocturnos realizados en la zona de introducci6n de la hormiga. Los moni-
toreos continuaran por dos aios adicionales para asegurar que no existen parches de hormi-
gas y para verificar el 6xito del program. En este articulo se discute los procedimientos para
erradicar W. auropunctata y para evaluar la eficacia de los m6todos utilizados, los costs del
program y su aplicabilidad en otros ecosistemas islenos.

Translation provided by the authors.

Ants are highly successful invaders of both is- creased distribution and when eradication is not
lands and continents (McGlynn 1999). Eradica- feasible, control techniques are costly and time
tion of recently introduced populations is funda- consuming. Moreover, they are often ineffective in
mental to preventing their dispersal and subse- preventing species from spreading through natu-
quent impacts on native biodiversity. Eradication ral mechanisms, and most importantly with the
is especially important for areas of high conserva- aid of humans (Suarez et al. 2001). Historically,
tion value where the loss of endemic fauna, in eradication has involved small, recently intro-
particular invertebrates, is at risk. The probabil- duced populations at ports of entry. Techniques
ity of successful eradication decreases with in- for eradication at a larger scale are still in the

Florida Entomologist 88(2)

early stages of development and there have been
few success stories: the removal of Wasmannia
auropunctata (Roger) from 3 ha on Santa Fe Is-
land in the Galapagos (Abedrabbo 1994) is an ex-
ample. Also, early results suggest that infesta-
tions of the ants Pheidole megacephala F. (up to
10 ha) and Solenopsis geminata (F.) (up to 3 ha) in
Kakadu National Park, Australia may have been
eradicated (Hoffman & O'Connor 2004). Yet, be-
cause of their unusual social organization and re-
productive strategies (Passera 1994; Tsutsui &
Suarez 2003), some species of ants are good can-
didates for eradication.
The little fire ant, Wasmannia auropunctata
has been listed as one of the 100 worst invaders in
the world by the Invasive Species Specialist
Group of The World Conservation Union (IUCN)
(Lowe et al. 2002). It is easily transported on
fruits and vegetables, and growing trade between
countries has facilitated this Neotropical insect's
colonization in many parts of the world. In the
last 25 years, at least seven Pacific island groups
including Hawaii and recently Tahiti have been
successfully colonized by W auropunctata (Wet-
terer & Porter 2003; E. Loeve, Fenua Animalia,
Tahiti, pers. comm.). Attributes that make W au-
ropunctata a successful invader include its adapt-
ability to a wide range of habitats, polyphagous
feeding habits, high interspecific aggression, and
lack of intraspecific aggression which leads to
unicoloniality (Ulloa-Chac6n & Cherix 1990; Le
Breton et al. 2004). Colonies are polygynous
(H11dobler & Wilson 1977), increasing the likeli-
hood that small numbers of ants that are split off
from the colony or are transported by man are
able to found a new colony.
Introduced into the Galapagos archipelago be-
tween 35 and 70 years ago, W auropunctata has
colonized eight large islands; Santa Cruz, San
Crist6bal, Isabela, Floreana, Santiago, Santa Fe,
Pinz6n, and Marchena (Silberglied 1972; Lubin
1984; Abedrabbo 1994). It also has been found re-
cently on some of the smaller islands: Champion,
Mao, Cousins, Albany, and Eden (C. E. C., unpubl.
data). The ants were most likely transported be-
tween the inhabited islands on plants, food, and
in soil. The uninhabited islands, on the other
hand, are less frequently visited and then only by
scientists and park rangers, and illegally by fish-
erman. Ants may have been transported in camp-
ing provisions and equipment or may have ar-
rived on vegetation rafts.
Known locally as the "hormiga colorada",
W auropunctata has had a wide-ranging impact
on biodiversity in the Galapagos, in particular to
native invertebrates (Clark et al. 1982; Lubin
1984; Roque-Albelo et al. 2000; Mieles 2002). It
also negatively affects the nesting activities and
young of reptiles and birds and its painful sting
makes it a significant pest to farmers and conser-
vation workers (Lubin 1985; Roque-Albelo et al.

2000; C. E. C., unpubl. data). Additionally,
W auropunctata aids the build up and spread of
populations of the cottony cushion scale (Icerya
purchase Maskell). Honeydew produced by this
scale insect is exchanged for transportation and
protection from predators (Causton 2001).
Mitigation of the impacts of W auropunctata
has been recognized as a priority for conservation
organizations in Galapagos. On the larger islands
the little fire ant is now distributed over thou-
sands of hectares and is beyond the means of cur-
rent methods of control. However, eradication
programs are expected to be more successful on
the smaller islands or areas that have been re-
cently colonized where distributions are less than
a few dozen hectares. This was demonstrated
with the successful removal of W auropunctata
from Santa Fe Island (Abedrabbo 1994). Eradica-
tion was also considered feasible for the recently
invaded Marchena Island, a near pristine island
in the northern part of the Archipelago (Roque-
Albelo et al. 2000).
Wasmannia auropunctata was first discovered
in 1988 at a campsite on Playa Negra, a large
black sand beach on the southwestern side of
Marchena Island (Fig. 1) (Roque-Albelo et al.
2000). In 1992, the area infested by W auropunc-
tata was estimated at 0.5 ha (Fig. 2a, b) and a con-
trol program was initiated by the Galapagos Na-
tional Park Service (GNPS) and the Charles Dar-
win Research Station (CDRS) adopting the meth-
odology previously used to eradicate W
auropunctata from Santa Fe Island (Abedrabbo
1994). Between 1993 and 1996, three attempts
were made to eradicate W auropunctata with Am-
dro (Zuiiga 1994; Roque-Albelo et al. 2000). Fol-
low up surveys indicated that the poison bait ap-
plications were only partially successful (Fig. 2a,
b), probably because populations were missed and
the area of infestation was underestimated. In
1996, W auropunctata still occupied 1.5 ha, but
the eradication program was suspended due to
lack of funding. By 1998, an El Nino year, the area
had increased to 17 ha (Fig. 2a, b) (Roque-Albelo
et al. 2000). High precipitation rates during El
Nino may have accounted for a rise in ant num-
bers. Lubin (1984, 1985), measured W auropunc-
tata spread at a rate of 170m/year in Santa Cruz
Island, increasing to 500 m in El Nino years. Nev-
ertheless, it is highly unlikely that El Nino was
solely responsible for such dramatic population
growth on Marchena, further suggesting that ear-
lier assessments had missed some populations.
Two years later in 2000, the infested area was es-
timated at 24 ha (Roque-Albelo et al. 2000). This
proved to be an overestimate as later calculations
showed the actual infested area to be 19.3 ha; an
increase of approximately 2.3 ha. from 1998.
What was evident from surveys carried out
post 1996 was that the distribution of W auro-
punctata in Marchena was still expanding and

June 2005

Causton et al.: Eradication of W auropunctata


,I t

= Playa Negra
1 Infested area
I) Vegetation
S Lava with small patches of vegetation

2 0 2 4 Kilometers
.. ....r

Fig. 1. Marchena Island (130 km2, inset shows location within the Galapagos Archipelago) and location of W. au-
ropunctata infestation in 2001 (-20.5 ha).

that there was a striking contrast between the
composition of ant communities in habitats where
W auropunctata was present and areas that had
not been invaded (Roque-Albelo et al. 2000). Was-
mannia auropunctata typically infested only veg-
etated areas and in Marchena, vegetation covers
only 25% of the total area of the island (130 km2).
If W auropunctata continued to spread at the
same rate, it could eliminate many of the native
invertebrate species that occupy these habitats,
especially those that are localized in distribution.
Paradoxically, the reproductive strategies that
make W auropunctata a successful colonizer and
enable it to expand its distribution also facilitate
the success of any program aimed at reducing
population numbers. This is primarily because
new colonies are typically formed by budding
(Hdlldobler & Wilson 1977), which restricts the
dispersal capacity of W auropunctata and con-
tains it to areas immediately adjacent to existing
colonies. As a consequence, eradication was still
thought to be possible and in 2001 a program was
initiated to eradicate W auropunctata from
Marchena Island. This paper evaluates the meth-
ods used in the current eradication program and

discusses their applicability to other areas of con-
servation value.


Description of Area Infested by W. auropunctata

Colonies ofW auropunctata were found between
0-50 m elevation. Marchena Island is arid and the
infested area was principally covered by areas of
dry eroded soil and fresh lava fields. Vegetation was
dense in parts, particularly in the rainy season, and
was composed of dry forest dominated by Bursera
graveolens (HBK) Trian. and Planch., Croton
scouleri Hook. f, Waltheria ovata Cav., Lantana pe-
duncularis Anderss., Opuntia helleri K. Scum., and
Castela galapageia Hook. f. (Hamman 1981). In the
Galapagos, January to May is the warm/wet season
with occasional rain and is followed by a cooler/dry
season from May to December with little or no rain
and lower temperatures. Annual meteorological
records do not exist for this island. Day time tem-
peratures recorded during field trips from 2001 to
2004 ranged from 24C to 44C with a relative hu-
midity of between 52 and 65%.

Florida Entomologist 88(2)

Ivem atlon
I Lava Mlh mial palche Illof vltaon

0 -----
1992 93 94 96 98 2000 01 02 0304

Fig. 2. A) Expansion of W. auropunctata at Playa
Negra in Marchena Island, 1992-2001. B) Change in
size of infested area. Arrows indicate applications of poi-
son bait (Amdro; see text for details).

Calculating the Size of the Treatment Area

In March 2001, 50 m longitudinal transects
were cut outwards from the perimeter established
in 2000 (Fig. 3) at 20 m intervals. Hot dogs (~5 mm
thick, made of beef) were placed on the lower ends
of 30-cm wire flags that were placed in the ground
at 5-m intervals along these transects. Baits were
checked after 45 min. In the event that W auro-
punctata was recorded, transects were extended
and additional baits placed at 5-m intervals until
ants had not been detected for 50 m. The perimeter
of the treatment area was established at least 50 m
from the last infested point found in each transect
to create a buffer zone between the infested and W.
auropunctata-free areas (Fig. 3). The perimeter
was tracked with a handheld GPS unit and the
size of the area calculated with ArcView GIS (Ver-
sion 3.2a, Environmental System Research Insti-
tute 1999). The area infested by W auropunctata
was estimated to be 20.5 ha. Wasmannia auro-
punctata was found up to 75 m away from where it
was recorded in 2000. Including the buffer zone
(6.1 ha), the area in which poison was applied and
monitoring was conducted was estimated to be
26.6 ha. These measurements are two-dimensional
and did not consider the topography of the area.

Preparation of Treatment Area

To enable the homogenous application of poison
and facilitate monitoring, the treatment area was
divided into five sectors (A, B, C, D, and I) based on
the old perimeters and natural divisions provided
by the terrain. In each sector 1.5-m wide longitu-
dinal transects were cut with machetes through
the vegetation at approximately 10-m intervals.
By 2003, a total of 352 longitudinal transects had
been cut in the treatment area ranging between 58
and 289 m in length (Fig. 3). Short latitudinal
transects were cut in areas of especially dense veg-
etation. The sectors and transects were mapped by
tracking with GPS units.

Control Techniques

Amdro (Hydramethylnon with soybean oil,
0.88% active ingredient), a product developed for
Solenopisis fire ants was used (Collins et al.
1992). This insecticide was the most attractive to
W. auropunctata of four fire-ant products tested
by Williams & Whelan (1992) and was also used
to successfully control it on Santa Fe Island (Abe-
drabbo 1994). Amdro was considered to be a
minimum risk to non-targets because of its low
toxicity to vertebrates, because it cannot be ab-
sorbed through the insect cuticle, and because it
is not known to accumulate in the environment
(Vander Meer et al. 1982; Extension Toxicology
Network 1996; Bacey 2000). Some scavenging ar-
thropods and arthropod predators, in particular
ants, were expected to feed on the bait, but any lo-
calized non-target impacts that might occur
would be negligible following re-colonization of
the treatment area by invertebrates. However,
the disadvantages of using this toxic bait are that
it decomposes quickly and cannot be applied dur-
ing or soon after rainfall (Vander Meer et al.
1982). Before each trip to Marchena, Amdro was
sampled at random and tested to ensure that it
was still attractive to the W auropunctata.
Amdro was applied after 15.00 h to reduce ex-
posure to sunlight. The bait was hand broadcast by
groups of field assistants walking parallel to each
other along adjacent transects. An average of 4.9
kg ofAmdro per hectare was applied to all sectors
in the treatment area in March and June 2001 (Ta-
ble 1). This was over double the quantity recom-
mended by specialists (2.2 kg/ha) (D. Williams,
University of Florida, Gainesville, pers. comm.),
but was considered necessary because of the hilly
terrain and the presence of caves and dense vege-
tation. Amdro was only applied to sectors A and B
in October 2001 after W auropunctata had not
been detected by the monitoring program in Sec-
tors C, D and I in June and October 2001. In April
and October 2002, the application of Amdro was
restricted to areas where small populations of
W auropunctata were found (Table 1).

June 2005

Causton et al.: Eradication of W auropunctata

Fig. 3. Treatment area for eradicating W. auropunctata (Sector A: 6.8 ha, 65 transects; Sector B: 7.5 ha, 62
transects; Sector C: 4.4 ha, 103; Sector D: 3.9 ha, 107 transects; Sector I: 4.1 ha, 15 transects). The inset is a closer
view of the 3-4 m grid of bait monitoring stations.

Monitoring the Effectiveness of Amdro Applications

The intensity of monitoring increased as Am-
dro applications decreased. In June 2001, three
months after the first application, the primary ob-
jective was to detect any further spread of W au-
ropunctata in outlying Sectors C and D (Fig. 3).
Hot dog baits were placed every 5 m in alternate
transects and techniques were similar to those
used to calculate the size of the treatment area in
March 2001. Six months after the first applica-
tion of Amdro (October 2001), a more intensive
monitoring program was begun placing peanut
butter baits in grids 3-4 m apart (Fig. 3). In Octo-
ber 2001, the grid system was applied to sectors
C, D, and part of I (bait stations were placed at 3-
4 m intervals only along the length of the
transects in the remaining sectors). For the last
four trips (April and October 2002, May 2003, and
April 2004), all sectors of the treatment area were
monitored with a 3-4 m grid of bait stations. Ad-
ditionally, in the area of introduction of W auro-
punctata (Sector A, Fig. 3) and in the areas where
small populations were found in April and Octo-

ber 2002, the distance between bait stations was
reduced to every 1 m.
Peanut butter baits were used instead of hot
dog baits because of the high proportion of hot dog
baits that were eaten by lizards and hermit crabs
on the first survey, and because peanut butter
baits were easier to use in large numbers. The
baits consisted of a wooden kebab stick (30 cm
long) painted with a fluorescent marker on one
end. Peanut butter was applied to the unpainted
end from midway down. The pointed end of the
stick was placed firmly in the ground to avoid re-
moval by lizards and doves. Monitoring activities
took place between 05:40-10:30 and 15:00-18:00 h
and were not carried out on rainy days or during
hours of intense sunlight when ants are less
abundant. Bait stations were placed every 3-4 m
along each of the longitudinal transects in the
treatment area. Additional bait stations were
placed every 3-4 m to the left and to the right of
each of these bait stations until the bait stations
on the adjacent transect were reached, thus form-
ing a grid of 3-4 m squares (Fig. 3). The number of
bait stations placed on each trip is shown in Table

Florida Entomologist 88(2)


Date Sectors Area treated (ha) Amdro (kg) kg/ha

Mar 01 A, B, C, D, I 26.6 130 4.9
Jun 01 A, B, C, D, I 26.6 134 5.0
Oct 01 A, B 14.3 60 4.2
Apr 02 Part of A and I 3.4 27 7.8
Oct 02 Part of A, all of I 10.9 45 4.2
Total 396

"Amdro applied in response to finding colony fragments ofW. auropunctata.

2. To ensure that the entire area was covered by
bait stations, 4-5 groups of field workers worked
parallel to each other along adjacent transects.
Bait was left for one hour after which it was in-
spected for ants. Field workers were trained to
identify and record W auropunctata and the three
most common ant species that they might en-
counter: Tapinoma melanocephalum (Fabricius),
Cardiocondyla emeryi Forel and Monomorium flo-
ricola (Jerdon). When a field worker believed that
they had detected W auropunctata, the ants were
collected and the site was marked. Bait sticks
were counted at the end of each transect to check
that all bait stations had been collected.

Nocturnal Monitoring

Nocturnal monitoring was carried out because
it is possible that W auropunctata resorts to feed-
ing more actively in the night when diurnal tem-
peratures are high and humidity is low (Meier
1994). Because time was limited, we surveyed
only where W auropunctata had initially been in-
troduced in Sector A (Fig. 3). Bait stations were
laid out on one night of each trip between 20:00
and 21:00 h. In October 2001, hot dog baits were
placed every 10 m for 100 m along three transects
with 30 m between transects. The number of
transects was intensified on subsequent trips and
peanut butter baits were used. Baits were placed
at 5-m intervals along the first 50 m of 10
transects in April 2002, and along 42 transects in

October 2002, May 2003, and April 2004 (Table
2). In all cases transects commenced at the beach.
Additionally, baits were laid out every 1 m in the
areas where colony fragments were found on pre-
vious trips.

Estimation of Colony Fragment Size

To determine the size of the remnant colonies
discovered in April and October 2002, we placed
peanut butter sticks in a grid with 1-m intervals
centered on the bait station where W auropunc-
tata was detected. Baits were checked after an
hour and in the event that ants were found, the
flags were left in place and the grid amplified
until ants had not been observed for 10 m in each


Wasmannia auropunctata numbers

Wasmannia auropunctata was not detected at
700 non-toxic bait stations in Sectors C and D
three months after the onset of the eradication
program in June 2001 (Table 2). After two appli-
cations of Amdro (October 2001), W auropunc-
tata was not recorded at 11,058 bait stations
placed in all sectors of the treatment area. In
April 2002, one year after the first toxic bait ap-
plication W auropunctata was recorded at three
of 33,638 non-toxic bait stations (Table 2). A pop-


Monitoring dates

Jun 01 Oct 01 Apr 02 Oct 02 May 03 Apr 04

Man hours (in the field) -392 -432 504 743 698 735
Total number of diurnal bait stations 700 11,058 33,638 36,251 44,142 40,100
Total number of nocturnal bait stations 33 110 570 780 780
Stations with W. auropunctata 0 0 3 2 0 0
Stations with other ant species 897 35 6,530 3,408 10,812

June 2005

Causton et al.: Eradication of W auropunctata

ulation of ants was located in a dry streambed in
Sector I. Along a 1-m grid of non-toxic baits the
infestation size was estimated to be at least 87 m2
and measured about 6 m by 18 m. Wasmannia au-
ropunctata was also found on two out of 36,251
non-toxic bait stations in October 2002 in Sector
A (Table 2). With baits set in a 1-m grid, the col-
ony was estimated to measure 99 m2 in an irregu-
lar patch up to 10 m wide and 21 m long. After
discovery of these populations, Amdro was ap-
plied to the infested areas and the surrounding
non-infested areas (Table 1). Wasmannia auro-
punctata was not registered in these areas on
subsequent trips. In the last two monitoring sur-
veys in May 2003 and April 2004, W auropunc-
tata was not detected at 44,142 and 40,100 bait
stations, respectively. Wasmannia auropunctata
was not recorded at any of the non-toxic bait sta-
tions placed at night during the monitoring sur-
veys (Table 2).
Based on an equidistant point method on Arc-
view, the number of non-toxic bait stations placed
in the last four monitoring surveys was calculated
to be similar to or higher than the number of sta-
tions required for complete coverage with 3 m be-
tween points on a two-dimensional plane (Table 2).

Presence of Other Ant Species

Tapinoma melanocephalum, C. emeryi, and
M. floricola were recorded from the non-toxic bait
stations in varying intensities on all four monitor-
ing trips. In October 2001 these ant species were
present in approximately 8% of the total number
of non-toxic bait stations, whereas in April and
October 2002 ants occupied 0.1% and 18% of the
stations, respectively. In May 2003 and April
2004, these ant species were again found in 7.7%
and 27% of the stations (Table 2). It is possible
that some of the identifications of C. emeryi may
have been Cardiocondyla nuda (Mayr), as these
two species are visually similar.


Efficacy of Chemical Applications

To our knowledge, this is the largest eradica-
tion program that has been attempted for W au-
ropunctata. Monitoring results suggest that the
application of Amdro along a series of closely cut
linear transects is an effective means of reducing
W auropunctata populations rapidly. Following
three applications of poison bait over a 9-month
period we have detected only two small patches of
ants in approximately 0.1% of the area originally
infested by W auropunctata. A larger number of
nest remnants was expected, especially given the
difficult terrain. Negative results with intensive
monitoring techniques in Sectors C and D sug-
gests that W auropunctata spread was contained

and that ants may have been eradicated from this
area after two applications of Amdro. Neverthe-
less, as intensive monitoring was only carried out
in all sectors beginning with the third applica-
tion, we cannot make any determinations about
the effectiveness of each individual application in
eradicating ants from the entire infested area. It
may have been that there were only a few survi-
vors after the first application. However, because
funding was limited and we wanted to guarantee
that the populations were hit hard, additional
bait applications and land clearing activities were
given priority over monitoring surveys.
The apparent effectiveness of the chemical ap-
plications may have been augmented by the effect
of extended dry periods following the first bait ap-
plication in March 2001. The lower elevations of
Marchena Island are typically very arid during the
dry season (May to December) and there was very
little green vegetation on subsequent trips in June
and October 2001. Wasmannia auropunctata pre-
fers moist habitats and only dominates arid zones
when temperature and humidity are high (Clark
et al. 1982; Lubin 1984, 1985). These dry condi-
tions probably inhibited nest-founding activities in
any nests that were not destroyed by the first
chemical application. Both W auropunctata den-
sity and the production of sexual appear to be in-
fluenced by humidity, and decreases in both occur
in the drier months (Clark et al. 1982; Ulloa-
Chac6n 1990). Furthermore, ant nests are typi-
cally found above ground (Lubin 1984, 1985; Ulloa-
Chac6n & Cherix 1990; Ambrecht & Ulloa-Chac6n
2003) and are susceptible to drying out when hu-
mid nesting sites are less abundant (Lubin 1984).
During the monitoring surveys, two small pop-
ulations of W auropunctata were discovered.
These may have been missed by the chemical ap-
plications because of the hilly and volcanic ter-
rain, particularly at the beginning when the
methodology was still being worked out and the
distance between transects was larger. This is the
most likely explanation for the small population
discovered in Sector I in April 2002 where Am-
dro was only applied twice. However, it is less
likely that the population discovered in Sector A
in October 2002 was missed, as Amdro was ap-
plied three times in this area. The topography of
the land may have influenced the success rate of
some of the applications, while some poison bait
may have been deactivated by high temperatures
during shipment to the Galapagos. Intensive mon-
itoring along a 3-4 m grid was only initiated in
these sectors in April 2002 and may explain why
these populations were not detected earlier. It is
also possible that nests that were partially hit by
the Amdro applications in 2001 may have taken
some time to build up population numbers and ini-
tiate foraging activities. For example, in New
Zealand non-toxic baits did not attract the Argen-
tine ant, Linepithema humile (Mayr) nine months

Florida Entomologist 88(2)

after treatment with toxic bait although searching
revealed their presence (Harris et al. 2002).
Conversely, at least three sympatric ant spe-
cies (T melanocephalum, C. emeryi, and M. flori-
cola) were collected from the non-toxic baits fol-
lowing the application of Amdro. These intro-
duced species were present in the W auropunc-
tata infested area before treatment began (Roque-
Albelo et al. 2000; Mieles 2002) suggesting that
they either were not affected as much by the toxic
bait or had re-invaded rapidly after treatment.
Fluctuations in ant numbers corresponded to pat-
terns observed in non-infested areas during the
same period and appear to be related to climate
(Mieles 2002).

Effectiveness of Monitoring

The monitoring techniques used during this
program should have been sufficient to detect the
presence of W auropunctata. Baits were made up
of peanut butter, which has been shown to be
highly attractive to W auropunctata in the labo-
ratory and in the field (Williams & Whelan 1992).
Studies on the foraging behavior of W auropunc-
tata suggest that if ants were present in the area
they would have been attracted to the non-toxic
baits under most climatic conditions (including
strong wind, heavy rain, and full sunlight) and at
all times of the day, although ant numbers may
vary (Clark et al. 1982; Meier 1994; Delsinne et
al. 2001). The distance between bait stations
should have permitted ants to reach the baits
within an hour. Wasmannia auropunctata typi-
cally makes superficial nests under most environ-
mental conditions (Lubin 1984, 1985; Ulloa-
Chac6n & Cherix 1990; Ambrecht & Ulloa-
Chac6n 2003). Although little is known about the
foraging distances of W auropunctata, workers
have been observed to forage up to 2 m high in
trees (de la Vega 1994; Meier 1994). With a mean
foraging speed of between 15-18 cm/min (Meier
1994), and assuming that ants could detect baits
up to ~2.1 m away (radius of the circle defined by
the grid size), ants should have been recruited to
the baits within an hour. Extended drought peri-
ods, however, are associated with lower ant abun-
dance (Clark et al 1982; Ulloa-Chacon 1990) and
have been known to cause hypogaeic nesting in
other parts of the Galapagos (Abedrabbo 1994;
Meier 1994). Thus, some nests may not have been
able to locate the bait within an hour under these
conditions. Although, we did not find any nests
below the ground in Marchena, we have re-
stricted our monitoring efforts in the last two
years to the end of the wet season when surviving
colonies are expanding and food is in demand.
Populations that have been reduced by toxic
baits also may be slow in reacting to the non-toxic
bait stations. When peanut butter bait stations
were set 3-4 m apart, W auropunctata was de-

tected at only three bait stations in April 2002 and
two bait stations in October 2002. Yet, on both oc-
casions, the area occupied by W auropunctata
proved to be larger (87 m2 and 99 m2, respectively),
as was discovered when the distance between bait
stations was reduced to 1 m. Approximately 13 bait
stations should have picked up W auropunctata at
3-m intervals. This may be because the popula-
tions were small and workers took longer to find
and recruit to the baits at wider spacing.
Studies on the foraging behavior of W auro-
punctata have not been repeated in different cli-
matic conditions sufficiently to identify optimal
conditions for monitoring. While it is likely that
non-toxic bait stations were laid out when W au-
ropunctata was active, we suggest that repeated
experimental trials be carried out with different
population sizes (including those that have been
partially hit by toxic bait applications) to deter-
mine foraging speed, distance, and peak foraging
hours under all climatic conditions and that mon-
itoring activities are modified accordingly. Never-
theless, provided that monitoring is maintained
for several years it is likely that any surviving
pockets of W auropunctata should grow large
enough to be detected at the level of intensity be-
ing employed in this study.

Could W. auropunctata Exist Outside the Containment

Current evidence suggests that outlying popu-
lations are unlikely on Marchena unless indepen-
dent introductions have occurred elsewhere on
the island. Wasmannia auropunctata has not
been collected from six batteries of pitfall traps
randomly placed within a 500-m radius of the
treatment area on six occasions between 2000
and 2004 (Mieles 2002; A. Mieles, CDRS, Galapa-
gos, pers. comm.), nor has it been collected in sur-
veys that have been initiated on other parts of the
island (C. S., unpubl. data).
These findings seem to indicate that W auro-
punctata has not used long distance dispersal as a
means for spreading in Marchena. In areas where
it has been introduced, W auropunctata typically
forms new colonies by budding, where insemi-
nated queens are accompanied by workers on foot
to a nearby site (e.g., Hdlldobler & Wilson 1977;
Lubin 1984). This leads to well-demarcated
boundaries of infested versus non-infested areas
as shown by Clark et al. (1982) and which were
observed on Marchena Island. This dispersal
strategy is corroborated by observations in the
laboratory of intranidal mating and by the fact
that queens were unable to establish new colonies
in the absence of workers (Ulloa-Chac6n 1990; Ul-
loa-Chac6n & Cherix 1990). Furthermore, work-
ers have been observed moving winged queens on
Santa Cruz Island in the Galapagos (Clark et al.
1982), and until recently nuptial flights of W au-

June 2005

Causton et al.: Eradication of W auropunctata

ropunctata had never been observed either in the
field or laboratory (Spencer 1941; Sielberglied
1972; Lubin 1984; Ulloa-Chac6n 1990). Mating
flights have been reported, however, among
W auropunctata populations in Puerto Rico
(Torres et al. 2001). It is evident that there are
still many gaps in our knowledge of the popula-
tion biology of W auropunctata, highlighting the
need for continued monitoring in Marchena and
further studies in its native and introduced range
to better understand the mechanisms used for
colony reproduction.

Future Needs and Application to Other Island

Given the track record of W auropunctata, it
can only be a matter of time before it is introduced
into other islands, especially in the Pacific. We
strongly recommend that early warning systems
are set up and that rapid response plans are
available in the event that W auropunctata is de-
tected. We also recommend pre-approval of effec-
tive bait treatment because this can save months
of delays if suitable treatment products are not
currently registered for use.
Our results from Marchena indicate that these
eradication techniques are effective for limiting
the spread and possibly also for eradicating well
established populations of W auropunctata of up
to at least 20 ha in size. Aerial application should
be considered if the infestation is more than a few
hectares and suitable aircraft are reasonably
available. Amdro is relatively safe to use in con-
servation areas but we recommend that toxicity
studies be carried out on non-targets before apply-
ing the poison bait to areas where re-colonization
of invertebrates from outlying areas is not possi-
ble. Caution should also be used in areas with wa-
ter sources because of its toxicity to fish (Extension
Toxicology Network 1996) and possible impact on
aquatic invertebrates. Chemical applications are
likely to be more successful at the beginning of the
dry season when the reproductive potential of W.
auropunctata is reduced and toxic bait applica-
tions are more effective. Post application surveys
are crucial to the success of the program and are
the only way to ensure that W auropunctata has
been eliminated. Surveys should be carried out
only in the rainy season. Although labor intensive,
there is no substitute for detailed mapping of the
area with bait sticks. The smaller the grid size the
greater the accuracy in evaluating the effective-
ness of the poison bait applications.
Ultimately the intensity of the monitoring will
depend upon financial and manpower constraints,
but we believe that intense early monitoring may
provide savings in the long run. To date, the pro-
gram in Marchena has cost approximately
$212,736 US (this includes time spent preparing
for the field trips, field and laboratory work, and

overhead). The cost for the purchase and shipping
ofAmdro was approximately $10,700. Assuming
that no more ants are found on the next two mon-
itoring trips, the total projected cost for removing
W auropunctata for each hectare of infested area
is estimated at $13,680. Personnel costs accounted
for approximately 47% of the total spent on this
program and can be reduced by using trained vol-
unteers. Approximately 25% of these costs were
for inter-island transport and surveys to evaluate
the response of native invertebrate communities
and may not be needed elsewhere. Additional
studies on foraging behavior and the refinement of
bait application and monitoring procedures
should help us improve these techniques and
make them less labor intensive and costly.

We acknowledge the work carried out by Victor Car-
ri6n, R. Jimenez, F. Gaona, C. Gaona, E. Cadena, W. Es-
pinoza, F. Azuero, and A. Ballesteros and the rest of the
team from the Galapagos National Park Service, our
partners in this project. We thank P. Ulloa-Chac6n for
providing information on the reproductive biology of W.
auropunctata. Our thanks also to H. Snell and J. Calle-
baut for help with Arcview; R. Adams, R. Harris, J. Le
Breton, H. Rogg, and D. Williams for reviewing the
manuscript; and to the many volunteers who have
helped on this program. A special thanks to L. Roque-
Albelo, one of the cofounders and strongest supporters
of this program, for his input into the study design, com-
panionship in the field, and critical review of the manu-
script. We are grateful to BASF and TopPro Specialities
for the donation of Amdro; B. Smith, K. Miller, H.
Kraus, J. Turner, and M. Haslett for patience in organiz-
ing its shipment; and SESA, Ecuador for the necessary
import permits. This project was partially financed by
the UNF/UNFIP project SCO-LAC-99-072-Control
and Eradication of Invasive Species: a Necessary Condi-
tion for Conserving Endemic Biodiversity of the Galapa-
gos World Heritage Site. This is contribution No. 1004
from the Charles Darwin Research Station.
The use of trade, firm, or corporation names in this
publication is for the information and convenience of
the reader. Such use does not constitute an official en-
dorsement or approval by the Charles Darwin Founda-
tion, United States Department of Agriculture, or the
Agricultural Research Service of any product or service
to the exclusion of others that may be suitable.

ABEDRABBO, S. 1994. Control of the little fire ant Was-
mannia auropunctata, on Santa Fe Island in the
Galapagos Islands, pp. 219-227 In D. F. Williams
[ed.], Exotic Ants: Biology, Impact, and Control of In-
troduced Species. Westview Studies in Insect Biol-
ogy, Boulder, CO.
AMBRECHT, I., AND P. ULLOA-CHACON. 2003. The little fire
ant Wasmannia auropunctata (Roger) (Hymenop-
tera: Formicidae) as a diversity indicator of ants in
tropical dry forest fragments of Colombia. Environ.
Entomol. 32(3): 542-547.
BACEY, J. 2000. Environmental fate of Hydramethyl-
non. Report for Environmental Monitoring and Pest

Florida Entomologist 88(2)

Management Branch, California Department of Pes-
ticide Regulation, Sacramento.
CAUSTON, C. E. 2001. Dossier on Rodolia cardinalis
Mulsant (Coccinellidae: Cocinellinae), a potential bi-
ological control agent for the cottony cushion scale,
Icerya purchase Maskell (Margarodidae). Charles
Darwin Foundation, Galapagos.
ant Wasmannia auropunctata: autoecology and ef-
fects on ant diversity and distribution on Santa Cruz
Island, Galapagos. Biotropica 14: 196-207.
LADNER 1992. Seasonal trends in effectiveness of
hydramethylnon (AMDRO) and fenoxycarb (LOGIC)
for control of red imported fire ants (Hymenoptera:
Formicidae). J. Econ. Entomol. 85: 2131-2137.
DE LA VEGA, I. 1994. Food searching behavior and com-
petition between Wasmannia auropunctata and na-
tive ants on Santa Cruz and Isabela, Galapagos
Islands, pp. 73-79 In D. F. Williams [ed.], Exotic Ants:
Biology, Impact, and Control of Introduced Species.
Westview Studies in Insect Biology, Boulder, CO.
Premieres donn6es sur la monopolisation de resources
par l'envahisseur Wasmannia auropunctata (Roger) au
sein dune myrmecofaune de fort s6che Neo-Cal6doni-
enne. Actes Coll. Insectes Sociaux 14: 1-5.
ArcView GIS Version 3.2 for Windows. Redlands, CA.
formation Profiles. Hydramethylnon. Cooperative
Extension Offices of Cornell University, Oregon State
University, the University of Idaho, and the Univer-
sity of California at Davis and the Institute for Envi-
ronmental Toxicology, Michigan State University.
HAMMAN, O. 1981. Plant communities of the Galapagos
Islands. Dansk Botanisk Arkiv. 34(2): 1-163.
HARRIS, R. J., J. S. REES, AND R. J. TOFT. 2002. Trials to
eradicate infestations of the Argentine ant, Linepi-
thema humile (Hymenoptera: Formicidae), in New
Zealand, pp. ??-?? In S. C. Jones, J. Zhai, and W. H.
Robinson [eds.], Proc. of the 4th International Con-
ference on Urban Pests. Pocahontas Press, VA.
HOLLDOBLER, B., AND E. O. WILSON. 1977. The number
of queens: an important trait in ant evolution.
Naturwissensch. 64: 8-15.
HOFFMAN, B. D., AND S. O'CONNOR. 2004. Eradication of
two exotic ants from Kakadu National Park. Ecol.
Mgmt Restor. 5: 98-105.
JEAN, AND H. JOURDAN. 2004. Experimental evi-
dence of large-scale unicoloniality in the tramp ant
Wasmannia auropunctata (Roger). J. Insect. Behav.
(In press).
LOWE, S., M. BROWNE, AND S. BOUDJELAS. 2002. 100 of
the world's worst invasive alien species. A selection
from the invasive species database. Invasive Species
Specialist Group, Auckland, New Zealand.
LUBIN, Y. D. 1984. Changes in the native fauna of
Galapagos Islands following invasion by the little
red fire ant Wasmannia auropunctata. Biol. J. Lin-
nean Soc. 21(1-2): 229- 242.
LUBIN, Y. D. 1985. Studies of the little red fire ant, Was-
mannia auropunctata, in a Nino year, pp. 473-493 In
G. Robinson and E. M. del Pino [eds.], El Niio en las
Islas Galapagos: el Evento de 1982-1983. Charles
Darwin Foundation, Quito, Ecuador.

MCGLYNN. T. P. 1999. The worldwide transfer of ants:
geographical distribution and ecological invasions.
J. Biogeography 26: 535-548.
MEIER, R. E. 1994. Coexisting patterns and foraging be-
havior of introduced and native ants (Hymenoptera
Formicidae) in the Galapagos Island (Ecuador), pp.
44-62 In D. F. Williams [ed.], Exotic Ants: Biology,
Impact, and Control of Introduced Species. West-
view Studies in Insect Biology, Boulder, CO.
MIELES, A. 2002. Densidad y distribuci6n de invertebra-
dos terrestres en la Isla Marchena; antes, durante y
despues se erradica la hormiga colorada (Wasman-
nia auropunctata). Undergraduate thesis. Facultad
de Ciencias Sociales y de la Educaci6n (Escuela de
Quimica y Biologia), Universidad T6cnica de
Manabi, Ecuador.
PASSERA, L. 1994. Characteristics of tramp species, pp.
23-43 In D. F. Williams [ed.], Exotic Ants: Biology,
Impact, and Control of Introduced Species, West-
view Studies in Insect Biology, Boulder, CO.
and the introduced insects in the Galapagos Islands:
Different dispersal strategies, similar effects. Notic.
Galapagos 60: 30-36.
2000. The ants of Marchena Island, twelve years af-
ter the introduction of the little fire ant Wasmannia
auropunctata. Notic. Galapagos 61: 17-20.
SILBERGLIED, R. 1972. The little fire ant, Wasmannia
auropunctata, a serious pest in the Galapagos Is-
lands. Notic. Galapagos 19: 13-15.
SPENCER, H. 1941. The small fire ant Wasmannia in cit-
rus groves-a preliminary report. Florida Entomol.
24: 6-14.
SUAREZ, A. V., D. A. HOLWAY, AND T. J. CASE. 2001. Pat-
terns of spread in biological invasions dominated by
long-distance jump dispersal: Insights from Argen-
tine Ants. Proc. Nat'l. Acad. Sci., USA 98: 1095-1100.
Seasonal and nocturnal periodicities in ant nuptial
flights in the tropics (Hymenoptera: Formicidae). So-
ciobiology 37: 601-625.
TSUTSUI, N. D., AND A. V. SUAREZ. 2003. The colony
structure and population biology of invasive ants.
Conservation Biology 17: 48-58.
ULLOA-CHACON, P. 1990. Biologie de la reproduction
chez la petite fourmi de feu Wasmannia auropunc-
tata (Roger) (Hymenoptera, Formicidae). Ph.D. the-
sis, Universit6 de Lausanne, Facult6 des Sciences.
ULLOA-CHACON, P., AND D. CHERIX. 1990. The little fire
ant Wasmannia auropunctata R. (Hymenoptera:
Formicidae), pp. 281-289 In R. K. Vander Meer, K.
Jaffee, and A. Cedeio [eds.], Applied Myrmecology:
A World Perspective. Westview Press, Boulder, CO.
GREN. 1982. Degradation of the toxicant AC 217,300 in
Amdro imported fire ant bait under field conditions.
J. Agricultural & Food Chemistry Nov/Dec, 1045-1048.
WETTERER, J. K., AND S. D. PORTER 2003. The little fire
ant, Wasmannia auropunctata: distribution, impact,
and control. Sociobiology 42(1): 1-41.
WILLIAMS, D. F., AND P. WHELAN. 1992. Bait attraction
of the introduced pest ant, Wasmannia auropunctata
(Hymenoptera: Formicidae) in the Galapagos Is-
lands. J. Entomol. Sci. 27(1): 29-34.
ZUNIGA, T. 1994. Erradicaci6n de animals introduci-
dos. Field trip report, Galapagos National Park Ser-
vice, Santa Cruz Island, Galapagos.

June 2005

Fleming et al.: Butterfly Population Dynamics


'University of Miami, Department of Biology, Coral Gables, FL 33124, USA

2Present address: Department of Entomology and Nematology, University of Florida, Gainesville, FL 32611, USA

3Present address: Centro de Ecologia, Instituto Venezolano de Investigaciones Cientificas
Apartado Postal 21827, Caracas 1020-A, Venezuela


We studied the population dynamics of the zebra longing butterfly, Heliconius charithonia
(Nymphalidae), in a 0.05 ha garden in Miami, Florida, for 2 years to answer the following
questions: (1) How stable is a suburban, subtropical population of this widespread neotropi-
cal butterfly? (2) What are the major factors influencing its population dynamics? (3) What
are the implications of adult mobility regarding gene flow within and between fragmented
urban populations of this species? A mark-recapture study indicated that adult population
size averaged 59 individuals (range: 9-115 individuals). Peak numbers occurred in mid-wet
season (September) in 1996 and in the late dry and early wet seasons (April through June)
in 1997. Fluctuations in size of the adult population paralleled changes in biomass of the lar-
val food plant, Passiflora incense. Population sex ratio was consistently male-biased (68%
males). Reproduction occurred year-round, and parasitism by a trichogrammatid wasp
killed about 50% of Heliconius eggs throughout the year. Recruitment of adults from chrys-
alises produced in the garden and deaths, rather than immigration and emigration, ac-
counted for most numerical changes. Both males and females apparently adjust their home
range locations in response to changes in the biomass of Passiflora plants. Females search
these plants for suitable oviposition sites, and males search them for female pupae (mates).
In addition to relatively high mortality from egg parasitism, fluctuations in the number of
suitable oviposition sites and amount of larval food limited recruitment into the adult but-
terfly population. High adult mobility probably results in substantial gene flow within and
between populations of this butterfly in urban south Florida.

Key Words: population dynamics, food limitation, egg parasitism, Heliconius charithonia


Estudiamos la dinamica poblacional de la mariposa "cebra de alas largas", Heliconius chari-
thonia (Nymphalidae), en un jardin de 0,05 ha en Miami, Florida, por dos aios para re-
sponder a las siguientes preguntas: (1) Qu6 tan stable es una poblaci6n suburban
subtropical de esta mariposa neotropical ampliamente distribuida? (2) Cudles son los princi-
pales factors que influencian su dinamica poblacional? (3) Cudles son las implicaciones del
movimiento de adults en el flujo de genes dentro de y entire poblaciones urbanas fragmen-
tadas? Un studio de marca-recaptura indic6 que el tamaio de la poblaci6n adulta promedi6
59 individuos (rango: 9-115 individuos), con picos en la mitad de la estaci6n lluviosa (Septi-
embre) de 1996 y al final de la estaci6n seca e inicio de la estaci6n lluviosa (Abril a Junio) de
1997. Las fluctuaciones en el tamaio de la poblaci6n adulta fueron paralelas a los cambios
en biomasa de la plant hu6sped del estadio larval, Passiflora incense. La proporci6n de sexos
en la poblaci6n fue consistentemente sesgada hacia los machos (68% machos). La reproduc-
ci6n ocurri6 durante todo el aio, pero una especie de avispa tricogrammatidae parasita mat6
alrededor del 50% de los huevos de H. charithonia producidos en el aio. Mas que la inmi-
graci6n y emigraci6n, el reclutamiento de adults, a partir de crisalidas producidas en el
jardin, y la mortalidad controlaron el tamaio de la poblaci6n. Tanto machos como hembras
aparentemente modifican sus ambitos hogareios en respuesta a cambios en la biomasa de
las plants Passiflora. Las hembras visitan estas plants en busqueda de sitios 6ptimos para
la oviposici6n, mientras que los machos lo hacen para buscar pupas hembras (parejas para
reproducci6n). Ademas de la relative alta mortalidad causada por el parasitismo de los hue-
vos, las fluctuaciones en el nmmero de sitios 6ptimos para la oviposici6n y la cantidad de co-
mida para las larvas limitaron el reclutamiento hacia la poblaci6n adulta de mariposas. La
alta movilidad de los adults probablemente result en un flujo sustancial de genes dentro
de y entire las poblaciones de esta mariposa en el sur de la Florida.

Translation provided by the authors.

Florida Entomologist 88(2)

Butterflies of the genus Heliconius are com-
mon and conspicuous members of forest habitats
throughout the neotropics. Studies of their popu-
lation dynamics indicate that Heliconius butterfly
populations (i) occur in low population densities
(<5 per ha), (ii) have biased sex ratios (usually to-
wards males), (iii) reproduce year-round, (iv) have
maximum adult lifespans ranging from 90 to 180
days, and (v) contain individuals that exhibit
home range behavior (Ehrlich & Gilbert 1973;
Cook et al. 1976; Saalfeld & Araujo 1981;
Quintero 1988; Gilbert 1991; Ramos & Freitas
1999). Compared with most temperate butterflies,
species of Heliconius are low density, long-lived
insects with extended reproductive lifespans. The
degree to which Heliconius populations are genet-
ically open or closed has been controversial. Bra-
zilian workers (e.g., Romanowski et al. 1985;
Haag et al. 1993; Silva & Araujo 1994) have re-
ported that populations of Heliconius erato are
'insular' and inbred whereas Mallet (1986) sug-
gested, based on mark-recapture data, that indi-
viduals of H. erato in Costa Rica disperse consid-
erable distances and are not likely to be inbred.
What are the structure and dynamics of Heli-
conius butterfly populations near the northern
geographic limits of the group? As has been re-
ported for some species of temperate butterflies
(e.g., Thomas et al. 1994; Shreeve et al. 1996; Van-
Strien et al. 1997), are subtropical populations
more variable in size than tropical populations?
Are mortality rates higher and reproductive sea-
sons shorter in northern populations? To answer
these questions, we studied the dynamics of a
population of the zebra longing butterfly, Heli-
conius charithonia L. (Nymphalidae), for two
years in Miami-Dade County, Florida. H. chari-
thonia is widely distributed in the southeastern
United States, the West Indies, and Central and
northern South America (Opler & Krizek 1984).
According to Gilbert (1991), this species is more
common in disturbed than in undisturbed sites
and prefers seasonal tropical sites to evergreen
forests. H. charithonia is a phylogenetically ad-
vanced member of its genus the larvae of which
are feeding specialists on non-woody, short-lived
species of Passiflora (passion vines) found in suc-
cessional habitats. Adults tend to collect smaller,
less nutritious pollen grains than less-advanced
heliconiines. Pupal mating occurs in this species,
and females lay their eggs gregariously on fresh
shoots. Larvae are non-aggressive and feed on
older leaves only after fresh shoots are depleted.


Study Site

This study was conducted for two years (mid-
December 1995 to mid-December 1997) in a 0.05-
ha. garden in suburban Miami, Florida (2549'N,

8017'W). Most observations were made in a 15-m
x 20-m section of the garden, which was planted
in 'butterfly' plants, including Pentas lanceolata,
s'r. .. ., r.,, .i..r.i, fruiticosa, and Hamelia patens,
that provided nectar and pollen for adults. One
individual of Passiflora incense, a non-native (hy-
brid) larval food plant (Vanderplank 2000), was
planted at the base of a clothes pole that served as
an arbor in January 1995. By late 1995 rhizomes
of this plant had spread over a relatively large
portion of the 300 m2 intensive study area, and
new stems emerged continuously throughout the
study. Prior to 1 August 1997, we did little to con-
trol the growth of Passiflora in the garden. By
that date, however, vines had overgrown much of
the study area, and we removed >95% of the Pas-
siflora biomass. A native Passiflora (P suberosa)
occasionally 'volunteered' in the garden but was
quickly eliminated by Heliconius herbivory.
The climate of Miami includes a 5-month dry
season (November-April) and a 7-month wet sea-
son (May-October). About 75% of Miami's average
annual rainfall of 1,420 mm falls during the wet
season. Annual rainfall was above average in 1996
and 1997 and totaled 1,466 and 1,795 mm, respec-
tively. During the study, lowest temperatures of
2-3C occurred on one day each in January and
February 1996 and January 1997, and highest
temperatures of 34C occurred in June 1996 and
July and August 1997. In this study, we recognize
two seasons per year: dry season (1 November-30
April) and wet season (1 May-31 October).
In addition to H. charithonia, two other species
of heliconiid butterflies frequented the garden.
The gulf fritillary, Agraulis vanillae (L.), was
present in low numbers (but sometimes in equal
abundance to that of the zebra longing) during
most of the study. Its larvae co-occurred with
those of H. charithonia on Passiflora vines. Much
less common than the gulf fritillary was the julia,
Dryas iulia Clench, which apparently did not ovi-
posit in the study area.


Size of the adult population feeding, oviposit-
ing, and mating in the garden was estimated by
mark-recapture methods. One morning each
week, butterflies were captured with hand nets
and released during a period of 1.5-2 h. Captured
individuals were marked by writing a number on
the outside surface of both hindwings with a
Sharpie Extra Fine Point permanent marker. At
an individual's initial capture, we noted its sex,
wing length, and wing wear condition. On subse-
quent captures, we recorded ID number and wing
wear condition. We measured wing length to the
nearest 0.5 mm with a plastic ruler as the dis-
tance from the bottom of the left hindwing to the
tip of the left forewing. We scored wing wear con-
dition in one of five categories: 0 = no wear (bright,

June 2005

Fleming et al.: Butterfly Population Dynamics

intense color), 1 = early wear (color less intense
but still bright), 2 = medium wear (color faded but
wings still opaque), 3 = extensive wear (color very
faded and wings partially transparent), and 4 =
extreme wear (little color, wings extensively dam-
aged). Butterflies were retained for ca. 5 min at
each capture. At approximately weekly intervals
between 4 July 1996 and 24 March 1997, we
counted the number of adults night-roosting on a
Citrus tree in the garden after sunset.
We surveyed 2-10 major Passiflora stems, de-
pending on availability, for heliconiid eggs, lar-
vae, and chrysalises once each week Throughout
the study the "clothes pole" plant was the largest
Passiflora stem and was always surveyed for im-
mature life stages. Other stems >1 m in length
were surveyed whenever they contained fresh
growth. We could not distinguish between the
eggs of H. charithonia and A. vanillae, so our egg
counts include both species. Number of larvae in
each of three size categories (<1/3 maximum
length (= "small"), >1/3 but <1/2 maximum length
(= "medium"), and >1/2 maximum length (=
"large") was recorded separately for both helico-
niid species. Finally, we searched each plant for
zebra longwing chrysalises and counted and
marked each one with a small spot of fingernail
polish to distinguish "new" from "old" chrysalises.
We estimated egg mortality due to parasitism
by trichogrammatid wasps between 6 October
1996 and 21 November 1997 in two ways. First,
every two weeks between 6 October 1996 and 10
May 1997 we removed 15 randomly chosen eggs
from new leaves or stems and placed each one in
a labeled 1.5 mL microcapillary tube. Eggs were
examined daily and the fate of each egg (no larvae
or parasites emerged, larva emerged, parasitic
wasps emerged) was recorded for two weeks. Be-
cause all larvae emerging from eggs were those of
H. charithonia, these mortality estimates are not
confounded by the presence ofA. vanillae. Second,
six times in 1997 (twice in April and May, once in
July and November) a series of 4-109 eggs on 1-14
plants was marked with a dot of ink on the leaf
next to each egg. For the next 2-9 days the status
of each egg was scored as parasitizedd" (egg
turned dark gray), hatched (egg shell empty), or
"gone." Number of small larvae ofH. charithonia
and condition of the leaf bearing the eggs (intact
or chewed) also were recorded every day.
To estimate the amount of Passiflora biomass
present in the study area each week, we took notes
on the condition of each stem (i.e., each ramet)
that we surveyed. Each stem was given one of the
following "condition" scores: 0 = chewed back to
ground level; 1 = new or regrowth <1 m long; 2 = a
medium-sized, non-flowering plant; 3 = a large,
flowering plant; and 4 = a very large, flowering
plant. These scores represent the approximate
greatest length (in m) of each stem. The sum of the
scores of stems was multiplied by 2 to represent

the approximate biomass of Passiflora foliage (in
m2) present in the intensive study area each week.

Population Estimation

Capture-recapture data were analyzed in a set
of models implemented in program TMSURVIV;
J. Hines, unpublished), a modification of program
SURVIV (White 1983) developed to compute esti-
mates under the transient models of Pradel et al.
(1997). Transient models represent a generaliza-
tion of the standard Cormack-Jolly-Seber model
(Cormack 1964; Jolly 1965; Seber 1965) devel-
oped for open populations (i.e., populations that
can experience gains and losses between sam-
pling periods). The generalization basically in-
volves the possibility of different survival proba-
bilities for animals captured in any sampling pe-
riod, i, depending on whether or not the animal
has been captured previously (i.e., whether the
animal is marked or unmarked).
The most general model contains three kinds
of parameters indexed by both time (sampling pe-
riod, i) and sex (s):
q. = the probability that a marked butterfly of
sex s in the population at time i survives and
is present in the area exposed to sampling
efforts in period i+1;
7, = the ratio of survival probabilities (i to i+1) of
unmarked to marked butterflies of sex s (in
a population with transient individuals, this
ratio is the probability that an unmarked
butterfly is a transient);
p" = the probability that a butterfly of sex s, in
the population at time i, is captured at i.
The zebra longwing butterfly data set con-
tained 97 sampling occasions, and the most gen-
eral model thus contained a very large number of
parameters. Several reduced-parameter models
were created by eliminating one or more sources
of variation in model parameters. For example,
models with = 1 incorporated the assumption
of the standard Cormack-Jolly-Seber model that
survival probability does not depend on mark sta-
tus. Some models incorporated the assumption of
parameters constant over time (e.g.,p," = p"'), and
others assumed equality of parameters for males
(m) and females (f) (e.g., p,' = p," = p,). Although
most sampling was conducted on a weekly basis,
this was not strictly true, and the time intervals
separating successive sampling periods varied.
For this reason, the time constraint, (pI" = (pI
makes little biological sense. However, the hy-
pothesis of equal survival per unit time is reason-
able, so we reparameterized survival in the model
as (p: = (S, ), where t denoted the time period (in
weeks) separating sampling periods i and i+1.
Thus, we effectively scaled survival probability to
a weekly time interval, and the hypothesis of

Florida Entomologist 88(2)

equal weekly survival over time, S," = S", is bio-
logically plausible.
We fit 14 different models to the data and fol-
lowed the general suggestions of Lebreton et al.
(1992) for model notation. Model (S, p, p ,, the
general model of Pradel et al. (1997), was the most
general model in our model set. All three classes
of parameter include the subscript i, denoting full
time variation, and the superscript s, denoting
sex-specificity of parameters. Absence of a super-
script or subscript indicates that the source of
variation is omitted from the model parameter-
ization. For example, model (S", p", y) includes no
time-specificity for any parameter (denoted by ab-
sence of subscripts, i), and no sex-specificity either
for y. We considered models that assumed equal
survival probabilities for previously marked and
unmarked butterflies (the Cormack-Jolly-Seber
model assumption), and denoted this assumption
as y = 1. Thus, model (S, p, y = 1) denotes a simple
2-parameter model in which survival and capture
probability are constant over time and the same
for males and females and survival is the same for
marked and unmarked butterflies.
The software, TMSURVIV, provides maximum
likelihood estimates of the parameters and associ-
ated estimates of variances and covariances un-
der each model. Adequacy of fit of the most gen-
eral model was judged using a parametric boot-
strap approach. Under this approach, capture his-
tory data were simulated under the general model
by treating the parameter estimates as true val-
ues. The general model was then fit to each simu-
lated data set, and maximum likelihood estimates
obtained. The standard G2 goodness-of-fit statistic
was also computed for each simulated data set.
We ran 100 simulations and compared the G2 sta-
tistic from the actual data analysis with the dis-
tribution of statistics from the simulated data sets
to test for fit. In the event of poor model fit, we fol-
lowed the approach recommended by White &
Burnham (1999) of estimating a variance infla-
tion factor, c, as the ratio of the observed G2 to the
mean of the G2 values from the simulations. In the
event of poor model fit, variance estimates were
obtained using a quasi-likelihood approach (e.g.,
Burnham et al. 1987; Lebreton et al. 1992) as the
product 5 ua( (0), where 6 is the maximum likeli-
hood estimate of parameter 0 under the selected
model, var (6) is the associated model-based vari-
ance estimate, and 5 is the variance inflation fac-
tor associated with lack of model fit and estimated
as described above.
Model selection was based on QAICc values,
Akaike's Information Criterion adjusted for lack
of fit (quasi-likelihood) and sample size (Burn-
ham & Anderson 1998). AIC can be viewed as an
optimization criterion useful in model selection
(Akaike 1973; Burnham & Anderson 1998). The
criterion places value on good fit of the model to
the data and on describing the data with as few

parameters as possible (Burnham & Anderson
1992, 1998; Lebreton et al. 1992).
Population size was not a model parameter but
was estimated using the numbers of butterflies
caught at each period (n,' = number of males; n,"
= number of females; n,' + nlf = n, = number of
butterflies of both sexes combined) in conjunction
with the associated estimates of capture probabil-
ity (p,'). Specifically, population size and its asso-
ciated variance were estimated as:

s n(s)
Nz =- (1)

) (n,())2 var (ps)) ()(1 -p(s)
var((s) 4 (s)
(P ) ) (P ) )

Confidence intervals for NW' were approxi-
mated using the approach of Chao (1989; also used
and recommended by Rexstad & Burnham
(1991)). The estimation is based on the estimated
number of butterflies not detected, A' = N"( n ".
The In ( ,) is treated as an approximately normal
random variable, yielding the following 95% confi-
dence interval, (n," + ../C, n,. + C), where

var )
C = exp 1.96 n 1+ I( .
r. (s) 2
,1) )j2

Statistical Analyses

We used one- and two-way ANOVAs to test for
the effects of sex and season on individual and
population variables and Pearson correlation
analyses to assess relationships between rainfall
and butterfly population variables and Passiflora
biomass. Analyses were conducted with Statmost
ver. 3.5 (Dataxiom Software, Inc., Los Angeles,
CA) and Systat ver. 10 (SPSS, Inc., Chicago, IL).
For population and weekly count data, we tested
for effects of season using repeated measures (rm)
ANOVAs with Type III sums of squares (Zar, 1999)
using SPSS ver. 10.1.0 (SPSS 2000, SPSS, Inc.,
Chicago, IL). We used the Greenhouse-Geisser
correction to adjust degrees of freedom whenever
inequality of sphericity was rejected by Mauchly's
test. We used the "Standard Tests" module in Eco-
Sim (Gotelli & Entsminger 2003) to assess the re-
lationship between weekly adult population size
and Passiflora biomass. Unless otherwise noted,
means + 1 SE are reported throughout this paper.


Capture Statistics

In the 2 yrs, we marked a total of 1,476 adults,
including 929 males (62.9%) and 547 females, and

June 2005

Fleming et al.: Butterfly Population Dynamics

recorded a total of 2,729 captures and recaptures.
Capture statistics, including the number of cap-
tures per individual and the number of weeks be-
tween first and last capture for each individual by
sex and season, are summarized in Fig. 1A,B. A
two-way ANOVA indicated that number of cap-
tures per individual varied by sex and season (sex:
F1,,14 = 7.43, P = 0.007; season: F4,,,14 = 23.55, P <
0.0001; the interaction was also significant: F4,1470 =
4.79, P = 0.001). Males were generally captured
more times than females (grand means were 2.08 +
0.06 and 1.82 0.08 captures, respectively), and
number of captures per individual in the 1996-97
dry season was about 50% higher than in other sea-
sons (Fig. 1A). A two-way ANOVA revealed that the
number of weeks between first and last capture
also differed by sex and season (sex: F11471 = 8.64, P
= 0.003; season: F4,1471 = 20.91, P < 0.0001; the inter-
action was not significant (P = 0.23)). Males were
generally captured over a longer time period than
females (grand means were 2.05 0.05 and 1.81 +
0.07 weeks, respectively), and individuals had
longer capture periods in the first three seasons of
this study than in the last two (Fig. 1B). Reanalysis
of these data after data from the short dry season in
late 1997 were eliminated produced similar results.

The Population Model

The G2 goodness-of-fit statistic for general
model (SI,p ,',, was 692.83 and was larger than
all 100 of the G2s resulting from the bootstrap sim-
ulations. Because of the lack of fit of the general
model to the capture-recapture data, we computed
the quasi-likelihood variance inflation factor as
described above to obtain a = 1.21. Models with the
lowest QAICc values were those with parameters
constant over time and with no difference between
survival of unmarked and marked butterflies (y =
1) (Table 1). There was very little basis for select-
ing among the first few models, as indicated by the
small AQAICcs. Daily survival estimates based on
model (S, p", y = 1) were 0.944 0.0021 for males
and 0.939 0.0032 for females. Estimated capture
probabilities under this model were 0.479 0.0163
for males and 0.447 0.0231 for females. Parame-
ter estimates for males and females were very sim-
ilar, providing little evidence for sex-specific differ-
ences in capture probabilities. In fact, the very
simplest model (S, p, y = 1) with single survival
and capture parameters that were constant over
time and sex was among the most reasonable mod-
els for this data set. From this, we conclude that
survival rates and capture probabilities likely did
not vary with season or sex.

Adult Population Size and Sexual and Body Size

Estimates of the number of adults foraging in
or passing through the garden each week aver-



0 ------- -----------------------------
S Dry9596 Wet96 Dry9697 Wet97 Dry97
3 B


Dry9596 Wet96 Dry9697 Wet97 Dry97

48 C
47 -
46 -
45 -
41 -
Dry9596 Wet96 Dry9697 Wet97 Dry97
S eason
Fig. 1. Summary of the capture statistics for adults
of Heliconius charithonia. A. Number of captures and
recaptures per individual. B. Number of weeks between
first and last capture records. C. Size of adults. Data are
means 1 SE. Sample sizes range from 29-330.

aged 58.9 2.9 and ranged from a minimum of
9 (in August 1997) to a maximum of about 115
(in April 1997) (Fig. 2). In 1996 the population
ranged between 20 and 70 individuals except for
peaks of 85-100 individuals in the middle of the
wet season (September) and early in the 1996-97
dry season (December). In 1997, the population
steadily increased during the dry season and av-
eraged over 80 individuals through July before
declining rapidly. We removed most of the Passi-
flora in the garden after the decline began, and
numbers slowly increased after 1 August 1997.
Size of the adult population fluctuated in par-
allel with changes in the biomass of Passiflora in
the study area (Fig. 2). A simulation comprising
1,000 iterations indicated that the correlation be-
tween adult population size (Y) and Passiflora
biomass (X) was significant (rob.oled = 0.69; ru a e =
0.00, P = 0.00). One-way rm ANOVAs indicated
that seasonal differences in weekly size of the
butterfly population and Passiflora biomass were
significant (adult population size: Greenhouse-

Florida Entomologist 88(2)


Model' No. parameters estimated Log-likelihood AQAICc

S, p, = 1 3 -1366.9 0.0
S",p, y = 1 4 -1366.3 0.5
S,p", y=1 3 -1367.5 0.9
S,p, = 1 2 -1369.1 1.4
S,p,y 3 -1368.6 3.4
S",( ", ) 6 -1365.8 4.5
S,p,, = 1 97 -1258.7 17.1
S", p(s),y = 1 194 -1177.4 99.2

aAbbreviations: S = probability that a marked butterfly survives from sampling time i to i + 1, p = probability that a butterfly
that is in the population at time i is captured at i, and = the ratio of survival probabilities (from i to i + 1) of unmarked to marked
butterflies, s = sex.

Geisser-corrected F16,264 = 5.90, P = 0.012; Passi-
flora biomass: Greenhouse-Geisser-corrected
F5,286 = 13.54, P < 0.001; data from the abbrevi-
ated 1997 dry season not included in these analy-
ses). The adult butterfly population averaged 36%
larger in 1997 than in 1996, and Passiflora bio-

mass was 48% lower during the 1995-96 dry sea-
son than in the next three seasons (Fig. 3A).
Neither size of the adult butterfly population
nor Passiflora biomass appeared to respond to
rainfall seasonality. Correlations between weekly
estimates of both adult population size and pas-

0 10 20 30 40 50 60
Number of Weeks Since

70 80 90 100 110
15 Dec 1995

Fig. 2. Weekly estimates of the adult population size (+ 1 SE) of Heliconius charithonia and the biomass (in m2)
of the larval host plant Passiflora incense from 15 December 1995 to 15 December 1997. The upper rectangles in-
dicate dry seasons (open) and wet seasons (hatched).

June 2005

Fleming et al.: Butterfly Population Dynamics

A] (sex: F,1... = 12.50, P < 0.001; season: F4 15 =
A 24.49, P < 0.001). Except in the wet season of
1997, females were larger than males, and butter-
Sflies of both sexes were about 10% larger in late
Adults 1997 than in early 1996 (Fig. 1C).

Adult Survivorship and Wing Wear

Turnover rate of adults in this population was
- relatively high. Most adults were recaptured for
Biomass periods of less than three weeks (Fig. 1B), and
Maximum capture periods were 9-10 weeks in
Dry6596 Wet96 Dry9697 Wet97 Dry97 both sexes. Most adults (86%, n = 1,504) were re-
cently closed (condition score 0) at first capture;
individuals occurring in condition score catego-
0- B ries 0 or 1 accounted for 98% of all first captures.
In 1996, we tallied the wing wear condition of 57
0 and 30 recently closed males and females, re-
spectively, that were captured >3 times. Rate of
0- gs wing wear was high in both sexes. Most individu-
als had worn to very worn wings 6-8 weeks after
first capture.
o Results of the mark-recapture analysis also in-
Larvae dicated that adult turnover rate was high. Adult
0 males and females had daily survival probabili-
Chrysalises ties of about 0.94, which represents a weekly sur-
Svival probability of about 0.65. At this rate, adults
Nweeks 20 20 24 25 5 had a probability of 0.031 and 0.013 of being alive
season eight and ten weeks after eclosion, respectively.

Fig. 3. Seasonal summaries of (A) weekly estimates of
the adult butterfly population size and biomass of its lar-
val host plant, Passiflora incense, and (B) weekly num-
bers of butterfly eggs, larvae, and chrysalises. Data are
means + 1 SE. Sample sizes are indicated as N weeks.

sion vine biomass and rainfall were nonsignifi-
cant (Bonferroni-corrected Ps > 0.28). Similar re-
sults were obtained when rainfall records were
lagged by one and two weeks (Ps > 0.26).
Between 4 July 1996 and 24 March 1997, an
average of 20.3 2.6 (range = 0-46; n = 25 counts)
adults night-roosted in the garden. The number of
night-roosting adults was positively correlated
with adult population size (Pearson's r = 0.75, df
= 23, P < 0.001). At times, most of the adults for-
aging in the garden night-roosted there.
In most weeks the adult sex ratio was male-
biased and averaged 65.6 1.2% males overall (n
= 94 weeks). A one-way rm ANOVA indicated that
sex ratio varied among seasons (F3,5 = 2.95, P =
0.04; arcsine squareroot-transformed data and
the short 1997 dry season not included). Sex ratio
was higher in the dry season of 1996-97 (71.7 +
2.1% males) than in the wet season of 1997 (60.3
+ 2.7% males). Thirty of 72 adults (41.7%) that we
sexed at eclosure in the garden were males. Sex
ratios of marked and closed adults differed sig-
nificantly (X2 = 12.3, df = 1, P < 0.001).
A two-way ANOVA indicated that the size
(wing length) of adults varied by sex and season

Egg and Larval Surveys

Eggs and larvae were found in most weeks, in-
dicating that reproduction occurs year-round in
south Florida (Fig. 4). The number of eggs
counted each week averaged 55.3 5.6 (range: 0-
326) and exhibited no significant seasonal varia-
tion (one-way rm ANOVA: F,57 = 2.36, P = 0.08;
only the first four seasons included in this analy-
sis) (Fig. 3B). The number of zebra longwing but-
terfly larvae (of all sizes) counted per week aver-
aged 14.3 2.4 (range: 0-167) and also exhibited
no significant seasonal variation (Greenhouse-
Geisser-corrected one-way rm ANOVA: F,1,,, =
1.55, P = 0.23) (Fig. 3B). Finally, number of zebra
longwing butterfly chrysalises counted per week
averaged 5.0 0.78 (range: 0-38) and did not vary
seasonally (Greenhouse-Geisser-corrected one-
way rm ANOVA: F,, ,3 = 1.90, P = 0.17) (Fig. 3B).
Only the number of eggs per week was signifi-
cantly correlated with estimated adult population
size (Pearson's r = 0.32, df = 90, Bonferroni-cor-
rected P = 0.037). Total number ofH. charithonia
larvae per week but not chrysalises was signifi-
cantly correlated with number of eggs (Pearson's
r = 0.36, df = 92, Bonferroni-corrected P = 0.011).
When we applied a 3-week time lag (the average
egg-to-chrysalis duration) to the egg data, how-
ever, number of chrysalises was significantly cor-
related with number of eggs (Pearson's r = 0.25, df
= 89, Bonferroni-corrected P = 0.018). Finally,

Florida Entomologist 88(2)

June 2005


200 3




0 10 20 30 40 50 60 70 80
Weeks Since 15 Dec 1995

90 100 110


200 3

150 R

100 W


Weeks Since 15 Dec 1995
Fig. 4. Weekly counts of eggs (A) and larvae (B) of Heliconius charithonia between 15 December 1995 and 15
December 1997. Estimated weekly adult population size is included in each panel.

weekly numbers of eggs, larvae, and chrysalises
were not correlated with weekly Passiflora bio-
mass (Bonferroni-corrected Ps > 0.18).

Two species of heliconiid larvae co-occurred on
the Passiflora vines between weeks 30 and 85
(from July 1996 to August 1997). During this pe-

Fleming et al.: Butterfly Population Dynamics

riod, number of zebra longwing butterfly larvae av-
eraged 17.0 3.9 per week (range: 0-167), and
number of gulf fritillary larvae averaged 10.7 1.9
(range: 0-55); these means do not differ signifi-
cantly (paired t-test, t = 1.56, df = 52, P = 0.12).
Numbers of these two larvae were not correlated
(Pearson's r = 0.17, df= 51,P = 0.21), and hence the
larvae did not appear to be interacting in an antag-
onistic (i.e., competitive or predator-prey) fashion.

Egg Parasitism

Egg survival was low (about 14% based on ra-
tios of eggs to small larvae over all seasons), and a
major reason for this was parasitism by an uniden-
tified trichogrammatid wasp. We determined per-
cent parasitism of zebra longwing butterfly eggs by
this wasp on 22 occasions between 6 October 1996
(week 43) and 25 November 1997 (week 101). Per-
cent parasitism averaged 53.0 5.0% (range: 0-
100%). Parasitized eggs produced an average of 6.6
+ 0.6 wasps (n = 34; range: 1-14). There appeared
to be no seasonal pattern to wasp parasitism.

Rates of Population Increase and Decrease

Changes in population size result from the in-
teraction of four factors: births, deaths, immigra-
tion, and emigration. To what extent are the pop-
ulation increases and decreases we documented
in this study (Fig. 2) the result of demographic
(i.e., births, deaths) rather than behavioral (i.e.,
immigration, emigration) factors? To answer this
question, we analyzed events occurring during
three increase episodes (weeks 29-33, 46-50, and
58-64; Fig. 2) and three decrease episodes (weeks
33-43, 50-53, and 81-92). For the increase epi-

sodes, we determined the relative contribution of
"births" (i.e., eclosion of new adults) to the overall
population increase by comparing the number of
chrysalises produced during each episode to the
estimated number of individuals added to the
population. To avoid re-counting the same chrys-
alises, we counted only "new" (= unmarked)
chrysalises that we found beginning and ending
one week before the increase period began and
ended. For the decrease episodes, we compared
the number of individuals that were "lost" from
the population with the number of expected losses
based on a weekly survival rate of 0.65.
Results of these calculations (Table 2) indi-
cated that local births and deaths likely ac-
counted for most of the changes in population size
during these episodes. Eclosion of new adults ac-
counted for about 90% of the population in-
creases, and expected adult deaths accounted for
virtually 100% of the population decreases. Abso-
lute values of increase (=13.5 adults/wk) and de-
crease (=10.5 adults/wk) were similar during
these six episodes (Fig. 2).


Our results indicate that the size of a suburban,
subtropical population of H. charithonia ranged
from about 10-115 adults during a two-year period.
Weekly variation in adult population size was mod-
erate (coefficient of variation = 0.47), and changes
in the size of this population closely mirrored
changes in the biomass of the larval host plantPas-
siflora incense. Neither adult population size nor
passion vine biomass appeared to be influenced by
weekly rainfall. Reproduction occurred year-round,
and although the sex ratio at eclosion was female-


A. Population increases

Weeks (n) An n Chrysalises n Chrysal./An

29-33(4) +66.2 57 0.86
46-50 (4) +59.7 96 1.61= 1.00
58-64(6) +53.3 39 0.73
Mean =0.90

B. Population decreases

Weeks (n) An Expected An" Expected/observed

33-43(10) -87.5 -85.5 0.98
50-53(3) -40.5 -60.4 1.49
81-92(11) -102.4 -107.8 1.02
Mean =1.00

"Calculated as n, (n1*0.65"), where n, is the number of individuals in week 1 of the episode, 0.65 is the weekly adult survival
rate, and n is number of weeks in the episode.

Florida Entomologist 88(2)

biased, sex ratio of the adult population was
strongly male-biased. Rates of wing wear in adults
were high, and maximum adult longevity was 9-10
weeks. Given that the egg-to-adult period lasts 30
days in this species in the subtropics (Quintero
1988), maximum lifespan of H. charithonia in
south Florida is 13-14 weeks (91-98 days). Most
adults, however, live less than one month, so aver-
age lifespan is less than eight weeks (<56 days).
The best predictor of adult population size in
this study was biomass of the larval host plant,
Passiflora incense. It appeared that both males
and females were tracking biomass of this plant,
but for different reasons. Females track passion
vine biomass in searching for oviposition sites,
whereas males track this biomass looking for fe-
male chrysalises (mates). Of the two sexes, males
appeared to be more sedentary, as indicated by
their higher number of recaptures and the longer
time between first and last capture. Many males
virtually 'camped out' in the garden waiting for
female chrysalises to become sexually receptive.
Receptive females were quickly 'swarmed' by sev-
eral males, one of which ultimately mated with
her (THF, pers. obs.). Females, in contrast, ap-
peared to be "trap-lining" and passed through our
study site en route to other oviposition or feeding
sites. While in the garden, they laid one or more
eggs on new growth on several stems.
The shift from a female-biased sex ratio at
eclosion to a male-biased sex ratio in the local
population suggests that adult females are more
mobile than males. As in most birds (Greenwood
1980), females appear to be the dispersing sex in
H. charithonia. Many males apparently stay near
their natal sites to search for mates whereas fe-
males disperse some distance away from their na-
tal sites before establishing a home range. Be-
cause of high female mobility, however, males are
not likely to mate with close relatives, and rates
of inbreeding are likely to be low. This mobility
also likely results in substantial gene flow among
zebra longwing butterfly subpopulations.
In support of these predictions, Kronforst &
Fleming (2001) reported very low levels of in-
breeding (Wright's fixation index FN = -0.027, a
slight excess of heterozygotes) and low population
genetic subdivision (Wright's fixation index F. =
0.003) in H. charithonia over a wide area in Mi-
ami-Dade County. These results call into question
the suggestion (e.g., Haag et al. 1993) that Helico-
nius populations have an island-like structure
and are highly inbred.
Although the dynamics of all populations are
influenced by abiotic and biotic factors, biotic fac-
tors are likely to most strongly influence the pop-
ulation dynamics ofH. charithonia in south Flor-
ida. This conclusion is based on the absence of a
correlation between rainfall and adult population
size and Passiflora biomass as well as the absence
of a strong seasonal effect on numbers of eggs, lar-

vae, chrysalises, and rates of egg parasitism. But-
terfly reproduction, egg parasitism, and Passi-
flora growth occurred year-round and were not
strongly seasonal. H. charithonia in south Florida
is clearly behaving like a tropical butterfly.
Two biotic factors, host plant biomass, espe-
cially the availability of growing shoot tips, and
egg parasitism, appeared to affect the dynamics of
this population. Although numbers of eggs and
larvae showed no strong seasonal trends, their
numbers were highly variable from week to week.
Coefficients of variation of weekly number of eggs,
total larvae, and chrysalises, respectively, were
0.99, 1.64, and 1.52 compared to a value of 0.47 for
number of adults. Variation in egg number, in
part, reflected variation in the availability of fresh
growing tips which was determined, in turn, by
caterpillar herbivory. At times, herbivory elimi-
nated the growing tips from most stems, which re-
sulted in few eggs being laid. Egg parasitism by
trichogrammatid wasps also eliminated an aver-
age of 50% of potential Heliconius larvae. Parasit-
ism was especially high during weeks 61-69 (early
February-mid-April 1997) when it averaged 75%.
During that period, few larvae were produced,
and Passiflora biomass began to steadily increase.
Despite low larval numbers during this period,
adult butterfly numbers were high. This was the
only time during the study that immigration,
rather than in situ recruitment, appeared to be
responsible for high adult population numbers.
Many aspects of the population biology of H.
charithonia in south Florida are similar to those of
tropical Heliconius populations. Similarities in-
clude low, relatively stable population densities, bi-
ased adult sex ratios, year-round reproduction,
and the importance of larval host plants in deter-
mining the distributions and densities of adult
butterflies (e.g., Ehrlich & Gilbert 1973; Cook et al.
1976; Quintero 1988; Gilbert 1991; Ramos & Frei-
tas 1999). Coefficients of variation of population
size of H. charithonia in Costa Rica and Puerto
Rico ranged from 0.34 to 1.05 (based on data in
Cook et al. 1976; Quintero 1988) compared with
0.47 in this study; CV of mean population size in
Heliconius erato in southern Brazil was 0.73 (Ra-
mos & Freitas 1999). As in this study, Quintero
(1988) found no correlation between population
size and rainfall in Puerto Rico. Under normal con-
ditions, the population density of H. charithonia in
urban Miami is low, on the order of a few individu-
als per hectare in forested areas (Kronforst &
Fleming 2001). This low density undoubtedly re-
flects the low density and biomass of Passiflora in
these areas. Average lifespan was also similar in
populations ofH. charithonia in south Florida, Pu-
erto Rico, and Costa Rica. Compared with H.
ethilla the maximum lifespan of which in Trinidad
is about 180 days and H. erato in southern Brazil
with a maximum lifespan of about 150 days, H.
charithonia is a short-lived butterfly. Maximum

June 2005

Fleming et al.: Butterfly Population Dynamics

longevities are 90-116 days, and average adult
lifespans are less than one month after eclosion.
In conclusion, populations trends in H. chari-
thonia in south Florida do not appear to differ
qualitatively or quantitatively from those at lower
latitudes. The mild climate of subtropical Miami
permits this butterfly to act as though it is still in
the tropics. These results contrast with those re-
ported for butterflies in England, where marginal
populations of some species fluctuate much more
strongly than central populations (Thomas et al.
1994; Shreeve et al. 1996). Studies of the popula-
tion dynamics ofH. charithonia closer to its north-
ern geographic limits (e.g., in northern Florida)
are needed to determine whether population sizes
are more variable and more strongly influenced
by abiotic factors than in south Florida. Such
studies in H. erato in subtropical Brazil indicate
that climate-related (either cold temperatures or
drought) extinctions can occur in Heliconius but-
terflies at the southern limits of their distribution
(Saalfed & Araujo 1981).


We thank M. Fleming, M. Garg, and M. Requina for
assistance in data collection and J. D. Nichols and J.
Hines for implementation of the mark-recapture analy-
ses. J. Mallet provided useful comments about a previ-
ous draft. H. Romero provided the Resumen.


AKAIKE, H. 1973. Information theory and an extension
of the maximum likelihood principle, pp. 267-281 In
B. N. Petran and F. Csaaki [eds.], International Sym-
posium on Information Theory. Akadeemiai Kiadi,
Budapest, Hungary.
BURNHAM, K. P., AND D. R. ANDERSON. 1992. Data-
based selection of an appropriate biological model: the
key to modern data analysis, pp. 16-30 In D. R. Mc-
Cullough and R. H. Barrett [eds.], Wildlife 2001: Pop-
ulations. Elsevier Applied Science, New York, NY.
BURNHAM, K. P., AND D. R. ANDERSON. 1998. Model Se-
lection and Inference: A Practical Information-theo-
retic Approach. Springer-Verlag, New York, NY.
BROWNIE, AND K. P. POLLOCK. 1987. Design and
analysis of methods for fish survival experiments
based on release-recapture. American Fish. Soc.
Monograph 5: 1-437.
CHAO, A. 1989. Estimating population size for sparse
data in capture-recapture experiments. Biometrics
45: 427-438.
Population structure, dynamics and dispersal of the
tropical butterfly Heliconius charitonius. J. Animal
Ecol. 45: 851-863.
CORMACK, R. M. 1964. Estimates of survival from the
sighting of marked animals. Biometrika: 429-438.
EHRLICH, P. R., AND L. E. GILBERT. 1973. Population
structure and dynamics of the tropical butterfly Hel-
iconius ethilla. Biotropica 5: 69-82.
GILBERT, L. E. 1991. Biodiversity of a Central American
Heliconius community: pattern, process, and problems,

pp. 403-427 In P. W. Price, T. M. Lewinsohn, G. W.
Fernandes and W W. Benson [eds.], Plant-animal In-
teractions: Evolutionary Ecology in Tropical and Tem-
perate Regions. John Wiley and Sons, New York, NY.
Null models software for ecology. Version 7. Acquired
Intelligence, Inc. & Kesey-Bear. Burlington, VT 05465.
GREENWOOD, P. J. 1980. Mating systems, philopatry,
and dispersal in birds and mammals. Anim. Behav.
28: 1140-1162.
HAAG, K. L., A. M. ARAUJO, AND A. ZAHA. 1993. Genetic
structure of natural populations of Dryas iulia (Lep-
idoptera: Nymphalidae) revealed by enzyme poly-
morphism (RFLP). Biochem. Genet. 31: 449-460.
JOLLY, G. M. 1965. Explicit estimates from capture-re-
capture data with both death and immigration-sto-
chastic model. Biometrika 52: 225-247.
KRONFORST, M. R., AND T. H. FLEMING. 2001. Lack of
genetic differentiation among widely spaced subpop-
ulations of a butterfly with home range behaviour.
Heredity 86: 243-250.
ANDERSON. 1992. Modelling survival and testing bio-
logical hypotheses using marked animals: a unified ap-
proach with case studies. Ecol. Monographs 62:67-118.
MALLET, J. 1986. Dispersal and gene flow in a butterfly
with home range behaviour: Heliconius erato (Lepi-
doptera: Nymphalidae). Oecologia 68: 210-217.
OPLER, P. A., AND G. O. KRIZEK. 1984. Butterflies East
of the Great Plains. The Johns Hopkins University
Press, Baltimore, MD.
NICHOLS. 1997. Capture-recapture survival models
taking account of transients. Biometrics 53: 60-72.
QUINTERO, H. E. 1988. Population dynamics of the but-
terfly Heliconius charitonius L. in Puerto Rico.
Carib. J. Sci. 24: 155-160.
REXSTAD, E., AND K. P. BURNHAM. 1991. User's guide for
interactive program CAPTURE. Abundance estima-
tion of closed animal populations. Colorado State
University, Fort Collins.
Studies of the genetics and ecology of Heliconius erato
(Lepidoptera, Nymphalidae). III. Population size, pre-
adult mortality, adult resources, and polymorphism
in natural populations. Rev. Brasil. Biol. 45: 563-569.
SEBER, G. A. F. 1965. A note on the multiple-recapture
census. Biometrika 52: 249-259.
1996. Marginality: Scale determined processes and
the conservation of the British butterfly fauna.
Biodiver. Conserv. 5: 1131-1141.
SILVA, L. M., AND A. M. ARAUJO. 1994. The genetic struc-
ture of Heliconius erato populations (Lepidoptera:
Nymphalidae). Rev. Brasil. Genet. 17: 19-24.
creased fluctuations of butterfly populations towards
the northern edges of species ranges. Ecography 17:
VANDERPLANK, J. 2000. Passion Flowers and Passion
Fruit, 3rd ed. MIT Press, Cambridge, MA.
statistical power of two butterfly monitoring
schemes to detect trends. J. Appl. Ecol. 34: 817-828.
WHITE, G. C., AND K. P. BURNHAM. 1999. Program
MARK: survival rate estimation from both live and
dead encounters. Bird Study 46: S120-S139.

Florida Entomologist 88(2)


'Departamento de Entomologia Tropical, El Colegio de la Frontera Sur. Tapachula, Chiapas, M6xico 30700

2Universidad Aut6noma de Nuevo Le6n (UANL) Apartado Postal 7-F. San Nicolas de los Garza
Nuevo Le6n, M6xico 66450


Volatile compounds released by disturbed females of the bethylid wasp Cephalonomia steph-
anoderis Betrem were collected and analyzed by gas chromatography-mass spectrometry.
The origin of volatiles and their behavioral effects on conspecifics were also investigated.
The source of the volatile compounds was found to be the head, and more specifically, the
mandibular glands. These glands contain skatole as the main volatile component. Behav-
ioral bioassays demonstrated that extracts of parasitoid heads and synthetic skatole evoked
the same alarm behavior in this species. The possible function of this chemical is discussed.

Key Words: Cephalonomia stephanoderis, Hypothenemus hampei, alarm pheromone, skatole,
biological control


Los compuestos volatiles liberados por hembras molestadas del parasitoide betilido Cepha-
lonomia stephanoderis Betrem fueron colectadas y analizadas por cromatografia de gases y
espectrometria de masas. El origen de los volatiles y su efecto comportamental en los para-
sitoides conespecificos fueron tambi6n investigados. La fuente de los compuestos volatiles
fue localizada en la cabeza, y mas especificamente, en las glandulas mandibulares. Estas
glandulas contienen skatol como el principal component volatil. Los bioensayos comporta-
mentales demostraron que los extractos de cabezas del parasitoide y skatol sint6tico provo-
caron el mismo comportamiento de alarma en esta especie. Se discute la possible funci6n de
este compuesto quimico.

Translation provided by the authors.

The bethylid wasp Cephalonomia stephano-
deris Betrem (Hymenoptera: Bethylidae) is an
ectoparasitoid of larvae and pupae of the coffee
berry borer, Hypothenemus hampei (Ferrari) (Cole-
optera: Scolytidae), which is the most important
pest of coffee worldwide (Barrera et al. 1990;
Murphy & Moore 1990). Cephalonomia stephano-
deris is native to Central West Africa and has been
introduced to various coffee-producing countries
(Murphy & Moore 1990). Adults of C. stephano-
deris emit a strong odor when they are disturbed
or transported to be released in field (G6mez
2005). This odor can be detected by the human
nose (Infante et al. 2001). A number of parasitic
wasps are known to emit more or less pungent
odors (Townes 1939). The function of these odors
in wasp behavior remains largely unknown; in
some cases they have been thought to have a de-
fensive role (Buckingham 1975) or to play an im-
portant role in courtship (Williams et al. 1988).
However, only one Cephalonomia species, C. gal-

licola (Ashmead), a cosmopolitan ectoparasitoid
of anobiid beetles, has been reported to release an
odor when squashed by forceps (Kuwahara 1984).
The odor originated from the head, and the chem-
ical identified was skatole (3-methylindole). In-
fante et al. (2001) suggested that a similar secre-
tion could be released by C. stephanoderis, but no
studies have been carried out to identify the
chemicals released by this species.
We describe here behavioral evaluation, ori-
gin, and identification of the volatile compounds
emitted by adult female ofC. stephanoderis when


Biological Material and Experimental Conditions

Adult C. stephanoderis were obtained from the
laboratory colony maintained at El Colegio de la
Frontera Sur, Tapachula, Chiapas, Mexico. Al-

June 2005

G6mez et al.: Volatiles released by Cephalonomia stephanoderis

though both sexes emit the odor (J. G6mez unpub-
lished), females were chosen because of their im-
portance as biological control agents and because
the sex ratio is markedly biased in favor of fe-
males (Barrera 1994). The colony was established
with insects collected from coffee plantations
near Tapachula in 1999 and reared as described
by Barrera et al. (1991). Bioassays were con-
ducted in a room at 24 2C, 80 + 5% relative hu-
midity and lit with red light (10 lux). Parasitoids
used in the bioassays were collected from adult
emergence jars on the day of the tests and placed
in the bioassay room 3 h before testing.

Headspace Bioassay
Two groups of 20 C. stephanoderis females were
placed in a separate glass vial (50 mm high x 20
mm diameter) and a plastic vial (70 mm high x 20
mm diameter) connected to each other by a plastic
tube (Fig. 1). The first group of insects was strongly

Air flow


shaken for 1 min. Preliminary observations showed
that during this time the insects released the odor.
The second group of insects was not disturbed. A
disposable syringe was used to inject 35 ml of clean
air into the glass vial in order to blow the volatiles
through the tube on to undisturbed insects. Test in-
sects were observed for any change in their behav-
ior for 10 min after the influx of air. Alarm behavior
was considered to happen if insects showed move-
ments such as agitated running or attempts to take
flight. No change in behavior was considered a lack
of observable response. Clean air and the odor from
undisturbed insects were used as controls. Six rep-
licates per treatment were performed.

Bioassays with Extracts and Synthetic Skatole
Four different extracts were made with 20 and
40 heads, 40 thoraces, and 40 abdomens of C.
stephanoderis females. Heads, thoraces, and abdo-
mens were macerated in 200 pl of hexane. For eval-

Nylon gauze


Nylon gauze

2 cm

Fig. 1. Design of olfactometer used to determine the response of C. stephanoderis females to volatiles from dis-
turbed and undisturbed insects and clean air.

Florida Entomologist 88(2)

uating the biological activity of these extracts, 10
pl of the chosen extract (equivalent to one or two
heads, two thoraces, or two abdomens depending
on the extract used) was applied to a piece (1 cm2)
ofWhatman No. 2 filter paper. The solvent was al-
lowed to evaporate for 1 min and then the piece of
filter paper was placed in a Petri dish (10 cm diam-
eter). An upside down glass tube (70 mm high x 15
mm diameter) containing 20 undisturbed female
parasitoids (prepared 2 h prior to the test) was
placed over the filter paper. An alarm response was
noted if the insects exhibited any movement. The
number of insects showing alarm behavior was
counted every 5 min for 40 min. In other experi-
ments, several doses (0.1, 0.5, 1, 10, and 100 ng) of
synthetic skatole in hexane were tested in the bio-
assay described above. Hexane alone was used as
control in all experiments. Ten replicates were per-
formed for all treatments. Skatole (98%) was ob-
tained from Sigma Chemical, Co. (Toluca, Mexico).

Solid Phase Micro-Extraction (SPME)

SPME was conducted with a holder and a 100-
pm poly-(dimethylsiloxane)-coated fiber which
were obtained from SUPELCO (Toluca, Mexico).
Twenty C. stephanoderis females were placed in-
side a glass vial (7.5 mm high x 1.5 cm diameter)
with a foam cap. Sampling was performed by in-
serting the SPME needle, through the foam cap,
into the headspace of the glass vial. Volatiles were
allowed to be adsorbed onto the fiber for 5 min.
Subsequently, it was removed from the vial and
volatiles desorbed inside the heated injection port
of a gas chromatograph for 5 min.

Solid Sampling Preparation for Chemical Analysis

Females of C. stephanoderis were killed by
placing them in a refrigerator for 24 h. Dissec-
tions were made under a binocular microscope
with fine forceps and entomological pins. Three
heads, thoraces, and abdomens, and ten mandi-
bles with the gland attached were dissected in
distilled water and placed in thin-walled soft
glass tubes previously sealed at one end; the open
end was then sealed in a micro-flame for analysis
by coupled gas chromatography-mass spectrome-
try (Morgan 1990). In addition, three whole in-
sects were analyzed by this technique.

Chemical Analysis

Gas chromatography-mass spectrometry (GC-
MS) was conducted with a Varian Star 3400 CX gas
chromatograph linked to a Varian Saturn 4D mass
spectrometer. The samples were analyzed in a
fused silica column (30 m x 0.25 mm) coated with
poly-(5%-diphenyl-95%-dimethylsiloxane) pro-
grammed from 50C to 250C at 15C/min. The
flow rate of helium through the column was main-

trained at 1 ml/min. The injector port temperature
was held at 200C. The glass capillaries containing
either the glands, heads, abdomens, or thoraces
were directly inserted into the injection area and
heated and crushed as described by Morgan (1990).

Scanning Electron Microscopy

Mandibular glands of C. stephanoderis females
were washed and fixed in a solution of 3% glutaral-
dehyde in phosphate buffer (0.1 M, pH 7.2). Glands
were washed twice for 5 min with distilled water,
and then passed through increasing grades of eth-
anol, from 10% to absolute ethanol, 30 min each.
Finally they were dried to critical point of CO,,
mounted in aluminum stubs and sputter coated
with gold-paladium (Dykstra 1993). The samples
were examined and photographed in a Topcon SM-
510 electron microscope operated at 5 kV.

Statistical Analysis

Results were subjected to one-way analysis of
variance (ANOVA) (SPSS for Windows 8.0. SPSS
Inc.), except the data for insect response to head
extracts and skatole over time were analyzed by
repeated measures ANOVA. When F values were
significant, means were compared by Tukey's test
at a = 0.05.

Headspace Bioassay

Undisturbed C. stephanoderis females showed
a significant response to volatiles from disturbed
females compared with the response to volatiles
from undisturbed females and clean air (F = 85.5;
df= 2, 15; P < 0.001). From a total of 120 insects
tested, 81 females were affected by the odor from
disturbed females; the rest (39) remained station-
ary. In contrast, most of the individuals remained
stationary when clean air and air from undis-
turbed insects (111 out of 120 in each treatment)
was passed over the parasitoids.

Female Response to Extracts

Females showed a significant alarm response
(e.g., agitated running) to head extracts, but not to
thorax extracts or abdomen extracts or hexane con-
trol (F = 90.7; df= 4, 45; P < 0.001) (Fig. 2). Of the
total of females tested, 64% were disturbed when a
one-head equivalent extract was introduced on the
filter paper, whereas 80% were disturbed when a
two-head equivalent extract was used.

Chemical Analysis

The SPME and GC-MS analysis of the parasi-
toid volatiles showed that agitated C. stephano-
deris females released at least two compounds (Fig.

June 2005

G6mez et al.: Volatiles released by Cephalonomia stephanoderis

b T


Control One




Fig. 2. Percentage of C. stephanod
responded to extracts containing equ
two heads, two thoraces, or two abdom
indicate the standard error of the me
ters over bars indicate that means are
ferent (ANOVA, followed by Tukey tes

skatole (F = 5.0; df = 4, 45; P = 0.002). Multiple
comparisons indicated that the dose of 1 ng ska-
tole elicited significantly larger alarm behavior
compared with those evoked by the doses of 0.1
and 100 ng of this compound. The alarm re-
sponses elicited by the doses of 0.5 and 10 ng of
skatole were intermediate between and not sig-
nificantly different from those evoked by the
C C doses of 1, 0.1, 100 ng of this compound (Fig. 4).
I EThe wasp response to head extracts and skatole
Two Two over time revealed that the type of chemical stim-
horaces abdomens uli used did not influence differently the alarm be-
havior ofC. stephanoderis (F = 1.23; df= 2, 27; P =
0.307). In contrast, time affected the wasp re-
eris females that sponse to the chemical stimulus (F = 6.34; df = 7,
ivalent of one or 189; P < 0.001). In the three treatments, the para-
ens. Vertical bars sitoids started to respond soon after the samples
an. Different let- were delivered to the vial reaching the highest re-
significantly dif- sponse at 15-20 min (35 min in the case of ska-
t, P < 0.05) tole), and after this time the insect response grad-
ually declined (Fig. 5). The chemical stimulus x
time interaction term was not significant (F =

3a). Compound 1 was identified as skatole (3-meth-
ylindole) by comparison of retention time and mass
spectrum with that of the synthetic standard.
Compound 2 with a Kovat's Index of 15.95 showed
mass spectrum fragment ions at m/z 55 (80%), 69
(70%), 83 (25%), 97 (100%), and 111 (25%). This
mass spectrum resembled that of a branched alk-
ene. Undisturbed insects did not release skatole or
compound 2 (data not shown). Solid sampling anal-
ysis of heads of C. stephanoderis showed that the
volatiles contained a mixture of hydrocarbons and
nitrogen compounds (Fig. 3b). The compounds de-
tected were skatole, unidentified compound 2, and
other nitrogen compounds which were tentatively
identified by mass spectral matching to a library
data base (NIST 2002) as (3) uric acid, (4) dl-ala-
nyl-l-leucine, (5) hexahydro-3-[2-methylpropyl]-
pyrrolo [1, 2-a] pyrazine-1, 4-dione and (7) oleam-
ide. Compounds 6, 8, 9, 10, 11, 12, 13, 14, 15, and
16, which were common in all solid samples were
cuticular in origin as confirmed by analysis of a
small fragment of cuticle. They have mass spectra
typical of hydrocarbons, which have been identified
previously from C. stephanoderis by Howard & In-
fante (1996). The analysis of the mandibular
glands showed that one of the components of the
glands was skatole with traces of all the other com-
pounds found in the head (Fig. 3c). Analysis of tho-
races and abdomens showed to these contained the
compounds 3, 4, 5, and 7, and hydrocarbons but not
skatole or unidentified compound 2 (Figs. 4d and
4e), confirming that compounds 1 and 2 are found
specifically in the head. Skatole content in the head
varied from 0.4 to 1.0 ng (n = 6; 0.5 + 0.1 SE).

Female Response to Synthetic Skatole

The alarm behavior of C. stephanoderis was
significantly influenced by the dose of synthetic

1.24; df= 14, 189;P = 0.25).

Scanning Electron Microscopy

The microphotograph showed that the man-
dibular gland is connected to the base of the man-
dible (Fig. 6a). The gland is comparable in size to
the mandible. A close-up shows that the gland
consists of a series of tubular structures attached
to the mandible (Fig. 6b).


The alarm behavior ofC. stephanoderis was ob-
served in individuals that were exposed to the
headspace volatiles collected from disturbed fe-
males. No such behavior was observed in wasps
exposed to the volatiles collected similarly from
undisturbed females. SPME and GC-MS analysis
showed that skatole was the main volatile com-
pound emitted by disturbed females. The presence
of skatole has been reported in two species of Neu-
roptera (Blum & Wallace 1973), one species of Tri-
choptera (Blum 1981), one species of Coleoptera
(Burger et al. 2002) and several species of Hy-
menoptera (Law et al. 1965; Smith & Roubick
1983; Kuwahara 1984; Keegans et al. 1993; Billen
et al. 1998). For instance, skatole was found in the
ant Pheidole fallax Mayr, although its function
was not determined (Law et al. 1965). This com-
pound is released from the abdomen of the army
ant, Labidus praedator (Smith) and functions as a
trail pheromone (Keegans et al. 1993). The man-
dibular gland of a leptanillinan ant, Leptanilla sp.
contains a large amount of skatole that is released
as an alarm pheromone (Billen et al. 1998). Ska-
tole is present in the mandibular gland of a sting-
less bee, Melipona interrupta triplaridis Schwarz
as a component of the alarm pheromone (Smith &

Florida Entomologist 88(2)


-. _, k _


VL ti ,14
b 9 15 t

20- 4
2 35 -, ,
_________-_-_-_- ^- ^ - --^
4Ccn d


54 6
35 ,

5 mill

Fig. 3. Representative chromatograms of compounds released by disturbed females collected by SPME (a); and
solid sampling analysis of crushed heads (b), mandibular glands (c), abdomens (d), and thoraces (e) of C. stephan-
oderis females. Skatole (1), unidentified (2), uric acid (3), dl-Alanyl-l-leucine (4) hexahydro-3-[2-methylpropyl]-pyr-
rolo [1,2-a] pyrazine-1, 4-dione (5), oleamide (7) and hydrocarbons (6), (8), (9), (10), (11), (12), (13), (14), (15) and (16).
*Contaminants from the solid phase fiber.

Roubik 1983). Kuwahara (1984) detected skatole
in the head of C. gallicola, another bethylid wasp,
and proposed a function as an allomone.
Female parasitoids ofC. stephanoderis showed
the highest alarm response to one and two-head
equivalent extracts and 1 ng of skatole. However,
response was reduced to higher doses of skatole,

which may indicate that higher doses of skatole
may disrupt the effective alarm communication
between parasitoids, as has been found in aphids
(El-Agamy & Haynes 1992). A high dose of (E)-p-
farnesene, the aphid alarm pheromone, produced
a rapid sensory habituation of aphids to this com-
pound (Calebrese & Sorenson 1978). In our study,

June 2005

G6mez et al.: Volatiles released by Cephalonomia stephanoderis





ab b


I 40-

0.1 0.5 1 10 100

Quantify of skatole (ng)

Fig. 4. Percentage of C. stephanoderis females that
responded to different quantities of skatole. Vertical
bars indicate the standard error of the mean. Different
letters over bars indicate that means are significantly
different (ANOVA, followed by Tukey test, P < 0.05).

the stimulatory effect of head extracts or skatole
was short-lived (<35 min); therefore the decrease
in the alarm response ofC. stephanoderis may be
due to sensory adaptation or habituation.
The odor released by agitated adults of
C. stephanoderis may have multiple functions as
reported for other insects (e.g., Blatt et al. 1998;
Staples et al. 2002; Wardle et al. 2003). In this
study, we analyzed only females, but preliminary
studies have indicated that male parasitoids also
release the same compounds (unpublished data).
The fact that both sexes produce the same compo-
nents in the secretion suggests that they do not
function as sexual pheromones. Generally, defen-
sive secretions are released by both sexes as has
been demonstrated for bugs (Leal et al. 1994),
thrips (Teerling et al. 1991), and cockroaches
(Farine et al. 2002). For example, the defensive
secretions of the glandular pouches of the adults
of both sexes of cockroaches Therea petiveriana

-A- one head
S- ---two heads
- - 1 ng skatole

5 10 15 20 25 30 35 40

Time (min)

Fig. 5. Percentage of C. stephanoderis females that
responded to one or two-head equivalent extracts and
skatole (1 ng), at different times of observation. Vertical
bars indicate the standard error of the mean.

(L.) contain volatile compounds that function as
an alarm pheromone for adults (Farine et al.
2002). A potential function for skatole in C. steph-
anoderis is as an alarm pheromone causing dis-
persal. Cephalonomia stephanoderis adults are
found in groups of sisters and brothers inside cof-
fee berries after they emerge from the cocoon and
they remain together 4-5 days to mate before they
disperse (Barrera et al. 1989). A pheromone could
promote the dispersion of the parasitoids after
mating. A prerequisite to the evolution of alarm
pheromones is the evolution of sociality (Nault &
Phelan 1984). Another possible function of ska-
tole in C. stephanoderis is as an epideictic phero-
mone, promoting spacing in the natural habitat
(Haynes & Birch 1985). Hypothenemus hampei,
the host of C. stephanoderis, reproduces inside
coffee fruits, which may represent a limited re-
source for both coffee berry borer and parasitoids,
thus an epideictic pheromone could well be ad-

Fig. 6. Scanning photomicrograph of the mandibular gland attached to the mandibula of C. stephanoderis, M =
mandible; MG = mandibular gland (a), Close up of the tubular structure of the gland (b).

Florida Entomologist 88(2)

aptative for individuals to reduce competition for
resources. Barrera et al. (1994) presented evi-
dence of a marking pheromone in C. stephano-
deris to avoid use of hosts previously parasitized,
but this possible pheromone seems to act over
short distances compared to the odor released by
agitated adults. A third possible function is that
the odor is released in direct response to threats.
Several species of spiders and ants have been re-
ported to attack to C. stephanoderis as well as a
bethylid wasp, Prorops nasuta Waterston in the
coffee plantations of Mexico (Henaut et al. 2001;
Infante et al. 2003). Some ant species occasionally
forage inside coffee fruits and prey upon imma-
ture and adult P nasuta and possibly C. stephan-
oderis (Infante et al. 2003). The hyperparasitoid
Alloxysta brevis (Thompson) applies defensive
compounds stored in mandibular gland reservoirs
against attacking ants and other generalist pred-
ators like spiders (Hubner & Dettner 2000). Fi-
nally, the odor could mediate the interactions be-
tween C. stephanoderis and two other species of
bethylid parasitoids of the coffee berry borer, P.
nasuta and C. hyalinipennis Ashmead. Female
parasitoids actively defend parasitized hosts and
their progeny when intruders attempt to take
possession of these resources (P6rez-Lachaud
et al. 2002). Thus, it is possible that the odor of
C. stephanoderis could function as an alarm pher-
omone, epideictic pheromone, or allomone de-
pending on the specific situation. For example,
the same compounds that function as an alarm
pheromone in the bedbug, Cimex lectularius L.,
also serve a defensive role, as they can effectively
repel Monormorium pharaonis (L.), a natural en-
emy of bedbugs (Levinson et al. 1974). These
glandular secretions also make bedbugs distaste-
ful to bat species which are predators to C. lectic-
ularius. It is believed that this defensive role may
have been the primary function of the secretion
and the alarm response of conspecifics evolved
secondarily (Nault & Phelan 1984).
Our microphotographs of the mandibular
gland of female C. stephanoderis show that the
gland is not typical of Hymenoptera, which have
been shown to contain a reservoir in the form of a
sac (e.g., Cruz-Landim 1990; Mayhe & Caetano
1994). The mandibular gland secretion in C.
stephanoderis presumably is produced by the tu-
bular structures attached to the mandible.
In conclusion, this study showed that the be-
havior of female wasps of C. stephanoderis was af-
fected by the introduction of volatiles released
from disturbed conspecific females. The source of
the volatile compounds was found to be the head,
and more specifically, the mandibular glands.
These glands contain skatole as the main volatile
component. Behavioral bioassays demonstrated
that extracts of parasitoid heads and synthetic
skatole evoked the same alarm behavior in this
species. An unidentified compound 2 released by

agitated females should be properly identified and
its biological activity evaluated alone and in com-
bination with skatole. Other unidentified com-
pounds found in the heads and mandibular glands
beside skatole may not be involved in the alarm
behavior of C. stephanoderis because they were
not detected in the headspace volatile analysis.


We thank Francisco Infante and Trevor Williams
(ECOSUR) for helpful reviews of the manuscript, Javier
Valle-Mora for statistical advice, and Guadalupe Nieto
for the scanning photomicrographs. J. G6mez received a
doctoral grant from CONACyT.


BARRERA, J. F. 1994. Dynamique des populations du sc-
olyte des fruits du caf6ier, Hypothenemus hampei
(Coleoptera: Scolytidae), et lutte biologique avec le
parasitoide Cephalonomia stephanoderis (Hymenop-
tera: Bethylidae), au Chiapas, Mexique. PhD thesis.
University Paul Sabatier, Tolouse, France.
dence for a marking pheromone in host discrimina-
tion by Cephalonomia stephanoderis (Hym.:
Bethylidae). Entomophaga 39: 363-366.
W. DE LA ROSA. 1989. Biologie de Cephalonomia
stephanoderis Betrem (Hymenoptera; Bethylidae)
en laboratoire. I. Cycle biologique, capacity d'oviposi-
tion et emergence du fruti du caf6ier. Caf6 Cacao Th6
33: 101-108.
PHY, AND C. PRIOR 1990. Biological control of the cof-
fee berry borer Hypothenemus hampei in Mexico and
possibilities for further action. Brighton Crop. Prot.
Conf. Pest and Diseases 4: 391-396.
ROSA, AND J. GOMEZ. 1991. Cria y manejo de Cepha-
lonomia stephanoderis y Prorops nasuta, parasi-
toides de la broca del caf6. Miscelaneas de la
Sociedad Colombiana de Entomologia 18: 76-86.
The mandibular gland, probably the source of the
alarm substance in Leptanilla sp. (Hymenoptera,
Formicidae). Naturwissenschaften 85: 596-597.
AND G. GRIES. 1998. Alarm pheromone system of the
western conifer seed bug, Leptoglossus occidentalis.
J. Chem. Ecol. 24: 1013-1031.
BLUM, M. S. 1981. Chemical Defenses of Arthropods.
Academic Press, New York.
BLUM, M. S., AND J. B. WALLACE. 1973. Skatole and
tridecene: Identification and possible role in a
chrysopid secretion. Insect Biochem. 3: 353-357.
BUCKINGHAM, G. R. 1975. The parasites of walnut husk
flies (Diptera: Tephritidae: Ragholetis) including com-
parative studies on the biology of Biosteres juglandis
Mues. (Hymenoptera: Braconidae) and on the male
tergal glands of Braconidae (Hymenoptera). PhD the-
sis. University of California, Berkeley.
Z. M. MUNRO. 2002. Semiochemicals of the Scara-
baeinae. VII: Identification and synthesis of EAD-

June 2005

G6mez et al.: Volatiles released by Cephalonomia stephanoderis

active constituents of abdominal sex attracting se-
cretion of the male dung beetle, Kheper subaeneus. J.
Chem. Ecol. 28: 2527-2535.
CALEBRESE, E. J., AND A. J. SORENSON. 1978. Dispersal
and recolonization by Myzus persicae following
aphid alarm pheromone exposure. Ann. Entomol.
Soc. Am. 71: 181-182.
CRUZ-LANDIM, C. 1990. Cephalic exocrine glands of
ants: A morphological view, pp. 102-118 In R. K.
Vander Meer, K. Jaffe and A. Cedeno [eds.], Applied
Myrmecology: A World Perspective. Westview Press,
Boulder, CO.
DYKSTRA, M. J. 1993. A Manual of Applied Techniques
for Biological Electron Microscopy. Van Nostrand
Reinhold Co., New York.
EL-AGAMY, F. M., AND K. F. HAYNES. 1992. Susceptibil-
ity of the pea aphid (Homoptera: Aphidae) to an in-
secticide and a predator in the presence of synthetic
aphid alarm pheromone. J. Econ. Entomol. 85: 794-
GRANDCOLAS, AND R. BROSSUT. 2002. Defensive se-
cretion of Therea petiveriana: Chemical identifica-
tion and evidence of an alarm function. J. Chem.
Ecol. 28: 1629-1640.
GOMEZ, J. 2005. Factores que afectan la capacidad de
dispersion de Cephalonomia stephanoderis y la
busqueda de su hospedero, la broca del caf6. PhD
Thesis. El Colegio de la Frontera Sur, Mexico.
HAYNES, K. F., AND M. C. BIRCH. 1985. The role of other
pheromone, allomones and kairomones in the behav-
ioral responses of insects, pp. 225-255 In G. A.
Kerkut and L. I. Gilbert [eds.], Comprehensive In-
sect Physiology, Biochemistry and Pharmacology,
Vol. 9. Pergamon Press. Oxford, UK.
IAMS. 2001. Retention, capture and consumption of
experimental prey by orb-web weaving spiders in
coffee plantations of Southern Mexico. Entomol. exp.
Appl. 98: 1-8.
HOWARD, R. W., AND F. INFANTE. 1996. Cuticular hydro-
carbons of the host-specific ectoparasitoid Cephalo-
nomia stephanoderis (Hymenoptera: Bethylidae)
and its host the coffee berry borer (Coleoptera: Sco-
lytidae). Ann. Entomol. Soc. Am. 89: 700-709.
HUBNER, G., AND K. DETTNER 2000. Hyperparasitoid
defense strategies against spiders: the role of chem-
ical and morphological protection. Entomol. exp.
Appl. 97: 67-74.
tion by native arthropods on the African parasitoid
Prorops nasuta (Hymenoptera: Bethylidae) in coffee
plantations of Mexico. Florida Entomol. 86: 86-88.
AND S. FOWLER 2001. Interespecific competition be-
tween Cephalonomia stephanoderis and Prorops na-
suta (Hym., Bethylidae), parasitoids of the coffee
berry borer, Hypothenemus hampei (Col., Sco-
lytidae). J. Appl. Entomol. 125: 63-70.
GOKCEN. 1993. Volatile glandular secretions of three
species of new world army ants, Eciton burchelli, La-

bidus coecus, and Labidus praedator. J. Chem. Ecol.
19: 2705-2718.
KUWAHARA, Y. 1984. Identification of skatole from a be-
thylid wasp, Cephalonomia gallicola (Ashmead)
(Hymenoptera: Bethylidae). Agric. Biol. Chem. 48:
Biochemical polymorphism in ants. Science 149:
LEAL, W. S., A. R. PANIZZI, AND C. C. NIVA. 1994. Alarm
pheromone system of leaf-footed bug Leptoglossus
zonatus (Heteroptera: Coreidae). J. Chem. Ecol. 20;
1974. Action and composition of the alarm phero-
mone of the bedbug, Cimex lectularis L. Naturwis-
senschaften 12: 684-685.
MAYHE, A., AND F. CAETANO. 1994. Ultramorphology
and comparison between the mandibular glands and
mandibles of two species of Acromyrmex (Hy-
menoptera: Formicidae). Naturalia 19: 17-27
MORGAN, E. D. 1990. Preparation of small-scale sam-
ples from insects for chromatography. Anal. Chim.
Acta 236: 227-235.
MURPHY, S. T., AND D. MOORE. 1990. Biological control
of the coffee berry borer, Hypothenemus hampei (Fer-
rari) (Coleoptera: Scolytidae): previous programmes
and possibilities for the future. Biocontrol News In-
form. 11: 107-117.
NAULT, L. R., AND P. L. PHELAN. 1984. Alarm phero-
mone and sociality in pre-social insects, pp. 237-256
In W. J. Bell and R. T Card6 [eds.], Chemical Ecology
of Insects, Chapman and Hall, London, UK.
LACHAUD. 2002. Insect gladiators: competitive inter-
actions between three species of bethylid wasps at-
tacking the coffee berry borer, Hypothenemus hampei
(Coleoptera: Scolytidae). Biol. Control 25: 231-238.
SMITH, B. H., AND D. W. ROUBIK. 1983. Mandibular
gland of stingless bees (Hymenoptera: Apidae)
chemical analysis of their contents and biological
function in two species ofMelipona. J. Chem. Ecol. 9:
WHITMAN. 2002. Chemical defense in the plant bug
Lopidea robiniae. J. Chem. Ecol. 28: 601-615.
GILLESPIE. 1991. Identification of alarm pheromone
in the western flower thrips, Frankliniella occidenta-
lis. J. Chem. Ecol. 19: 681-697.
TOWNES, H. K., JR. 1939. Protective odours among the
Ichneumonidae (Hymenoptera). Bull. Brooklyn En-
tomol. Soc. 34: 29-30.
GRIES. 2003. Volatile compounds released by dis-
turbed and calm adults of the tarnished plant bug,
Lygus lineolaris. J. Chem. Ecol. 29: 931-944.
VINSON. 1988. Hagen's gland morphology and chem-
ical content analysis for three species of parasitic
wasps (Hymenoptera: Braconidae). J. Chem. Ecol.
14: 1727-1736.

Florida Entomologist 88(2)


'Unidad de Zoologia, Facultad de Biologia, Universidad de Salamanca. 37071- Salamanca, Spain

2Dipartimento di Biologia Evolutiva, Universita degli Studi di Siena-Via A. Moro, 4, 35100, Siena, Italy

3Dipartimento di Biologia, Sezione di Zoologia e Citologia, Universita degli Studi di Milano-
Via Celoria, 26, 20133, Milan, Italy


The mature larva ofA. sikhimensis is described, illustrated, and compared with that of other
species of the genus so far described. The final-instar of this species and the different species
recently described, or redescribed, in this genus are A. kitcheneri, A. longispinosus, and A.
trifasciatus, and all can be differentiated on the basis of the following characters: (a) the de-
velopment of antennae; (b) the development of the atrium with respect to the subatrium; (c)
the number of the sensilla of the labrum and galea, and (d) the presence/absence of spinules
and papillae on the labium.

Key Words: Hymenoptera, Eumenidae, Ancistrocerus, mature larva, Nepal


Se describe, y compare con las ya descritas del g6nero, la larva madura de Ancistrocerus
sikhimensis. Los caracteres que permiten distinguir las larvas maduras, recientemente de-
scritas o redescritas, del g6nero Ancistrocerus: A. kitcheneri,A. longispinosus, yA. trifascia-
tus, radican en: (a) desarrollo de las antenas y del atrium con respect al subatrium; (b)
numero de sensilas del labrum y galeas, y (c) presencia/ausencia de espinulas y papilas en
el labium.

Translation provided by the authors.

Most of the taxonomy of eumenid wasps is
based on external adult morphology, and rela-
tively little attention has been paid to interspe-
cific differences in larval characters, even though
they could be useful. In this respect, of 2500-3500
species described (Yamane 1990), larval morphol-
ogy is only known for 42 (Tormos et al. 1998).
Within this set, only five species of the genusAn-
cistrocerus Wesmael, 1836, have been described:
A. trifasciatus (Miiller, 1776) (Enslin 1921; J0r-
gensen 1942; Tormos et al. 1998); A. oviventris
(Wesmael, 1836) (Micheli 1930); A. gazella (Pan-
zer, 1793) (Grandi 1961), A. kitcheneri (Dusmet,
1917) and A. longispinosus (de Saussure, 1885)
(Tormos et al. 1998). This study addresses the lar-
val morphology ofA. sikhimensis Bingham, 1898,
obtained during a study on the fauna of rubi-
colous species of Nepal.


The methods employed to prepare larval spec-
imens as well as the terminology for larval mor-
phology and the format used in the descriptions
follow Evans (1987) and Sime and Wahl (1998).
The following abbreviations are employed: d = di-

ameter, h = height, 1 = length, w = width. The ma-
terial is deposited at the "Torres-Sala" Entomo-
logical Foundation (Valencia, Spain).
The description is based on four mature larvae
obtained by R. Boesi in Nepal in 2003. Absolute
measurements, except for the body width and
length, are based on data for the one specimen.


A. sikhimensis Bingham, 1898 (Figs. 1-6)

Body (Fig. 1) (1 = 11.7-12.6 mm, maximum w =
2.1-2.3 mm) robust; first five abdominal segments
divided into two annulets by a transverse crease.
Anus terminal, in central position, as a trans-
verse slit. Pleural lobes developed. Integument
with scanty and disperse setae (1 = 9-11 pm) and
punctures. Spiracles (Fig. 2) with walls of atrium
with ridges and asperities; opening into sub-
atrium spinulose; subatrium (d = 84 pm) as wide
as atrium (d = 81 pm).
Cranium (Fig. 3) (w = 1.4 mm, h (exclusive of
labrum) = 1.1 mm) with sparse setae (1 = 9-12 pm)
and punctures. Coronal suture absent and pari-
etal bands present. Antennae (d = 65 pm) almost

June 2005

Tormos et al.: Larval Description ofAncistrocerus sikhimensis

0.5 cm

0.1 mm


1 mm

0.5 mm

0.1 mm

0.1 mm

6 b

Figs. 1-6. Mature larva of Ancistrocerus sikhimensis: (1) Body, lateral view; (2) Anterior thoracic spiracle (side
view) (atrium, subatrium and tracheal trunk); (3) Cranium (frontal view); (3 a) mandible; (4) Labrum; (5) Epiphar-
ynx; (6 a) Maxilla; (6 b) Labium.

Florida Entomologist 88(2)

flat, circular, with 3 sensilla. Clypeus with setae (1
= 4 pm) and punctures. Labrum (Fig. 4) (w = 685
pm) emarginate, with around 68 conical sensilla
(w = 8 pm). Epipharynx (Fig. 5) spinulose, with 16
sensilla (d = 2 pm).
Mouthparts. Mandible (Fig. 3a) (1 = 410 pm, w
= 250 pm) weakly tridentate. Maxilla (Fig. 6a) (w
= 292 pm) spinulose on the lacinial area and
withl6 setae (1 = 5-8 pm) on external part. Maxil-
lary palpus conical (h = 87 pm, w = 45 pm) with 4
protuberant apical sensilla (w = 3-6 pm); galea (1 =
130 pm, w = 50 pm) long, attenuated at apex, with
2 apical sensilla. Labium (Fig. 6b) (w = 375 pm)
spinulose dorsally to salivary orifice; labial palpus
(1 = 80 pm, w = 60 pm) with 4 apical sensilla (w =
3-6 pm); prementum with setae (1 = 12-15 pm);
salivary orifice a transverse slit (w = 130 pm).


The present description of the morphology of
the mature larva ofA. sikhimensis, together with
previous descriptions carried out by our team or
some of its members (Tormos et al. 1998), show
that the mature larvae of Ancistrocerus are very
similar, differing in (a) the presence/absence of
the coronal suture and setae of the labrum; (b)
more or less developed parietal bands; (c) the
number and arrangement of the sensilla of the
epipharynx, and (d) the development of apical
sensilla of the galeae (Table 1). Additionally,


1 2 3 4 5 6 7

A. gazella A A B A A B A
A. kitcheneri A B A B B B A
A. longispinosus A B B B B B B
A. oviventris B B B A A A B
A. sikhimensis B A A B A B A
A. trifasciatus A A A A B B A

Coronal suture: (1) present (A); absent (B). Epipharynx: (2)
with 16 sensilla (A); with 8 to 10-12 sensilla (B). Galea: (3)
longer (higher) than the maxillary palpus (A); as long as or
shorter than the maxillary palpus (B). Labrum: (4) with setae
(A); without setae (B). Maxillae: (5) with more than 6 setae on
the external margin (A); with 6 setae on the external margin
(B). Maxillary palpus: (6) with 6 sensilla at apex (A); with 4 sen-
silla at apex (B). Parietal bands: (7) well developed (A); very
weakly developed (practically absent) (B).

other differences can be observed between A.
sikhimensis and A. kitcheneri, A. longispinosus,
and A. trifasciatus, recently described or re-
described, by the authors. These differences are
(a) the development of antennae; (b) the develop-
ment of the atrium with respect to the subatrium;
(c) the number of the sensilla of the labrum and
galea, and (d) the presence/absence of spinules
and papillae on the labium.


We are indebted to E. Chiappa (Universidad
Catolica de Valparaiso) and J. Kojima (Ibaraki Univer-
sity) for comments on the manuscript. Financial sup-
port for this study was provided by the Junta de Castilla
y Le6n, project SA 18/96.


ENSLIN, E. 1921. Beitrage zur kenntnis der Hymenop-
teren II. 3. Biologie von Symmorphus sinuatus F. 4.
Biologie von Ancistrocerus trifasciatus F. Deutsch
entomologische Zeitschr. 279-285.
GRANDI, G. 1961. Studi di un entomologo sugli imenot-
teri superior. Bollettino dell'Istituto di Entomologia
della Universita di Bologna XXV: 1-659.
JORGENSEN, P. 1942. Biological observations on some
solitary vespides. Ent. Meddv. 22: 299-335.
MICHELI, L. 1930. Note biologiche e morfologiche sugli
imenotteri contributeo 2). Memorie della Societa En-
tomologica Italiana 9: 46-66.
SIME, R. K., AND D. B. WAHL. 1998. Taxonomy, mature
larva, and observations on the biology of Gnamp-
topelta obsidianator (Brulle) (Hymenoptera: Ichneu-
monidae, Ichneumoninae). J. Hymenoptera Res.
7(2): 157-164.
1998. Description of the mature larvae of Ancistro-
cerus kitcheneri (Dusmet, 1917), A. longispinosus
longispinosus (Saussure, 1885) and redescription of
that ofA. trifasciatus (Miller, 1776). (Hymenoptera,
Vespidae). Nouvelle Revue d'Entomologie (NS) 15:
EVANS, H. E. 1987. Order Hymenoptera, pp. 597-710 In
F. W. Stehr [ed.], Immature Insects, Kendall/Hunt
Publishing Company, Dubuque, Iowa.
YAMANE, SK. 1990. A revision of the Japanese Eu-
menidae (Hymenoptera, Vespoidea). Insecta-mat-
sumurana, N.S. 43: 1-189.

June 2005

Scientific Notes


'University of Florida, IFAS, Gulf Coast Research & Education Center, Wimauma, FL 33598

2Florida Department of Agriculture and Consumer Services, Fruit Fly Identification Laboratory, Palmetto, FL 34221

The pepper weevil, Anthonomus eugenii Cano,
is a serious pest of cultivated Capsicum spp. pep-
pers in the southern United States, Hawaii, Mex-
ico, Guatemala, Honduras, Costa Rica, and Puerto
Rico (Schuster et al. 1996). Eggs are deposited in
flower buds and fruit, where larvae and pupae
complete their development. Infested buds and
fruit often abscise, but larvae and pupae can com-
plete development if fallen buds and fruit do not
desiccate. Yield losses can reach 90% in Florida, if
the weevil is not controlled (Schuster & Everett
1982). Broad spectrum insecticides have been
used most often to manage the pest but may lead
to unintended consequences, including insecticide
resistance and outbreaks of non-target pests. Bio-
logical control could be an alternative or adjunct
to insecticides in managing the pepper weevil.
At least three species of predators and seven
species of parasitoids have been reported to at-
tack the pepper weevil (Riley & King 1994). The
most abundant parasitoid recovered from the
pepper weevil in Florida was Catolaccus hunter
Crawford (Hymenoptera: Pteromalidae) (Riley &
Schuster 1992). While natural enemies generally
are regarded as contributing little to control of
the pest (Elmore & Campbell 1954), 50% parasit-
ism of pepper weevil larvae by C. hunter was ob-
served in fallen jalapeno buds and over 20% par-
asitism in fallen bell pepper buds (Schuster et al.
1988). Augmentative releases of C. hunter on al-
ternative host plants during the summer off-sea-
son and on pepper at the initiation of flowering
have resulted in reduced or delayed damage by
weevil larvae (Schuster unpublished data). Be-
cause C. hunter has shown potential for bio-con-
trol of the pepper weevil, a method of rearing the
parasitoid in the laboratory is needed.
A commercial diet for rearing pepper weevil
larvae is available (Bio-Serv, Entomology Divi-
sion, Frenchtown, NJ); however, the diet was not
used due to low egg hatch (Toapanta 2001). Be-
cause rearing the pepper weevil in pepper fruit is
time and space consuming, an alternative host
was sought. The cowpea weevil, Callosobruchus
maculatus Fabricius (Coleoptera: Bruchidae),
was shown to be a suitable factitious host for rear-
ing Catolaccus grandis (Burks) (Rojas et al. 1998),
a closely related parasitoid of the boll weevil,
A. grandis grandis Boheman. The C. maculatus
larvae were encapsulated in Parafilm (Pechiney

Plastic Packaging, Inc., Menasha, WI) for presen-
tation to parasitoid adults. This method was de-
veloped for exposingA. grandis grandis larvae to
ovipositing C. grandis (Cate 1987) and was mech-
anized for mass production (Roberson & Harsh
1993). Methods also were developed for producing
C. maculatus larvae in pieces of garbanzo beans
(chick peas), Cicer arietinum L. (Leyva et al.
2002). The pieces were not large enough for lar-
vae to complete their development within, thus
forcing the larvae to exit the bean pieces. The lar-
vae then were easier to collect prior to encapsula-
tion. This method had been used successfully to
rear C. hunter in the laboratory. Life history pa-
rameters including pre-oviposition period, ovipo-
sition period, adult longevity, fecundity, and egg
to adult development period ofC. hunter onA. eu-
genii were found to be the same whether the par-
asitoid had originally been reared on either
C. maculatus or A. eugenii (Seal et al. 2002). Col-
lecting larvae and encapsulating them in Para-
film represents extra investments in time and
equipment. Therefore, a method was developed
for rearing C. hunter on C. maculatus larvae di-
rectly in garbanzo beans.
Two colonies of C. maculatus were maintained
in a room at a temperature of about 27C, relative
humidity of about 60% and a photoperiod of
14L:10D. The colonies were maintained on black-
eyed peas, Vigna unguiculata (L.) Walp., and on
garbanzo beans. The black-eyed peas were used to
maintain the colony of C. maculatus and the gar-
banzo beans were used for exposing C. maculatus
larvae to the C. hunter parasitoid.
Three times a week, six narrow-mouth 800-ml
"Mason" glass jars (Ball Corporation, Muncie, IN)
were filled with 300 g of black-eyed peas each.
About 100 C. maculatus adults were collected
with an aspirator connected to a vacuum pump
and were deposited in each jar, which then was
sealed with a screen, filter disc, and metal ring.
These jars were stored upright. A new generation
of bruchid adults emerged about every 30 d.
Three times a week, ca. 400 C. maculatus
adults were collected with an aspirator and put
into each of ten 800-ml glass jars that contained
300 g each of garbanzo beans. The C. maculatus
adults were removed 48 h later by placing the
beans and bruchids on a metal sieve placed in the
large opening of a 25-cm diam galvanized funnel,

Florida Entomologist 88(2)

the narrow end of which was attached to a wet/
dry vacuum cleaner. The vacuum was operated
until all C. maculatus adults were drawn through
the sieve. The beans then were returned to the
jars, which were laid on their sides. In about 21 d,
the hatching larvae were 4th instars, the lifestage
used previously for parasitism (Rodriguez-Leyva
et al. 2000). The larvae form pupation cells and
chew an emergence hole, leaving only the integu-
ment of the bean. These opaque "windows" can be
seen readily and aid in the selection of beans with
3rd instars present. These jars were moved to the
C. hunter rearing room, which was maintained
under the same conditions as the C. maculatus
rearing room.
The beans were placed in trays (9 x 8 x 2 cm)
with 115-125 beans in each tray. The trays were
plastic strawberry baskets with the sides
trimmed to 2 cm high (Fig. la). Corks were glued
to the bottoms of the trays to elevate them, thus
allowing more accessibility of the female parasi-
toids to the beans on the bottoms of the trays. Ev-

Fig. 1. Strawberry basket (a) for exposing garbanzo
beans with 4th instar Callosobruchus maculatus in ovi-
position containers to adult Catolaccus hunter (b) in the

ery Monday, Tuesday, and Wednesday, two trays
were placed in each of two oviposition containers
consisting of No. 6 (2.8 liter) plastic jars (Newell
Rubbermaid Co., Wooster, OH) laid on their sides
(Fig. Ib). Water was provided by inserting two wa-
ter-filled, cotton-plugged 1-dram vials through
two 1.3-cm diameter holes in the upper surface of
each container. A cloth sleeve was attached to the
mouth of each container and was sealed with a
rubber band when not in use. Drops of honey were
placed on the inside top of the containers to pro-
vide food and were replenished when consumed
by the parasitoids. About 50 female and 50 male
parasitoids were introduced into each oviposition
container. The trays in the containers were
changed three times a week for 26 days, at which
time the oviposition containers were disassem-
bled and cleaned for re-use.
The beans that had been exposed 2-3 days to
parasitoids were placed in No. 3, 2.4-liter rectan-
gular, plastic containers (Newell Rubbermaid Co.,
Wooster, OH) with screen covered square holes
cut in the lid to allow ventilation but prevent es-
cape of emerging C. maculatus adults. The beans
were divided into three containers and each con-
tainer was placed individually in Plexiglas (Ato-
fina Chemicals, Inc., Philadelphia, PA) incubation
cages (30.5 x 30.5 x 30.5 cm) with a cloth sleeve on
one end. Two sides of the cage were covered with
organdy fabric to allow ventilation.
After about 7 d, adult parasitoids began to
emerge and were collected with a vacuum pump
aspirator. The garbanzo beans were sifted to re-
move C. maculatus adults. The beans were then
placed on a wax paper-lined fiberglass lunchroom
tray (45 x 35 cm), one layer deep and the trays
were placed on the shelves of an emergence box
(Fig. 2a). The emergence boxes were constructed
of wood and had 4-8 shelves with individual, seal-
able doors for each shelf. The shelves did not ex-
tend to the back of the emergence box and the
bean-filled trays were not placed on the shelf all
the way to the back. Thus, an open space was cre-
ated at the back of the box from the bottom to the
top. At the top of this open space, a hole (5 x 20
cm) was cut and covered with metal window
screen that allowed passage ofC. hunter but pre-
vented that of the C. maculatus adults. A Plexi-
glas collection chamber (32 x 32 x 21 cm) (Fig. 2b)
was attached to the top of the emergence box over
the screen-covered slot. The sides of the Plexiglas
box had cloth sleeves installed, allowing access
for collecting parasitoid adults with a vacuum
pump aspirator. Two water-filled, cotton-plugged
vials were placed in the bottom of the Plexiglas
box and honey was streaked on the inside of the
top and front. Both were replenished as needed.
Trays were replaced within the emergence box
every 23 days as new parasitoid-exposed, C. mac-
ulatus-infested beans were added. Once a week,
the Plexiglas box was thoroughly cleaned with

June 2005

Scientific Notes

Fig. 2. Emergence box (a) with tray of garbanzo
beans infested with 3'd instar Callosobruchus maculatus
that have been exposed to adult Catolaccus hunter. As-
pirating C. hunter adults from the Plexiglas collection
chamber (b).

Kimwipes tissue (Kimberly-Clark Corp., Roswell,
GA) moistened with water.
Approximately 18,000 parasitoids were pro-
duced weekly by these rearing methods with 26
oviposition cages. Start-up costs include about
$133 for 180 "Mason" jars for rearing the C. macu-
latus; about $150 for a humidifier to maintain RH
at 60% in the C. maculatus rearing room; about
$315 for 26 C. hunter oviposition cages including
plastic jars, vials, cotton balls, honey, plastic berry
baskets, corks, fabric (also used for incubation
cages), twine rope, and rubber bands; about $80
for each of 9 C. hunter larval incubation cages;
and about $155 for labor and supplies to build each
of three adult emergence cages. About 22 h/wk
were required in the maintenance of both the C.
maculatus and C. hunter colonies. About 4 kg of
black-eyed peas and 6 kg of garbanzo beans were
used each week. At $8/h for labor and $1.22/kg for
the peas and $0.90/kg for beans, the estimated re-
curring cost of production was about $186/wk.

Anecdotal observations have indicated that bi-
weekly releases of 1,500 C. hunter along one edge
of pepper fields of different sizes during the sum-
mer and fall fallow season resulted in reduced in-
festations of the pepper weevil on pepper during
the following spring season. In addition, experi-
mental evidence on an organic farm indicated
that weekly releases of the parasitoid at about
7,400/ha delayed the pepper weevil infestation
(Schuster, unpublished data). In experimental
plots, weekly releases of 1,500 C. hunter in night-
shade during the fallow, off-season followed by
weekly releases at 7,400/ha in adjacent pepper in
the spring resulted in 65-75% fewer pepper fruit
infested by the pepper weevil.
It is estimated that for an organic grower with
a 1-ha block to make releases of 1,500 C. hunter
adults every 2 wk for 32 wks (16 releases during
the fallow off-season) would cost about $250 in re-
curring expenses. To add additional releases of
7,400/ha would cost about another $76/wk during
the early pepper season. Neither of these cost es-
timates includes the cost of labor to release the
parasitoid adults. The estimated cost for fallow
season releases is probably cost effective but the
in-season costs may be prohibitive; however, in
discussions with organic producers, this latter
cost may not be prohibitive in light of few effective
alternatives for managing the pepper weevil. The
current rate of parasitism in the C. maculatus
host is about 40%. If the rate of parasitism could
be increased without increasing production costs,
the cost for releases of C. hunter for managing
the pepper weevil during the spring season could
become more cost effective.
This research was supported in part by USDA,
Special Research Grants, Tropical/Subtropical
Agricultural Research, and in part by the Florida
Agricultural Experiment Station and approved
for publication as Journal Series No. R-10607.


Methodology was developed to rear Catolaccus
hunter Crawford, a parasitoid of the pepper wee-
vil (Anthonomus eugenii Cano), on an alternative
host, the cowpea weevil (Callosobruchus macula-
tus F.) in temperature controlled rooms at 27C,
60% relative humidity and 14L:10D photoperiod.
Black-eyed peas, Vigna unguiculata (L.) Walp.,
were used to maintain a colony of C. maculatus,
and garbanzo beans, Cicer arietinum L., were
used to expose the C. maculatus larvae to C. hunt-
eri females. About 250 garbanzo beans containing
4th instar C. maculatus were exposed 48 to 72 h to
50 female and 50 male C. hunter. Parasitoid-ex-
posed beans were held for about 7 days and were
placed into emergence boxes with screened-cov-
ered slots, which retained C. maculatus adults in
the box but allowed C. hunter adults to pass into
a Plexiglas collection chamber. With an invest-

Florida Entomologist 88(2)

ment of about 22 h/wk, about 18,000 parasitoids
can be reared weekly at an estimated recurring
cost of $186/wk for labor and supplies. Start-up
costs for rearing containers and a rearing room
humidifier totaled about $1,800.


CATE, J. R. 1987. A method of rearing parasitoids of boll
weevil without the host plant. Southwestern Ento-
mol. 12: 211-215.
ELMORE, J. C., AND R. E. CAMPBELL. 1954. Control of
the pepper weevil. J. Econ. Entomol. 47: 1141-1143.
LEYVA VAZQUEZ. 2002. Efecto de diferentes factors
sobre la cria de Callosobruchus maculatus (Co-
leoptera: Bruchidae) para la producci6n de Catolac-
cus spp. (Hymenoptera: Pteromalidae). Acta Zool.
Mex 86: 67-101.
RILEY, D. G., AND E. G. KING. 1994. Biology and man-
agement of the pepper weevil Anthonomus eugenii
Cano (Coleoptera: Curculionidae): a review. Ento-
mol. (Trends in Agric. Sci.) 2: 109-121.
RILEY, D. G., AND D. SCHUSTER. 1992. The occurrence of
Catolaccus hunter, a parasitoid of Anthonomus eu-
genii, in insecticide treated bell pepper. Southwest-
ern Entomol. 17: 71-72.
ROBERSON, J. L., AND D. E. HARSH. 1993. Mechanized
production processes to encapsulate boll weevil lar-
vae (Anthonomus grandis) for mass production of
Catolaccus grandis (Burks), pp. 922-923 In D. J.
Horber and D. A. Richter [eds.], Proc. Beltwide Cot-

ton Conf., Vol. 2, National Cotton Council of Amer-
ica, Memphis, TN.
BARCENAS, AND G. W. ELZEN. 2000. Biology of Cato-
laccus hunter (Hymenoptera: Pteromalidae), a par-
asitoid of the pepper weevil and boll weevil
(Coleoptera: Curculionidae). Ann. Entomol. Soc.
Amer. 93: 862-868.
SALDANA, AND S. M. GREENBERG. 1998. Use of a facti-
tious host and supplemented adult diet to rear and in-
duce oogenesis in Catolaccus grandis (Hymenoptera:
Pteromalidae). Environ. Entomol. 27: 499-507.
SCHUSTER, D. J., AND P. H. EVERETT. 1982. Control of
the beet armyworm and pepper weevil on pepper.
Proc. Fla. State Hort. Soc. 95: 349-351.
KRING. 1988. Pepper weevil and sweetpotato white-
fly management on pepper. Univ. Fla., IFAS,
Bradenton GCREC Res. Rpt. BRA1988-19.
C. CRUZ, AND R. ZAPATA. 1996. Prospects for inte-
grated management of the pepper weevil in the Car-
ibbean basin, pp. 71-72 In D. N. Maynard [ed.], Proc.
Natl. Pepper Conf., 8-11 Dec. 1996. Naples, FL.
Influence of temperature and host on life history pa-
rameters of Catolaccus hunter (Hymenoptera: Pter-
omalidae). Environ. Entomol. 31: 354-360.
TOAPANTA, M. A. 2001. Population ecology, life history,
and biological control of the pepper weevil, Anthono-
mus eugenii (Coleoptera: Curculionidae). Ph.D. dis-
sertation. University of Florida, Gainesville.

June 2005

Scientific Notes


'Embrapa Recursos Gen6ticos e Biotecnologia, Parque Estacao Biol6gica Final da Av. W5 Norte
70770-900 Brasilia (DF), Brazil

2University of Florida, IFAS, Entomology and Nematology Department, Citrus Research and Education Center
700 Experiment Station Road, Lake Alfred, FL 33850

3University of Florida, Entomology and Nematology Department, P.O. Box 110620, Gainesville, FL 32611-0620

The flat mite, Brevipalpus phoenicis (Geijkes),
is a polyphagous pest found in many subtropical
and tropical regions of the world (Childers et al.
2003). Brevipalpus phoenicis is recognized as the
most economically harmful virus vector species
in citrus areas where Citrus leprosis Virus
(CiLV) has been reported. So far, chemical sprays
have been the main approach adopted to control
the mite in citrus (Rodrigues & Machado 2003).
Otherwise, fungal pathogens are frequently
found causing diseases and epizootics in mite
populations (Alves 1998; Van Der Geest et al.
2000). Recently, the potential to control Brevi-
palpus populations by spraying with the fungus
Hirsutella thompsonii Fisher was suggested
(Rosas-Acevedo & Sampedro 2000; Rossi 2002).
However, the development ofH. thompsonii as a
bio-acaricide has been hampered by difficulties
in mass producing aerial conidia. In contrast,
the entomopathogenic fungus Metarhizium
anisopliae var. acridum Driver & Milner is eas-
ily produced in large scale and has been devel-
oped as a mycoinsecticide in several countries
(Magalhaes et al. 2000; Lomer et al. 2001). This
fungus is tolerant of high temperatures, an im-
portant characteristic for pathogens developed
for tropical agroecosystems. We studied the
pathogenicity of M. anisopliae var. acridum
against B. phoenicis to evaluate its potential use
as an acaricide.
Mites used in this study were 5-15 days old
adult females derived from clonal lineage estab-
lished from a single female isolated from a mite
colony collected from citrus in Plant City, Flor-
ida (Rodrigues et al. 2004). The fungus assayed
was the isolate CG423 of M. anisopliae var. acri-
dum, produced on SDAY medium, harvested,
and dried according to Magalhaes & Boucias
(2004). The viability of the conidia used in the
bioassays was determined by plate assay to be
>95%. The bioassay unit was a 90-mm diameter
Petri dish containing a 60-mm diameter plate
glued to the bottom. A 1.5-cm diameter circular
leaf arena clipped from Ligustrum lucidum Ai-
ton was placed over a layer of cotton wool which
was saturated with water. Twenty mites were
transferred to the leaf arena. Two h later the

pathogen was applied at a concentration of 108
conidia/ml in a 200-pl suspension applied by air
flow with a pressure of 150 psi with the aid of a
spray tower (Pereira 1991). Mite survival was
monitored daily and infection confirmed by ex-
amining the specimens with an epifluorescence
microscope. Dead mites were stained with Cal-
cofluor White (Sigma) (1 mg/ml) for 30 min and
washed three times. To prevent rapid fading, the
stained material was immersed in a mounting
medium containing DABCO (Sigma) (100 mg),
glycerine (30 ml) and Hepes (10 ml) for micro-
scopic examination.
In preliminary experiments, the mites were
sprayed with fungus + 0.5% Tween 20 (v/v)
(polyoxyethylenesorbitan monolaurate, P-1379
(Sigma)) in deionized water, water + 0.5%
Tween 20, and control (no treatment). Each
treatment was repeated five times. Observa-
tions showed very low survival rates (<23%) 24
h after treating the mites with the fungus +
Tween 20 but also with the Tween 20 alone.
Otherwise, the mortality rates in the controls
(water treatment) were insignificant, indicating
acaricidal activity by the wetting agent. There-
fore, we tested the direct mortality to B. phoen-
icis with 0.5, 0.05, 0.005, 0.0005, and 0.00005%
of Tween 20 to determine a safe concentration
for subsequent studies. The results from a linear
regression confirmed that mite survival was de-
pendent on the Tween 20 concentration (P <
0.001) (Zar 1998). When Tween 20 was applied
at a concentration 0.5%, survival at 24 h after
inoculation was lower than 35%. In a separate
experiment, similar mortality was observed
when the leaf substrates were treated with
Tween 20 prior to transferring the test mites
(data not shown).
Survival of adult female mites treated with
0.00005% Tween 20 was not significantly differ-
ent from the control and was therefore used in the
fungal assays. The fungus was able to germinate
and penetrate the treated adult female mites
(Fig. 1). The SEM observations confirmed these
findings. At 8 days post-treatment, the fungus
caused 90% mortality (Fig. 2), where infection
levels were confirmed by fluorescence microscopy.

Florida Entomologist 88(2)

Fig. 1. (A) Microphotograph of Metarhizium anisopliae var. acridum infecting Breuipalpusphoenicis 3 days after
inoculation by fluorescence microscopy. (B, C, and D) SEM of infection and colonization process. E) Control sprayed
with water. (Cn = Conidium, Ap = Appressorium, Hf = Hypha).

Fungal infection also reduced the production
of eggs by B. phoenicis females in comparison to
treatments that received only Tween 20 or water
(Table 1). These results are important and sup-
port the possibility of exploiting M. anisopliae
var. acridum as a biological acaricide against
B. phoenicis. Additional research is needed to de-
termine the acaricidal effects of Tween 20 and
similar products for false spider mite control.
This research was supported by the Florida
Agricultural Experiment Station and EMBRAPA.
The manuscript was approved for publication as
Journal Series No. R-10508. We thank J. L. Cap-
inera and J. Stimac, University of Florida for pro-
viding facilities and laboratory space. Thanks to

J. L. Capinera and P. Inglis for reviews of this


Metarhizium anisopliae var. acridum Driver &
Milner isolate CG423 was demonstrated to be
pathogenic to the false spider mite Brevipalpus
phoenicis Geijskes (Acari: Tenuipalpidae). Effects
on mite survival and egg production were as-
sessed. The fungus was able to infect treated
adult mites at least 4 days after inoculation and
reached 90% mortality by the 8th day. We also
demonstrated that Tween 20 shows acaricidal ac-
tivity at low concentrations to B. phoenicis.

June 2005

Scientific Notes

Fungus + Tween 20

1 2 3 4 5 6 7 8

Tween 20

1 2 3 4 5 6 7 8








Fungus + water

1 2 3 4 5 6 7 8

D ---------.


1 2 3 4 5 6 7 8


Fig. 2. Survival curves of Brevipalpus phoenicis females treated with Metarhizium anisopliae var. acridum sus-
pended in Tween 20 and water. The fungus suspension was sprayed at concentration of 108 conidia/ml and
0.00005% Tween 20. Vertical bars = Standard Error.

zium anisopliae VAR. acridum. AVERAGE OF FIVE REPLICATES.

Treatment Females alive Eggs produced Eggs/Female (S.E.)

Water (W) 19.8 (+0.2) a* 17.2 (4.1) a* 0.87 (+0.21) a
Tween 20 (T) 18.6 (+0.8) a 9.64 (3.7) ab 0.51 (+0.20) ab
Fungus + W 17.0 (+0.4) a 4.2 (+0.2) b 0.25 (+0.01) b
Fungus + T 18.2 (+0.9) a 3.8 (+1.5) b 0.20 (+0.08) b

*Means followed by the same letter do not differ at P < 0.05; Tukey (a = 0.05); S.E. = Standard Error.


ALVES, S. B. 1998. Fungos entomopatog6nicos, pp. 289-
382 In S. B. Alves [ed.], Controle Microbiano de In-
setos. 2 ed. Piracicaba-SP: Fealq, V.1
BOURN. 2003. Host plants ofBrevipalpus californicus,
B. obovatus, and B. phoenicis (Acari: Tenuipalpidae)
and their potential involvement in the spread of one
or more viral diseases vectored by these mites. Exp.
Appl. Acarol. 30: 29-105.
WALD, AND M. THOMAS. 2001. Biological control of lo-
custs and grasshoppers. Annu. Rev. Entomol. 46:
MAGALHAES, B. P. AND D. BOUCIAS. 2004. The effects of
drying on the survival ofMetarhizium anisopliae var.
acridum conidiospores. J. Orthoptera Res. 13(1): 155-

SCHMIDT, AND W. D. GUERRA. 2000. Field trial with
the entomopathogenic fungus Metarhizium anisopliae
var. acridum against bands of the grasshopper
Rhammatocerus schistocercoides in Brazil. Biocon-
trol Sci. Technol. 10: 427-441.
PEREIRA, R. M. 1991. Evaluation of the entomopatho-
genic fungus Beaveria bassiana on the red imported
fire ant, Solenopsis invicta. Ph.D. Thesis. University
of Florida. 135 pp.
ROSSI, L. S. 2002. Selecio de fungos entomopatog6nicos
e infeccao de Hirsutella sp. em Brevipalpus phoenicis
(Geijskes, 1939). M.S. Dissertation. University of
Sao Paulo (ESALQ/USP). 92 pp.
RODRIGUES, J. C. V., AND M. A. MACHADO. 2003. Virus-
Brevipalpus-plant relationships on citrus leprosis
pathosystems, pp. 768-770 In J. L. Albrigo [ed.],
Proc. Int. Soc. Citriculture Congr. 2000, Orlando,
FL, Dec. 3-7. Vol. II.







Florida Entomologist 88(2)

C. CHILDERS, AND B. J. ADAMS. 2004. Mitochondrial
DNA and RAPD polymorphisms in the haploid mite
Brevipalpus phoenicis (Acari: Tenuipalpidae). Exp.
Appl. Acarol. (in press).
biol6gico de Brevipalpus spp. en Citrus aurantifolia

en Guerrero, M6xico. Man. Int. de Plagas (Costa
Rica) 55: 56-59.
WER, AND E. A. M. BEERLING. 2000. Diseases of
mites. Exp. Appl. Acarol. 24: 497-560.
ZAR, J. H. 1998. Biostatistical Analysis (4 ed.). Prentice
Hall. 929 pp.

June 2005

Scientific Notes


Entomology and Nematology Department, University of Florida, Gainesville, FL 32611-0630

Smetana (1982) described a new genus (Oxy-
bleptes) to include four species of Xantholinini from
America north of Mexico. One species, 0. davisi
(Notman), was previously known from the District
of Columbia, New Jersey, and New York. The re-
maining three species (0. kiteleyi Smetana, 0. hat-
chi Smetana, and 0. pusio Smetana) were de-
scribed as new. None of these species was reported
to occur farther south than North Carolina.
In 1984, J. H. Frank submitted to Ales Smet-
ana (Canadian National Collection, Ottawa, On-
tario) North American specimens of Xantholini-
nae that either could not be identified using Smet-
ana's (1982) keys, or gave evidence of range exten-
sions. Thereafter, Smetana (1988) described a new
species (0. meridionalis Smetana) based upon five
specimens. One paratype was retained by the Ca-
nadian National Collection (CNC), the holotype
and one paratype were in 1987 deposited in the
Florida State Collection of Arthropods (FSCA),
and two paratypes were retained in the collection
of J. H. Frank. All were from the grounds of what
is now called Florida Medical Entomology Labora-
tory (of the University of Florida), at Vero Beach,
Indian River County, Florida, a location a few
hundred meters west of the Indian River (intra-
coastal waterway) on Florida's Atlantic coast, and
at =2736'N. Four were collected in August 1973
and one in May 1976, in oak-palm hammock. The
collection method was unusual: Frank had been
operating 5-gallon (=20-liter) green plastic water-
filled tubs to attract ovipositing Culex mosquitoes.
Four Oxybleptes specimens were collected drowned
or drowning from the surface of the water in the
tubs, and one alive from the conical lid with which
half of the tubs at any time were fitted. This lid
was of light-reflecting sheet aluminum. The
attractant (if there really was an attractant) for
the beetles may have been the reflectance of the
water surface or the aluminum. Although Frank
routinely ran an ultraviolet light trap, and later a
Malaise trap, at this and other nearby locations,
no specimens of Oxybleptes were caught in those
traps. The habitat and behavior of these curious
little beetles remained unresolved. All five speci-
mens collected were males. Smetana (1988),
therefore, was able to describe only the male of
this species, and he found it to be bicolored, like
that of 0. davisi and contrasting with males of the
three other known species.
In 2002, J. L. Foltz and D. T. Almquist operated
various kinds of traps for insects at Bee Island
in the Myakka River State Park, 2715.12'N,

8215.09'W. Most of the park is in Sarasota
County, Florida, but Bee Island is in Manatee
County. This park is within 25 km of Florida's Gulf
of Mexico coast. In November 2002, Almquist
asked Frank to identify staphylinid specimens col-
lected in pine flatwoods at Bee Island. Frank rec-
ognized some of them as being 0. meridionalis, ex-
plained their apparent rarity, and urged Almquist
to collect more, alive if possible for behavioral
study. The initial collection in the park was on 16
October, and included eight specimens (7 male, 1
female). The second collection was on 18 December
and included 18 specimens (13 male, 5 female).
Again, the collection method was curious. All
specimens were collected in water, with the Octo-
ber 2002 collection from the water-filled, white
plastic top of a Lindgren funnel trap (Phero Tech,
Inc., Delta, BC, Canada), and the December collec-
tion from plastic shoe boxes with a few cm of soapy
water. The collections again suggested attraction
to a reflective surface. All collections yielded a pre-
ponderance of males (for 1973-1976, 5:0 males:
females, October 2002, 7:1, December 2002, 13:5)
but the significance of this imbalance in sex ratio
is unclear. Adult activity is known to occur in May,
August, October, and December. Eggs and larvae
were not obtained, nor was any behavioral infor-
mation obtained except that Foltz remembers that
the specimens collected on 18 December were
trapped in the early afternoon, thus in daylight.
The female (Fig. 1), similar in size and coloration
to male, is "bicolored", testaceous to rufotestaceous;
the head and elytra are piceous to piceous black,
and the abdominal apex darkened. The antenna of
female with less swollen apical antennomere and
with antennomere III distinctly shorter than II
(Fig. 2a) cf. as long as II in male (Fig. 2b). The max-
illary palpus of the female (Fig. 2c) has all pal-
pomeres shorter and therefore relatively broader
than in male (Fig. 2d). The meso- and metatarsus
are shorter than the respective tibia (cf. as long in
male), and tarsomeres II-IV are relatively shorter
than in male. Such sexual dimorphism occurs also
in other species of the genus (Smetana 1982). Fe-
male specimens will be deposited in FSCA and
CNC together with additional male specimens.
Heretofore there is no record of any species of
Oxybleptes other than 0. meridionalis in Florida.
Among staphylinids collected by C. W. O'Brien
was one female specimen of Oxybleptes labeled
USA, Florida, Franklin Co., 3 mi NW of Alligator
Point, 17-IV-1974, Berlese funnel extract of pine
litter, O'Brien and Marshall. This specimen is not

Florida Entomologist 88(2)

Fig. 1. Habitus of female Oxybleptes meridionalis.
The scale line represents 1.0 mm.

distinctly bicolored, is considerably larger than
the diminutive 0. meridionalis, and most closely
matches the description of 0. kiteleyi by Smetana
(1982). Thus, in the northwest of Florida, distant
by about 380 km from the nearest known location
for 0. meridionalis and in a harsher climatic zone,
seems to exist a population perhaps of 0. kiteleyi.
If this identification is correct, it would represent
the southernmost record for 0. kiteleyi, which un-
til now had not been known south of North Caro-
lina. Smetana (1982) noted the difficulty of distin-
guishing females of 0. kiteleyi from those of

C. N D. NI
Fig. 2. Oxybleptes meridionalis, 2a. Antenna of fe-
male, 2b of male, 2c maxillary palpus of female, 2d of
male. The scale line represents 0.25 mm.

0. davisi, so there is doubt in this determination,
and it would be advantageous to examine male
specimens from northwestern Florida.
This is Contribution Number 56 of the National
Fire and Fire Surrogate Network Project, and was
supported by funding from the U.S. Joint Fire Sci-
ence Program. We thank C. W. O'Brien (formerly of
Florida A.&M. University) for the gift of one Oxy-
bleptes specimen (cf. 0. kiteleyi). We thank M. C.
Thomas and P E. Skelley for critical comments on
a manuscript draft. This is Florida Agricultural
Experiment Station Journal Series No. R-10491.

Oxybleptes meridionalis Smetana (Staphylin-
idae: Staphylininae: Xantholinini) was previously
known only from five males collected in 1973-1976
from Indian River County, Florida, at 2736'N. It is
here reported from Manatee County, on the other
side of the Florida Peninsula, at 2715'N. The new
collections include six females together with 20
males. The female is colored similarly to the male,
but differs in the structure of antennomeres III
and XI, the maxillary palpi, and the meso- and
metatarsi, which are here described. One female
specimen of another species of Oxybleptes, of un-
certain identity, was collected in Franklin County,
Florida, some 380 km farther north.

SMETANA, A. 1982. Revision of the subfamily Xantholini-
nae ofAmerica north of Mexico (Coleoptera: Staphylin-
idae). Mem. Entomol. Soc. Canada 120: i-viii, 1-389.
SMETANA, A. 1988. Revision of the subfamily Xantholin-
inae of America north of Mexico (Coleoptera: Sta-
phylinidae) Supplementum 1. Canadian Entomol.
120: 525-558.

June 2005

r ,'

Scientific Notes


University of Florida, Institute of Food and Agricultural Sciences
Department of Entomology and Nematology, Fort Lauderdale Research and Education Center
3205 College Avenue, Fort Lauderdale, FL 33314, U.S.A.

Two non-endemic and highly destructive spe-
cies of Coptotermes Wasmann occur in Florida. In
1980, the Formosan subterranean termite, Copto-
termes formosanus Shiraki, was first discovered
in Florida infesting condominiums along the
Atlantic Ocean and the Intracoastal Waterway in
Broward County (Anonymous 1980; Koehler 1980).
Later, C. formosanus was found in neighboring
Dade Co. and in central Florida (Orange Co.) and
in the western Panhandle (Escambia Co.) (Thomp-
son 1985). A 1987 survey of structure-infesting
termites of Florida (Scheffrahn et al. 1988)
showed a low incidence ofC. formosanus in urban
areas of the peninsula compared with other pest
species. Based on published reports and personal


0 C. formosa
@ C. formosa

communications but no voucher specimens,
Woodson et al. (2001) added Okaloosa, Hillsbor-
ough, Santa Rosa, Palm Beach, Marin, Citrus,
and Leon as additional Florida counties where
C. formosanus is distributed.
The Asian subterranean termite, Coptotermes
gestroi (Wasmann), was first discovered in Florida
in 1996 infesting a storefront in Miami (Su et al.
1997). Originally classified as C. havilandi
Holmgren, this species was recently synonymized
with C. gestroi by Kirton & Brown (2003). Since
1996, no additional reports have been published
on the distribution of C. gestroi in Florida. Copto-
termes formosanus and C. gestroi can be differen-
tiated from each other by soldier (Scheffrahn et al.

e 'J sonville

crystal 6De

nus Orlando
Ts shiprne .



Jupiter a
S Cape Coral
nus on land Y ngs
nus shipborne ns FL3 Laud.

O C. gestroi on land
B C. gestroi shipbome
* urban areas

Fig. 1. Distribution of Coptotermes formosanus and C. gestroi in Florida. Inset on left shows greater detail of the
southeast coast. Grey areas represent urbanized land zones.

Florida Entomologist 88(2)

1990) and imago (Scheffrahn & Su 2000) morphol-
ogy. Both species separate into different clades
when sequences of their respective mtDNA 16s
RNA genes are compared (Scheffrahn et al. 2004).
Over the last 19 years, we have directly col-
lected or have received more than 3,000 termite
colony samples from throughout Florida includ-
ing those of the two now well established Copto-
termes species. We herein report on the current
distribution of Coptotermes in Florida based on
locality records of 168 C. formosanus and 35
C. gestroi samples. All samples are cataloged and
housed in the University of Florida Termite Col-
lection, Ft. Lauderdale Research and Education
Center, Ft. Lauderdale, Florida.
The geographical distribution of Coptotermes
in Florida, including incidences where these spe-
cies have been collected aboard boats or ships, is
mapped in Fig. 1. Most urban centers throughout
Florida, with the exception of Pinellas Co. (St.
Petersburg), now support populations of C. formo-
sanus. Coptotermes localities in Florida are listed
in Table 1 by county, city, and year of discovery.
Including shipborne infestations, C. formosanus
has been collected from 20 counties and 40 cities
in Florida, while C. gestroi has been collected in
4 counties and 8 cities (Table 1). All populations
of C. gestroi are currently restricted to tropical
southeastern Florida. Dade and Broward Coun-
ties, Florida, are the only geographies worldwide
where these two species have sympatric distribu-
tions. Both species are exclusively synanthropic
in Florida and have only been collected in or
within foraging distance of a structure.
The tendency of both C. formosanus and
C. gestroi to colonize boats (<40-m-long) and ships
(>100-m-long) likely contributed to the dispersal
of these and other Coptotermes species from their
other ranges to exotic localities, such as Florida
(Scheffrahn et al. 2004). Colonies have been ob-
served reaching maturity aboard watercraft and
dispersal flights from watercraft could initiate
land-based infestations near dockage.
One particular infestation is worth reporting
here because it provides compelling evidence for
shipboard establishment and movement of Copto-
termes colonies over long distances. In January
2001, a 29 meter-long yacht docked off the Intra-
coastal Waterway in Ft. Lauderdale was found to
be infested with C. gestroi. Since 1993, the yacht's
winter dockage was at the Turtle Cove Marina in
Providenciales, Turks and Caicos Islands, British
West Indies. Providenciales is some 930 km
southeast of Ft. Lauderdale. Coptotermes gestroi
was first collected in the Turks and Caicos at Tur-
tle Cove in 1990 (R. Scheffrahn, unpubl.). It is
very probable that during one or more preceding
winters in Providenciales, lighting on the yacht
attracted alates during a nocturnal dispersal
flight in the Turtle Cove vicinity (C. gestroi dis-
persal flights in Florida and the West Indies occur


County City Year

Coptotermes form





Palm Beach

Santa Rosa

Dania Beach
Fort Lauderdale
Hillsborough Beach'
Lighthouse Point'
Pembroke Pines
Pompano Beach'
Wilton Manors
Crystal River
Marco Island
Florida City
North Miami Beach
Temple Terrance
Bonita Springs
Cape Coral
Holms Beach'
Jensen Beach
Boca Raton
Lake Park
North Palm Beach
Palm Beach Gardens
Riviera Beach
West Palm Beach

Coptotermes gestroi




St. Lucie

Fort Lauderdale'
Miami Beach
Key West
Stock Island
Fort Pierce'




Collected from boat only.
Collected previously from ship in 1999.

June 2005

Scientific Notes

at night from January to March). Dealates then
proceeded to colonize the boat. At least one colony
became established on board and subsequently
grew until the infestation was detected and the
boat fumigated in Ft. Lauderdale before return-
ing to Providenciales.
It is likely that shipboard infestations will con-
tinue to contribute to the intrastate spread of
Coptotermes in Florida, where the overwhelming
number of infestations are within one kilometer
of marine boat dockage. Eighteen C. formosanus
and 3 C. gestroi samples in our collection data-
base were taken aboard boats and ships. In
inland locations, such as the Orlando and Talla-
hassee areas, land-based commodities such as
railroad ties and landscape materials harboring
incipient colonies may have served as vehicles of
We thank the many pest control professionals
and others that have collected and submitted
samples that were included herein, including
Paul Ban, Bob Benham, Ron Box, Lyle Buss, Gabe
Centeno, Jim Chase, Mary Cohen, Bruce and Jeff
Edwards, Louis Giacone, Terry Harper, Jim
Maler, John Mangold, Bruce Ryser, Jeff Stotts,
and Mark Weinberg. William H. Kern Jr. and
Brian J. Cabrera reviewed this contribution R-
10474 of the University of Florida Agricultural
Experiment Station Journal Series.


Confirmed Florida localities of the Formosan
subterranean termite, Coptotermes formosanus,
and the Asian subterranean termite, C. gestroi,
are reported. Coptotermes formosanus has been
collected from 20 counties and 40 cities in Florida,
while C. gestroi has been collected in 4 counties
and 8 cities. Dispersal of both Coptotermes species
have been facilitated by shipboard infestations
and land-based commodities.


ANONYMOUS. 1980. Formosan termites now in Florida.
Pest Control Magazine, November: 20, 113.
KIRTON, L. G., AND V. K BROWN. 2003. The taxonomic
status of pest species of Coptotermes in Southeast
Asia: Resolving the paradox in the pest status of the
termites, Coptotermes gestroi, C. havilandi and
C. travians (Isoptera: Rhinotermitidae). Sociobiol-
ogy 42: 43-63.
KOEHLER, P. G. 1980. The Formosan subterranean ter-
mite. Florida Coop. Ext. Service, Univ. Florida Inst.
Food Agric. Sci. circular ENT-51.
SCHEFFRAHN, R. H., AND N.-Y. SU. 2000. Asian subter-
ranean termite, Coptotermes gestroi (= havilandi)
(Wasmann) (Insecta: Isoptera: Rhinotermitidae).
2000. Retrieved 13 October 2004 from University of
Florida, Department of Entomology and Nematology
"Featured Creature" Web site: http://creatures.ifas.
survey of structure-infesting termites of peninsular
Florida. Florida Entomol. 71: 615-630.
SCHEFFRAHN, R. H., N.-Y. SU, AND B. DIEHL. 1990. Na-
tive, introduced, and structure-infesting termites of
the Turks and Caicos Islands, B.W.I. Florida Ento-
mol. 73: 622-627.
AUSTIN, AND J. NIXON. 2004. Establishment of the
African termite, Coptotermes sjostedti (Isoptera:
Rhinotermitidae), on the Island of Guadeloupe,
French West Indies. Ann. Entomol. Soc. America 97:
A new introduction of a subterranean termite, Copto-
termes havilandi Holmgren (Isoptera: Rhinotermiti-
dae) in Miami, Florida. Florida Entomol. 80: 408-411.
THOMPSON, C. R. 1985. Detection and distribution of
Formosan termites (Isoptera: Rhinotermitidae) in
southeastern Florida. J. Econ. Entomol.78: 528-530.
WOODSON, W. D., B. A. WILTZ, AND A. R. LAX. 2001. Cur-
rent distribution of the Formosan subterranean ter-
mite (Isoptera: Rhinotermitidae) in the United
States.Sociobiology 37: 661-671.

Florida Entomologist 88(2)


University of Florida, Department of Entomology and Nematology, Gainesville, FL 32611

The relationship between mouthpart structure
and diet has been known for years. This connec-
tion between mouthpart morphology and specific
food types is incredibly pronounced in the class In-
secta (Snodgrass 1935). As insects have evolved
and adapted to new food sources, their mouthparts
have changed accordingly. This is an extremely im-
portant trait for evolutionary biologists (Brues
1939) as well as systematists (Mulkern 1967).
Isley (1944) was one of the first to study grass-
hopper mouthparts in detail. He described three
groups of mandibles according to general struc-
ture and characteristic diet. These three groups,
still used today, were graminivorous (grass-feed-
ing type) with grinding molars and incisors typi-
cally fused into a scythe-like cutting edge, for-
bivorous (forb or broadleaf plant-feeding type)
which have a molar region consisting of a depres-
sion surrounded by raised teeth and sharp inter-
locking incisor teeth, and herbivorous (mixed-
feeding type) that have characteristics of both of
the aforementioned groups. The original findings
by Isley (1944) have since been proven to be wide-
spread in grasshoppers. Additional studies have
been conducted by Snodgrass (1928), Gangwere
(1965, 1966), Gangwere et al. (1976), and Patter-
son (1984) in North America; Lieberman (1968)
and Gangwere & Ronderos (1975) in South Amer-
ica; Williams (1954), Kaufmann (1965), and
Gangwere & Morales (1973) in Europe; Gangwere
& Spiller (1995) and Gangwere et al. (1998) in the
Mediterranean islands; Feroz & Chaudhry
(1975), Gapud (1968), and Kang et al. (1999) in
Asia; and Chapman (1964) in Africa.
The relationship between grasshopper mouth-
parts and food is far from precise. Mulkern (1967)
was convinced that only the grossest determina-
tions could be made between mandibular struc-
ture and diet (i.e., graminivorous, forbivorous, and
herbivorous). Occasionally, grasshoppers with
forb-feeding mandibles regularly feed on grasses
or vice versa (Chapman 1964). Nevertheless, there
is some value in assessing mouthpart structure
relative to predicting diet and habitat of grasshop-
pers, especially for the many rare or non-economic
species that are unlikely to be studied in detail.
Thus, the morphological characteristics and struc-
tural adaptations of the mouthparts of 36 of the 71
grasshoppers occurring in Florida were examined.
Grasshoppers were collected from various habi-
tats throughout north-central Florida in 2001 and
2002. Thirty-six of the most common Floridian
grasshopper species were identified with the tax-
onomic key found in Smith et al. (2004) and fro-

zen until examination. Mandibles were removed
from thawed specimens by lifting the labrum and
pulling out each mandible separately with for-
ceps. Only young adults were used in an effort to
avoid confusion of mandible type due to mandible
erosion (Chapman 1964; Uvarov 1977). An exam-
ple of moderate erosion can be seen in Figure 1 (I).
This process was replicated with 10 individuals
from each species. After air-drying, each mandi-
ble was glued to the head of a #3 or #2 insect pin,
depending on its size, for easier manipulation,
and examined microscopically.
We used Isley's (1944) description of mandible
types and their adaptive functions to divide the
mandibles into 3 major categories: forbivorous
(forb-feeding), graminivorous (grass-feeding),
and herbivorous (mixed-feeding).
Mandibles were lightly brushed with 80 per-
cent ETOH and distilled water in an effort to re-
move most of the sand and debris adhering to the
mouthparts. Photographs were taken with the
Syncroscopy Auto-Montage system (University of
Florida, Entomology and Nematology Dept.).
The mandible structure of 36 species of grass-
hopper, from five subfamilies (Acridinae, Cyrta-
canthacridinae, Gomphocerinae, Oedipodinae,
and Romaleinae), found in Florida was micro-
scopically examined. These grasshoppers were
collected from a variety of habitats including dis-
turbed freshwater marsh, high pine, swamp, and
oak hammocks. All grasshoppers had distinctive
mouthparts that could be described as forbivo-
rous (forb-feeding type), herbivorous (mixed-feed-
ing type), or graminivorous (grass-feeding type)
(Fig. 1, A-L). A list of each species studied and the
mandible type is given in Table 1.
Of the subfamilies examined, the Cyrtacan-
thacridinae demonstrated the most diversity in
mandible type; however, most of them displayed
either herbivorous or forbivorous mandibles, indi-
cating a tendency toward forb-feeding. These
grasshoppers can be found in a wide range of hab-
itats, usually in dense vegetation or woodland ar-
eas, and are quite active in both walking and fly-
ing. It is interesting to note that both the grass-
hoppers in this subfamily that did display
graminivorous type mandibles (L. marginicollis
and S. vitreipennis) also have extremely slender,
elongated bodies and can be found on the edges of
ponds or in freshwater marshes (Isley 1944; Squi-
tier & Capinera 2002b; Smith & Capinera 2005).
These grasshoppers typically grasp the stems of
emergent grass or grass-like vegetation such as
sedges or cattails, blending in almost perfectly.

June 2005

Scientific Notes



Fig. 1. Mandibles ofAmblytropidia mysteca, a representative graminivorous species: right incisor region (A), left
incisor region (B), right molar region (C), and left molar region (D); Schistocerca ceratiola, a representative forbiv-
orous species: right incisor region (E), left incisor region (F), right molar region (G), and left molar region (H); and
Spharagemon cristatum, a representative herbivorous species: right incisor region (I), left incisor region (J), right
molar region (K), and left molar region (L).

Florida Entomologist 88(2)


Graminivorous type mandibles Herbivorous type mandibles Forbivorous type mandibles

Acridinae Cyrtacanthacridinae Cyrtacanthacridinae
Metaleptea breuicornis (Johannson) Gymnoscirtetespusillus Scudder Aptenopedes aptera Scudder
Melanoplus bispinosus Scudder Aptenopedes sphenarioides Scudder
Cyrtacanthacridinae Melanoplus sanguinipes (Fabricius) Melanoplus keeleri (Thomas)
Leptysma marginicollis (Serville) Schistocerca alutacea (Harris) Melanoplus ordwayae Deyrup
Stenacris vitreipennis (Marschall) Schistocerca americana (Drury) Melanopluspropinquus Scudder
Schistocerca obscura (Fabricius) Melanoplus punctulatus Scudder
Gomphocerinae Schistocerca rubiginosa (Scudder) Melanoplus querneus Rehn & Hebard
Achurum carinatum (F. Walker) Melanoplus rotundipennis Scudder
Amblytropidia mysteca (Saussure) Oedipodinae Paroxya atlantica Scudder
Dichromorpha viridis (Scudder) Chortophaga australior (Rehn & Hebard) Paroxya clavuliger (Serville)
Eritettix obscurus (Scudder) Spharagemon crepitans (Saussure) Schistocerca ceratiola Hubbell & Walker
Mermiria intertexta Scudder Spharagemon cristatum (Scudder) Schistocerca damnifica (Saussure)
Orphulella pelidna (Burmeister)
Syrbula admirabilis (Uhler) Romaleinae
Romalea microptera (Beauvois)
Arphia granulata (Saussure)
Hippiscus ocelote (Saussure)
Pardalophora phoenicoptera (Burmeister)

However, the overwhelming majority of these
grasshoppers display either herbivorous or for-
bivorous mandibles (Isley 1944; Gangwere 1965,
1966; Squitier & Capinera 2002a).
The Oedipodinae were split between two man-
dible types: graminivorous and herbivorous. This
signifies a more grass-dominated diet. However,
these grasshoppers are much more divergent and
some may be completely graminivorous or forbivo-
rous. Most of the species in this subfamily were
found on the ground in open areas on bare soil,
rarely on plants or grasses. As a general rule, the
Oedipodinae show the most mandible diversity of
all the grasshopper subfamilies. Isley (1944),
Gangwere (1966), and Kang et al. (1999) found a
fairly even distribution of the three mouthpart
types in this group.
The gomphocerinae all had graminivorous
mandibles, indicating a consistent diet of grasses.
These findings are reinforced by the preferred
habitats of this subfamily, usually open grassy
fields and pastures. Virtually all Gomphocerinae
are graminivorous (Lockwood et al. 1994), or at
least have graminivorous type mandibles. Occa-
sionally a gomphocerine will display graminivo-
rous type mandibles but feed entirely on forbs
(Gangwere & Morales 1973); however, these are
rare exceptions. In almost every study carried out
on orthopteran mouthpart morphology, the Gom-
phocerinae display graminivorous type mandi-
bles (Isley 1944; Gangwere 1965, 1966; Lockwood
et al. 1994; Kang et al. 1999).
Due to only one representative species from
both the subfamilies Acridinae and Romaleinae,
determination of the mandibular morphology of

these subfamilies was limited. The Acridinae are
typically considered to be grass-feeders, display-
ing the classic graminivorous type mandibles
(Chapman 1964; Isley 1944). Very rarely a species
in this subfamily will display herbivorous type
mandibles (Chapman 1964). The Romaleinae are
always forb feeders and always display forbivo-
rous type mandibles (Isley 1944; Squitier & Cap-
inera 2002a; Smith & Capinera 2005).
Many thanks to David Almquist for help in
taking photographs with the auto-montage sys-
tem. This research was supported by the Florida
Agricultural Experiment Station, and approved
for publication as Journal Series No. R-10456.


Mouthpart consistency within subfamilies in-
dicates that evolution is just as important as eco-
logical factors in determining food plants; for
most subfamilies there is a strong association
with a particular form of vegetation. It is evident
that the ability, or tendency, of grasshoppers to
change hosts is partly limited by the structure of
their mandibles. However, because there are ex-
ceptions to the strong association of cyrtacan-
thacridines with forbs, and gomphocerines with
grasses, we see evidence that behavioral plastic-
ity or ecological opportunism is present even in
relatively primitive taxa such as Orthoptera.


BRUES, C. T. 1939. Food, drink, and evolution. Science
90: 145-149.

June 2005

Scientific Notes

CHAPMAN, R. F. 1964. The structure and wear of the
mandibles in some African grasshoppers. Proc. Zool.
Soc. London 142: 107-121.
CHAPMAN, R. F. 1966. The mouthparts of Xenocheila
zarudnyi. J. Zool. 148: 277-288.
FEROZ, M., AND M. A. CHAUDHRY. 1975. Studies on man-
dibles of some grasshoppers of Lahore. Biologia
(Lahore) 21: 211-225.
GANGWERE, S. K. 1965. Food selection in the Oedi-
ponidae grasshopper Arphia sulphurea. American
Midi. Nat. 74: 67-75.
GANGWERE, S. K. 1966. Relationships between mandi-
bles, feeding behavior and damage inflicted on plants
by the feeding of certain acridids (Orthoptera). Mich-
igan Entomol. 1:13-16.
1976. The food habits and Biology ofAcrididae in an
old-field community in Southeastern Michigan.
Great Lakes Entomol. 9: 83-123.
AND R. G. BLAND. 1998. Food selection and feeding
behavior in selected Acridoidea (Insecta: Orthoptera)
of the Canary Islands, Spain. J. Orth. Res. 7: 1-21.
GANGWERE, S. K., AND A. MORALES. 1973. Food selection
and feeding behaviour in Iberian Orthopteroidea.
An. Inst. Nac. Invest. Agrar, Ser. Prot. Veg. Num. 3:
GANGWERE, S. K., AND R. A. RONDEROS. 1975. A synop-
sis of food selection in Argentine Acridoidea. Acrida
4: 173-194.
GANGWERE, S. K., AND D. O. SPILLER. 1995. Food selec-
tion and feeding behavior in selected Orthoptera
sen. lat. of the Balearic Islands, Spain. J. Orth. Res.
4: 147-160.
GAPUD, V. P. 1968. The external morphology of the head
and mouthparts of some Philippine Orthoptera.
Philippine Entomol. 1: 11-32.
ISLEY, F. B. 1944. Correlation between mandibular mor-
phology and food specificity in grasshoppers. Ann.
Entomol. Soc. Am. 37: 47-67.
KANG, L., Y. GAN, AND S. L. LI. 1999. The structural ad-
aptation of mandibles and food specificity in grass-
hoppers on Inner Mongolian grasslands. J. Orth.
Res. 8: 257-269.

KAUFMANN, T. 1965. Biological studies on some Bavar-
ian Acridoidea (Orthoptera), with special reference
to their feeding habits. Ann. Entomol. Soc. Am. 58:
LIEBERMANN, J. 1968. The mandibles of grasshoppers of
the subfamily Chilacridinae. Rev. Invest. Agropecu.,
Patol. Veg. Ser. 5: 53-62.
IAM. 1994. Comparison of grasshopper (Orthoptera:
Acrididae) ecology on the grasslands of the Asia
steppe in Inner Mongolia and the Great Plains of
North America. J. Orth. Res. 2: 4-14.
MULKERN, G. B. 1967. Food selection by grasshoppers.
Annu. Rev. Entomol. 12: 59-78.
PATTERSON, B. D. 1984. Correlation between mandibu-
lar morphology and specific diet of some desert
grassland Acrididae (Orthoptera). American Midl.
Nat. 111: 296-303.
Key to the grasshoppers (Othoptera: Acrididae) of
Florida. Florida Entomol. 87: 537-550.
SMITH, T. R., AND J. L. CAPINERA. 2005. Host prefer-
ences and habitat associations of some Florida
grasshoppers (Orthoptera: Acrididae). Environ. En-
tomol. 34: 210-224.
SNODGRASS, R. E. 1928. Morphology and evolution of
the insect head and its appendages. Smithsonian
Misc. Coll. 81: 1-158.
SNODGRASS, R. E. 1935. Principles of Insect Morphology.
McGraw-Hill. New York. 667 pp.
SQUITIER, J. M., AND J. L. CAPINERA. 2002a. Host selec-
tion by grasshoppers (Orthoptera: Acrididae) inhab-
iting semi-aquatic environments. Florida Entomol.
85: 336-340.
SQUITIER, J. M., AND J. L. CAPINERA. 2002b. Habitat as-
sociations of Florida grasshoppers (Orthoptera: Acri-
didae). Florida Entomol. 85: 235-244.
UVAROV, B. 1977. Grasshoppers and locusts. A Hand-
book of General Acridology, Vol. 2, Cambridge,
United Kingdom. 613 pp.
WILLIAMS, L. H. 1954. The feeding habits and food pref-
erences of Acridinae and the factors which deter-
mine them. Trans. R. Entomol. Soc. London 105:

Florida Entomologist 88(2)


Subtropical Agricultural Research Center, USDA-ARS, 2413 E. Highway 83, Weslaco, TX 78596

Ionizing irradiation is used in Florida and Ha-
waii to disinfest several fruits and sweetpotatoes
of fruit flies (Diptera: Tephritidae) or other insects
(Hallman 2004a). Importation of fruit irradiated
against 11 fruit fly species and the mango seed
weevil, Cryptorhynchus mangiferae (F.) (Coleop-
tera: Curculionidae), has been approved (APHIS
2002). The treatment shows promise for wide-
spread implementation, as it is safe, broadly effica-
cious, accepted by consumers, cost-effective, may
be applied after packing, and widely tolerated by
fresh agricultural commodities (Hallman 2002).
A large body of research has determined mini-
mum absorbed doses of ionizing radiation to pre-
vent development or reproduction for many differ-
ent species of arthropods iaea.org/ididas/start.htm>. Some of this research
has shown that hypoxia can reduce some of the
detrimental effects of irradiation to insects used
in sterile release programs. The effect of hypoxia
on irradiation disinfestation treatment efficacy
has not been studied until recently, although
some agricultural commodities are stored under
hypoxic conditions, a strategy that is increasing in
application. Hallman (2004a) found that while no
oriental fruit moth, Grapholita molesta (Busck)
(Lepidoptera: Tortricidae), fifth instars developed
to the adult stage when irradiated at 200 Gy in
ambient atmosphere, 5.3% of those irradiated in
hypoxic atmospheres developed to the adult.
Hypoxia caused a small increase in the ability of
apple maggot, Rhagoletis pomonella (Walsh)
(Diptera: Tephritidae), third instars to emerge as
adults after irradiation (Hallman 2004b).
Plum curculio, Conotrachelus nenuphar
(Herbst) (Coleoptera: Curculionidae), is native to
the eastern Neartic and is a quarantine pest of
stone and pome fruits exported from the United
States and Canada. Hallman (2003) determined
that a minimum absorbed dose of 92 Gy (the max-
imum recorded dose when 80 Gy was sought) pre-
vented reproduction of adults, the most radiotol-
erant stage of the insect. The objective of this re-
search was to determine the effect of hypoxia on
reproduction of the plum curculio.
Plum curculios originally collected near
Gainesville, Florida, were obtained from a colony
at the United States Department of Agriculture,
Agricultural Research Service facility in Byron,
Georgia. The insects were reared at about 25EC,
70% RH, 12: 12h (L: D), on immature apples that
were picked when about 3 cm in diameter. Larvae
emerging from the apples were placed on steril-
ized potting soil until adult emergence.

A radiation source of 137Cs (Husman Model
521A, Isomedix, Inc., Whippany, NJ) that deliv-
ered a gamma ray dose rate of about 40 Gy.min-1
was used in this research. Routine dosimetry was
done with radiochromic film (Gafchromic MD-55,
ISP Technologies, Inc., Wayne, NJ) and read with
a spectrophotometer at 510 nm (Milton Roy Spec-
tronic 401, Ivyland, PA).
Adult plum curculios were irradiated in ambi-
ent atmospheres and in atmospheres of mostly
nitrogen. Cylinders (polyvinyl chloride, 37.5 cm
inside length, 10 cm inside diameter) fitted on one
end with a screw cap sealed with vacuum grease
and on the other end with 2 brass, barbed-nipple
compression hose fittings (25 mm long, 4-mm in-
side diameter) were constructed. Two-week-old
plum curculio adults were placed inside the cylin-
der with a few immature apples, and the atmo-
sphere was purged through the hose fittings with
nitrogen at a pressure of about 3 kPa for 2 min-
utes 20, 16, and 2 h before irradiation with an ab-
sorbed dose of 40 Gy. After purging, the hose fit-
tings were sealed with rubber septa and the cyl-
inders held at about 24C. About 1.5 h after irra-
diation, the cylinders were opened to return the
insects to ambient atmosphere. There were 6 rep-
licates of 300-600 each for adults irradiated in
ambient or hypoxic atmospheres. Controls con-
sisted of 6 replicates of 30-100 insects each in a
cylinder under ambient atmosphere that were not
After irradiation, adults were maintained on
immature apples at about 25C, and mortality
was determined every week. The apples were re-
placed every week and maintained at 25C for
development of any immatures inside. Larvae
emerging from apples were collected every 1-3
days and placed on potting soil for pupation and
adult emergence. After larvae were no longer
emerging, the apples were opened and any re-
maining insects collected. Analyses of variance
were done with Prism 4 (www.graphpad.com).
The mortality rate for the irradiated plum cur-
culios was greater than that for the control until
15 weeks after irradiation when the rate for the
control accelerated (Fig. 1). There was no signifi-
cant difference among the treatments for time to
reach 95% mortality (overall mean = 27.4 2.6
weeks, F = 1.56, P = 0.26, df: 2,15).
Reproduction under irradiation and hypoxia
was increased by over 20-fold compared with irra-
diation in ambient atmosphere (Table 1). Repro-
duction was greater than the 34.4 Gy for 4th in-
stars per female reported by Hallman (2003) for

June 2005

Scientific Notes

S75 \\

50- \ ", ----- control

2 25- -
-- Ambient ', - Hypoxla

0 10 20 30 40 50
Weeks after irradiation
Fig. 1. Mortality rate for control and irradiated (40
Gy) plum curculios.

the control, but similar to that reported for 40 Gy
under ambient atmosphere (0.31). Although few
4th instars were produced with irradiation under
ambient conditions, significantly more of these
became adults compared with the control and ir-
radiation under hypoxia. Adults irradiated under
ambient conditions did not live significantly less
time than control or adults irradiated under hy-
poxia, but their reproductive period was shorter
and peaked earlier.
Previous studies on the effect of hypoxia on ir-
radiation phytosanitary treatment efficacy con-
cluded that, although an effect could be observed,
it did not necessarily threaten the ability of the
treatment to prevent an infestation. Although
5.3% of irradiated (200 Gy) oriental fruit moth 5th
instars developed to adults, they all died within
10 days of emergence without ovipositing, while
non-irradiated controls lived up to 28 days and laid
abundant eggs that hatched (Hallman 2004a).
Even though apple maggots irradiated as 3rd in-
stars in apples subject to hypoxia had an esti-
mated increase in the dose required to prevent
adult emergence of 17% compared with those irra-
diated in ambient atmosphere, no adults emerged

in large-scale testing (Hallman 2004b). In both of
these cases the measure of efficacy was preven-
tion of adult emergence, which allows for a
greater margin of error than a treatment against
adults where prevention of successful reproduc-
tion is the only viable efficacy standard. In a treat-
ment designed to prevent adult emergence, even
if some adults emerge, as long as they will not suc-
cessfully reproduce, establishment of the pest in a
new area is prevented. But in a pest where adults
may be present, such as plum curculio, any failure
in prevention of reproduction means that some in-
dividuals would be capable of a new infestation.
The difference in reproductive success be-
tween plum curculios irradiated under hypoxia
and ambient atmosphere is approximately of the
same order of magnitude as the difference be-
tween 20 and 40 Gy under ambient conditions
(Hallman 2003). This may give a rough indication
that hosts of plum curculio irradiated under low
oxygen storage might require twice the dose as
under ambient conditions, or about 180 Gy to pre-
vent reproduction. This dose would need to be
confirmed by large-scale testing before it could be
used as a phytosanitary treatment. Until that is
done, irradiation against plum curculio should
not be used for fruits under low oxygen storage.
Miguel Diaz and Sandra Ramos, USDA-ARS,
Weslaco, Texas, are acknowledged for technical
help. Lisa Neven, USDA-ARS, Wapato, Washing-
ton, is thanked for providing immature apples.
Walter P. Gould, USDA-APHIS-PPQ, Riverdale,
MD, and Lisa Neven, USDA-ARS, Wapato, WA,
are thanked for reviewing the manuscript. This
research was supported by funding from USDA-


Adult plum curculios irradiated in a hypoxic
atmosphere accomplished by flushing a cylinder
with nitrogen gas were more tolerant of ionizing
radiation than plum curculios irradiated in ambi-
ent atmosphere. Some hosts of plum curculio,


Mean 4th instars/ Final
4th instars Week of peak female during contiguous
Mean 4th developing reproduction week of peak Final week of week of
Treatment instars/female to adult (%) post irradiation reproduction reproduction reproduction

Control 89.8 + 10.1 a 42.5 1.9 b 4.8 0.79 a 18.4 3.6 a 24.3 3.1 a 17.3 0.67 a
Hypoxia 5.9 0.91 b 40.0 3.7 b 5.8 1.1 a 1.9 0.30 b 28.8 2.5 a 17.5 1.4 a
Ambient 0.27 + 0.12 c 64.7 2.9 a 1.5 0.34 b 0.18 0.08 c 8.8 1.5 b 2.8 0.48 b
-Results of statistical analyses (degrees of freedom for all are 2, 15):
F value 72.4 25.9 6.8 25.1 17.6 102
P value 0.0001 0.0001 0.014 0.0001 0.0005 0.0001

"Means followed by the same letter are not significantly different, Tukey's, 95% confidence.

Florida Entomologist 88(2)

such as apples, are stored under hypoxia. An irra-
diation quarantine treatment against plum cur-
culio for apples stored in hypoxia would probably
need to be greater than the 92 Gy determined to
be efficacious in ambient atmosphere.


[APHIS] U.S. Dept. Agric., Animal and Plant Health In-
spection Service. 2002. irradiation phytosanitary
treatment of imported fruits and vegetables. Federal
Register 67: 65016-65029.
HALLMAN, G. J. 2002. Quarantine treatments: achiev-
ing market access for agricultural commodities in

the presence of invasive pests, pp. 271-291 In G. J.
Hallman and C. P. Schwalbe [eds.], Invasive Arthro-
pods in Agriculture: Problems and Solutions. Science
Publishers, Enfield, New Hampshire.
HALLMAN, G. J. 2003. Ionizing irradiation quarantine
treatment against plum curculio (Coleoptera: Curcu-
lionidae). J. Econ. Entomol. 96: 1399-1404.
HALLMAN, G. J. 2004a. Ionizing irradiation quarantine
treatment against oriental fruit moth (Lepidoptera:
Tortricidae) in ambient and hypoxic atmospheres. J.
Econ. Entomol. 97: 824-827.
HALLMAN, G. J. 2004b. Irradiation disinfestation of ap-
ple maggot (Diptera: Tephritidae) in hypoxic and
low-temperature storage. J. Econ. Entomol. 97:

June 2005

Scientific Notes


'Marine Laboratory, University of Guam, UOG Station, Mangilao, GU 96913, USA

2Permanent address: Departamento de Biologia, Facultad de Ciencias del Mar y Ambientales
Universidad de Cadiz, Poligono Rio San Pedro s/n Ap. 40. 11510, Puerto Real, Cadiz, Spain

3Museo Nacional de Ciencias Naturales, CSIC. Jos6 Guti6rrez Abascal, 2. 28006, Madrid, Spain

Smedly & Eisner (1995, 1996) stressed the dif-
ficulties that terrestrial herbivores, including
Lepidoptera, suffer in order to fulfill their need
for sodium, an essential ion plentiful in the seas,
but in short supply in plants.
Adult Lepidoptera drink from puddles, edges of
streams, carrion, and excreta from which they ob-
tain sodium and proteins (Beck et al. 1999). Pud-
dling behavior is typically, although not exclu-
sively, carried on by males (Boggs et al. 1991; Scul-
ley & Boggs 1996). It has been demonstrated that
sodium acquisition from puddles enables males to
provide mates with sodium, presumably via the
spermatophore (Pivnik & McNeil 1987; Smedly &

Eisner 1996). Sodium intake by males affects
their reproductive success, while the transfer of
sodium from male to female enhances the repro-
ductive successes of both females and eggs (Pivnik
& McNeil 1987) since females subsequently trans-
fer sodium to their eggs (Smedly & Eisner 1996).
Location of resources, which are usually rare
and patchily distributed (i.e., puddles with the
appropriate salt concentration), is not a simple
matter for Lepidoptera, which might use both vi-
sual (Papilionidae, Pieridae) and olfactory stimuli
(Lycaenidae, Nymphalidae) to locate them (Beck
et al. 1999). The sea, an easy to locate source of so-
dium, is basically unexploited by butterflies.

0 -
. .... --- -- Tn-r-

-9 ~.,

* a

Fig. 1. Reef shelf at Ipan (Guam) at low tide, showing four specimens of Papilio polytes scattered through the
shelf on green algae mats.

kft, -~i~C-- -L~----_. .~~- I.~i C



i: -

Florida Entomologist 88(2)

Fig. 2. Specimen of Papilio polytes drinking seawater on Ipan reef shelf (Guam). Note the proboscis introduced
into the water.

On August 28, 2004, on a sunny afternoon, we
observed about 20 male specimens ofPapilio poly-
tes Linnaeus, 1758, drinking seawater at low tide,
on the Ipan reef shelf on the southeast coast of the
island of Guam (Micronesia, USA) (Fig. 1). The
butterflies were extending their proboscis directly
into the sea while standing on green algae float-
ing mats or on exposed coral structures (Fig. 2), at
a distance from the shore ranging from 0.3 to 15
m. This behavior was observed for about 1 h until
the butterflies left. No other Lepidoptera were
seen on the reef, but numerous specimens of
Euploea eunice Quoy, 1815, were observed on the
beach shrubs.
Sodium may be a rare resource on the oceanic
island of Guam, as it is often the case in out-
washed tropical soils (Ross & Dykes 1996). How-
ever, inland sodium resources are widespread
enough to permit butterflies' persistence without
drinking marine water. Seawater intake by
P polytes in Guam is not likely the only possibility
for this butterfly to obtain minerals, but because
of the number of specimens seen, it seems a fa-
vored option. Higher salt concentration as a result
of evaporation during sunny low tide, and resting

places on the water, are factors that favored the
presence of P polytes in the shallow open waters
of the reef shelf. The absence of these factors alone
cannot explain why the sea is not more widely
used by Lepidoptera as a source for sodium.
We hypothesize that water temperature is a
critical factor diverting butterflies from seawater
as a sodium source. Water temperature recorded
at the sea reef platform at low tide reached 36 to
42C on the surface, a temperature often reached
in mud-puddles used by butterflies in tropical and
Mediterranean regions. This hypothesis, together
with differences in water availability (Launer et
al. 1996), might explain the intraspecific regional
differences found in the use of mud puddles, much
less visited by the same or closely related species
in the cooler climate of central Europe than in the
dry steppe biomes of the Mediterranean region
(Beck et al. 1999).
We thank the University of Guam Marine Lab
friends and colleagues for help and assistance
during our visit and especially to Elaine Pinder.
MP's stay in Guam was supported by a short-term
fellowship of the Ministerio de Educaci6n, Cul-
tura y Deportes of Spain.

June 2005

Scientific Notes


Adult Lepidoptera rarely uses seawater as a
source for sodium. We observed specimens of
Papilio polytes drinking seawater on the coast of
Guam. Based on our observations, we hypothe-
size that water temperature might play a key role
while choosing among puddling sites.


puddling behavior in tropical butterflies: In search of
proteins or minerals? Oecologia 119: 140-148.
BOGGS, C. L., AND L. A. JACKSON. 1991. Mud puddling
by butterflies is not a simple matter. Ecol. Entomol.
16: 123-127.
MAN, S. B. WEISS, AND P. R. ERLICH. 1996. Puddling

behavior by bay checkerspot butterflies (Euphydryas
editha bayensis). J. Res. Lepid. 32: 45-52.
PIVNIK, K., AND J. N. MCNEIL. 1987. Puddling in butter-
flies: sodium affects reproductive success in
Thymelicus lineola. Physiol. Entomol. 12: 461-472.
Ross, S. M., AND A. DYKES. 1996. Soil conditions, ero-
sion and nutrient loss on steep slopes under mixed
dipterocarp forest in Brunei Darussalam, pp. 259-
270 In D. S. Edwards, W. E. Booth, and S. C. Choy
[eds.], Tropical Rainforest Research-Current Is-
sues. Kluwer, Dordrecht
SCULLEY, C. E., AND C. L. BOGGS. 1996. Mating sys-
tems and sexual division of foraging effort affect
puddling behaviour by butterflies. Ecol. Entomol.
21: 193-197.
SMEDLY, S., AND T. EISNER 1995. Sodium uptake by
puddling in a moth. Science 270: 1816-1818.
SMEDLY, S., AND T. EISNER 1996. Sodium: A male
moth's gift to its offspring. Proc. Nat. Acad. Sci. USA
93: 809-813.

Florida Entomologist 88(2)


USDA, ARS Biological Control of Pests Research Unit, P.O. Box 67, Stoneville, MS 38776

Pseudacteon curvatus Borgmeier is one of three
species of phorid decapitating flies currently ap-
proved for release in the U.S. for the suppression
of the red, black, and hybrid imported fire ants,
Solenopsis invicta Buren, S. richteri Forel, and S.
invicta x richteri, respectively. Two biotypes of P
curvatus are established in the U.S. The Las
Flores, Argentina biotype prefers black and hybrid
imported fire ants (Porter & Briano 2000) and is
established at sites in Alabama, Mississippi, and
Tennessee (Graham et al. 2003; Thead et al. 2003;
Vogt & Streett 2003; Vail et al. 2004; Ward et al.
2004). The Formosa, Argentina biotype prefers red
imported fire ants and is established at sites in
Florida (R. J. Vasquez & S. D. Porter, pers. comm.).
This study reports dispersal of flies of the Las
Flores biotype, first released in spring, 2002 in a
grazed pasture (Knox site) in Clay Co., MS (3.25
ha, 3340'05.87"N, 8834'48.02"W) (Fig. 1) (Vogt
& Street 2003). By Sept. 2002, flies had estab-
lished on a mixture of black and hybrid imported
fire ants and had spread up to 600 m from the
original release site (Vogt & Streett 2003). Addi-
tional releases, with the same protocol, were
made during spring 2002 and 2003 in a grazed
pasture (Prima site) in Clay Co., MS, about 8.8

km and 149.7 SE of the Knox site (Fig. 1). Fly
presence was confirmed at both sites during 2003
(J. T. V. & L. G. T., unpubl. data).
Observations were made outside the release
sites on 23 dates from May-Sept. 2004, between
09:25 and 15:45 hours at 134 active fire ant
mounds. Sampling areas were randomly selected
and located on roadsides that were bordered by
forests with overhanging vegetation or by grazed
pastures. The presence of P. curvatus was deter-
mined by making a round depression (about 4-5
cm wide and 5-10 cm deep) in black and hybrid
imported fire ant mounds. Hovering flies were
counted within and around the depression. Ants
were macerated and dropped into the depression
to release semiochemicals that attract the flies
(Porter et al. 2004). All sampled areas were geo-
referenced. Mounds were observed for up to 35
min. If flies were found in an area, we moved and
sampled farther from the release sites. An area
was re-sampled later unless flies were found far-
ther from the release sites along a similar com-
pass bearing. Average air temperature during
sampling was 29.9 2.6C (SD), with a mean rel-
ative humidity of 66.3 18.5%, and a mean wind
speed of 1.2 1.45 km/h.


.0 r
0O o0 MS AL
o Legend
Monroe / ite
*0 Prima
St Flies
Lamar Flies
S 0 No Flies

Fig. 1. Dispersal, as of 2004, of the decapitating fly P. curvatus from 2002 and 2003 releases at two sites in Clay
Co., MS.

June 2005

Scientific Notes

A total of 130 flies were recorded attacking
ants in approximately 33% (44/134) of the
mounds. Each of 44 mounds had from 1 to 14 flies
with an average of 3.0 2.7 flies. Time of fly ar-
rival at a mound ranged from about 10 sec to 20
min, averaging 6.1 5.3 min.
A modified electric livestock prod, TheBlue-
OneTM LMPlus, (Hot-Shot Products Co., Inc.,
Savage, MN) was used randomly to shock the
ants to provoke alarm pheromone release (Vander
Meer et al. 2002; Barr 2004) in approximately
40% (53/134) of the mounds. Flies were attracted
to ants in about 25% (13/53) of the stimulated
mounds and approximately 38% (31/81) of un-
stimulated mounds. It took longer for flies to ar-
rive at stimulated mounds (8.2 6.7 min) than
unstimulated mounds (5.1 4.3 min). Unlike
P tricuspis (Barr 2004),the use of the electric live-
stock prod on ants did not appear to attract P cur-
vatus as quickly or at a greater rate.
Regression analysis showed no correlation (P >
0.05) between time to fly arrival and temperature,
relative humidity, wind speed or time of day. A sig-
nificant correlation (P < 0.001) between time to fly
arrival and day sampled explained 61% of vari-
ance, but confounding of sampling farther from
the release sites as the season progressed made
the model (y = 2E 1Ox443) unreliable. Therefore, a
significant correlation (P < 0.001) between time to
fly arrival and distance from Knox site is shown
(Fig. 2). It is likely that there were fewer flies at
the more distant sites (Porter et al. 2004).
Fly dispersal from the Knox and Prima sites is
illustrated in Fig. 1. By spring 2004, P. curvatus
populations at the Knox and Prima sites had
merged. By Sept. 2004, flies had spread over 44
km (356 NW), 37 km (162 SE), and 24 km (70



-E 15



0 5 10 15 20 25 30 35 40 45
Distance from Release Site (km)
Fig. 2. Time to fly arrival at a fire ant mound as a
function of distance from Knox release site. A logarith-
mic regression is plotted (y = 3.18 LN(x) 1.63, R2 =
0.30, P< 0.001).

NE) from the center of the Knox site. The outer
boundaries of fly expansion to the north, south,
and east of the release sites may have extended
farther than we were able to observe in this study.
Time restraints prevented sampling beyond Sept.
2004. Fly movement to the west of the release
sites appeared to be slower than in other direc-
tions. By Aug. 2004, dispersal was about 11 km
(291 NW) from the Knox site. No flies were found
farther than 11 km west of the Knox site in Sept.
2004. Habitat variation or sampling effort may
explain slower dispersal (Porter et al. 2004).
P curvatus occupied an area that encompassed
more than 2249 km2 (>224,914 ha) by Sept. 2004
(Fig. 1). Dispersal was 11 to at least 44 km in 21/2
We thank Mary Vowell, Evita Gourley and Dan
Harsh for assistance with fly surveys and Jimmy
Bryan, B. Bryan Farms, for access to the release
sites. Comments by Jason Oliver, Ken Ward, and
two anonymous reviewers helped improve the
manuscript. Mention of trade names or commer-
cial products in this publication is solely for the
purpose of providing specific information and
does not imply recommendation or endorsement
by the U.S. Department of Agriculture.


The fire ant decapitating fly, Pseudacteon cur-
vatus, first released in Clay County Mississippi
during spring 2002, occupied an area of over 2249
km2 by Sept. 2004. Dispersal was at least 44 km
in 212 years.


BARR, C. L. 2004. Phorid fly detection enhancement
with a modified electric livestock prod, pp. 139-141
In D. Pollet [ed.], Proceedings Annual Red Imported
Fire Ant Conference. Baton Rouge, LA. 203 pp.
D. DOROUGH, AND A. T. KELLEY. 2003. Field releases
of the decapitating fly Pseudacteon curvatus
(Diptera: Phoridae) for control of imported fire ants
(Hymenoptera: Fonnicidae) in Alabama, Florida,
and Tennessee. Florida Entomol. 86: 335-340.
PORTER, S. D., AND J. A. BRIANO. 2000. Parasitoid-host
matching between the little decapitating fly Pseudac-
teon curvatus from Las Flores, Argentina and the black
fire ant Solenopsis richteri. Fla. Entomol. 83: 422-427.
RISON. 2004. Establishment and dispersal of the fire
ant decapitating fly Pseudacteon tricuspis in North
Florida. Biol. Control. 29: 179-188.
2003. Establishment of Pseudacteon curvatus vs.
black and hybrid fire ants in Tennessee, In W. G.
Haun [ed.], Proceedings 30th Annual Meeting Ten-
nessee Entomol. Soc., The Firefly. Nashville, TN.
J. T. VOGT, AND A-M. CALLCOTT. 2004. Phenomenal

Florida Entomologist 88(2)

fire ant decapitating fly recovery in Tennessee, In W. G.
Haun [ed.], Proceedings 31st Annual Meeting Tennes-
see Entomol. Soc., The Firefly. Nashville, TN.
SON. 2002. Semiochemicals released by electrically
stimulated red imported fire ants, Solenopsis invicta
(Hymenoptera: Formicidae). J. Chem. Ecol. 28 (12):
VOGT J. T., AND D. A. STREETT. 2003. Pseudacteon cur-
vatus (Diptera: Phoridae) laboratory parasitism, re-

lease and establishment in Mississippi. J. Entomol.
Sci. 38(2): 317-320.
R. N. WARD, AND J. T. VOGT. 2004. Release of Pseu-
dacteon curvatus into black and hybrid imported
fire ant populations in northern Alabama, south-
ern Tennessee and eastern Mississippi, pp. 158-
160 In D. Pollet [ed.], Proceedings Annual Red Im-
ported Fire Ant Conference. Baton Rouge, LA. 203

June 2005

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs