ARTICLE
Selective Oviposition by Aedes aegypti (Diptera: Culicidae) in
Response to Mesocyclops longisetus (Copepoda: Cyclopoidea)
Under Laboratory and Field Conditions
JOSE´ LUIS TORRES-ESTRADA, MARIO H. RODRI´GUEZ,1 LEOPOLDO CRUZ-LO´PEZ,2 AND
JUAN I. ARREDONDO-JIMENEZ
Centro de Investigacio´n de Paludismo, Instituto Nacional de Salud Pu´blica, Tapachula, Chiapas, Mexico
J. Med. Entomol. 38(2): 188—192 (2001)
ABSTRACT The inßuence of predacious Mesocyclops longisetus Thiebaud on the selection of
oviposition sites bypreyAedes aegypti (L.)was studiedunder laboratory andÞeld conditions. Inboth
cases, gravid Ae. aegypti females were signiÞcantly more attracted to ovitraps containing copepods
or to ovitraps with water in which copepods were held previously than to distilled water. Mono-
terpene and sesquiterpene compounds including 3-carene, a-terpinene, a-copaene, a-longipinene,
a-cedrene, and d-cadinene were found in hexane extracts of copepods by gas chromatography and
mass spectrometry analyses. These compounds may be responsible for attracting gravid Ae. aegypti
females and may increase the number of potential prey for the copepod.
KEY WORDS Aedes aegypti, Mesocyclops longisetus, oviposition behavior, attractants, copepod
essence, predator-prey relationships
SELECTION OF OVIPOSITION sites by gravid female mos-
quitoes is a crucial event for the survival of their
species. Gravid females follow visual or olfactory cues
to appropriatewater collections and, guided by chem-
ical cues and physical factors in the water, assess the
quality of the water before a decision to lay eggs is
made (Muir 1988). The presence of aquatic predators
of mosquito eggs, larvae, or pupae is a risk that some
species take into account before oviposition (Chesson
1984, Lowenberger and Rau 1994, Ritchie and Laid-
law-Bell 1994, Zahiri and Rau 1998).
Copepods have been reported to be effective in
controllingAedes aegypti (L.) andAe. albopictusSkuse
in natural and artiÞcial larval habitats (Marten 1990,
Marten et al. 1994). However, in preliminary Þeld
studies using Mesocyclops longisetus Thiebaud to con-
trol Ae. aegypti in southern Mexico, we observed that
variable numbers of Aedes larvae coexisted with this
copepod. In a series of experiments to deÞne the
predator-prey population dynamics, Ae. aegypti fe-
males preferred to oviposit in water containing cope-
pods. We present herein the results of experiments
that document this observation and identify the vol-
atile compounds released by the copepods that prob-
ably are involved in this attractive effect.
Materials and Methods
Copepods. Mesocyclops longisetus were obtained
fromFacultaddeCienciasBiolo´gicas,UniversidadAu-
to´noma de Nuevo Leo´n, Me´xico, and a colony was
established at the Centro de Investigacio´n de Palud-
ismo (CIP). Copepods were maintained in water at
258C, in 4-liter containers, and fed a mixture of milk
powder and Paramecium caudatum L. (Sua´rez et al.
1992).
Mosquitoes. Aedes aegypti were insectary-reared
from a colony from Tapachula established in 1994,
following CIP standard rearing procedures. Larvae
were reared at a density of 800 larvae per tray (40 by
30 by 5 cm) at 28 6 18C water temperature and a
photoperiod of 12:12 (L:D) h; and they were fed
powdered wheat germ. Female wing length averaged
2.70 6 0.2 mm. Before the experiments, adults were
kept in 73-cm3 screened cages at 27 6 18C, 70% RH,
and fed 10% sugar on soaked cotton pads. Mated fe-
males blood fed on rabbits and were allowed to de-
velop their eggs before experimentation.
Treatments. Gravid females were offered three
types of water as oviposition substrates in standard
ovitraps (1-liter metallic cans painted black inside
with a wooden paddle covered in Þlter paper, Service
1993): (1) distilled water (0.8 liter) containing 40 M.
longisetus (W/C), (2) distilled water (0.8 liter) in
which 40 copepods were held for 48 h but removed
before experiments (H/C), and (3) distilled water
(0.8 liter) (S/C).
Laboratory Bioassays. Seventy gravid Ae. aegypti fe-
males were offered oviposition substrates (H/C, W/C,
1 Centro de Investigaciones Sobre Enfermedades Infecciosas, In-
stituto Nacional de Salud Pu´blica. Av. Universidad 655, Col. Santa
Marı´a Ahuacatitla´n. Cuernavaca, Morelos 62508 Mexico.
2 El Colegio de la Frontera Sur, Apartado Postal 36, Tapachula,
Chiapas, 30700 Mexico.
0022-2585/01/0188—0192$02.00/0 q 2001 Entomological Society of America

andS/C) in standardovitraps in a screenedcage(170by
100 by 120 cm) for 24 h. After exposure, ovitraps were
examined and eggs were counted. Three sets of exper-
imentswere repeated for20consecutivedays,usingnew
mosquitoes and fresh ovitraps on each day. To eliminate
possible bias due to location, ovitraps were set at the
center of the cage and rotated, allowing all three treat-
ments to be in every possible orientation. Effects of
positioning of ovitraps and oviposition site preference
were evaluated using two-way factorial analysis of vari-
ance(ANOVA)of log(x11)transformeddatafollowed
by Bonferroni-Dunn posthoc tests for multiple compar-
isons (Dunn 1961, Zar 1999).
Oviposition activity was expressed as the oviposi-
tion activity index (OAI, Kramer and Mulla 1979)
where OAI 5 Ni 2 Nc/Ni 1 Nc, and Ni 5 number of
eggs deposited in treatment ovitraps and Nc 5 the
number of eggs deposited in control ovitraps. Re-
sponses using this system ranged from one (attrac-
tion) to 21 (repellence), with zero denoting no pref-
erence. In this context, attraction refers to the
deposition of eggs as the end result of a complex
sequence of behaviors including orientation toward
the source and stimulation to oviposit. These behav-
iors were not differentiated under the current bioas-
say and only the endpoint of the pre- and oviposition
behaviors (i. e., oviposition) was used to determine
attraction or repellency.
Field Bioassays. Field experiments were carried out
in the cemetery of Ciudad Hidalgo, Chiapas, Me´xico
(148 409 N, 928 099 W, elevation 20 m). Twenty tombs
attractive for oviposition to gravid Ae. aegypti females
(shaded by trees, Bentley and Day 1989) were se-
lected. Ovitraps with each treatment (H/C, W/C, and
S/C, 0.8 liter each) were placed on each tomb fol-
lowing the same triangular orientation as used in the
laboratory assays. Similarly, three series of 20 exper-
iments, one per day, with 20 replicates each (one per
tomb) were run, rotating treatment positions. Positive
ovitraps were recorded, according to treatment and
rotation, and the eggs counted. Because very few eggs
were laid in each ovitrap, oviposition site preferences
were analyzed only as proportions of ovitraps positive
for eggs of Ae. aegypti (with at least one egg). Effects
of positioning of ovitraps and oviposition site prefer-
ence were evaluated using two-way factorial ANOVA
of arcsine (p)one-half transformed data followed by
Bonferroni—Dunn tests for multiple comparisons
(Dunn 1961, Zar 1999).
Distances of Response Bioassays. Additional bioas-
says assessed the maximum distance at which cope-
pod-released substances affect substrate discrimina-
tion among gravid Ae. aegypti females. Ten attractive
tombs were selected in the cemetery and H/C—ovi-
traps were set at 20, 30, 40, and 50 cm from control
(S/C) ovitraps. Traps were inspected every 24 h for 5
consecutive days for the presence of eggs. Paired t-
tests (Zar 1999) comparedpositive sites forAe. aegypti
eggs between treatment and control ovitraps.
Chemical Analysis. Gas chromatography-mass
spectrometry (GC-MS) analyses were carried out on
samples fromall treatments (W/C,H/C,S/C).Brießy,
samples (distilled water with 200 copepods, distilled
water that previously held 40 copepods for 48 h or
distilledwater) inglass capillarieswereanalyzedusing
a fused silica column (30 m by 0.25 mm) coated with
poly (5% diphenyl/95% dimethyl siloxane) in a gas
chromatograph (Varian Star 3400 CX, Mulgrave, Aus-
tralia) linked to a mass spectrometer (Varian Saturn)
programmed from 50 to 2508C at 158C /min. The car-
rier gaswashelium.The injectorport temperaturewas
held at 2008C. Also, hexane-extracts of 200 macerated
copepods, previously thoroughly washed with sterile
distilled water, were injected directly into the GC-MS
system using the same settings.
Results
Laboratory Bioassays. Positioning of containers
(i.e., the three rotations of each treatment) had no
signiÞcanteffect in theANOVA(F52.702; df52, 171;
P 5 0.757). Treatments were signiÞcant (F 5 948.4;
df 5 2, 171; P , 0.0001), and Bonferroni—Dunn
posthoc tests showed that signiÞcantlymore eggs (P ,
0.0001) were laid in containers containing water with
copepods (W/C, mean no. eggs laid 6 SE 5 650.8 6
16.2) than in distilled water (S/C, mean number eggs
laid 5 162.8 6 6.1). In addition, more eggs were laid
in water that had held copepods for 48 h (H/C, mean
number eggs laid 5 1003.5 6 34.9) than in water with
copepods (W/C) or in distilled water (P , 0.0001)
(Fig. 1). The interaction of treatments with rotation
was not signiÞcant (F 5 0.74; df 5 4, 171; P 5 0.56).
Fig. 1. Mean number of eggs laid by Aedes aegypti in
laboratory bioassays in three experiments where ovitraps
were rotated (H/C 5 water that held copepods for 48 h;
W/C 5 water with copepods; S/C 5 sterile distilled water
without copepods).
March 2001 ARREDONDO-JIME´NEZ ET AL.: EFFECT OF M. longisetus on Ae. aegypti 189