Electrophysiological Responses of Atta sexdens rubropilosa Workers
to Essential Oils of Eucalyptus and its Chemical Composition
Luciane G. Batista-Pereiraa, João B. Fernandesa, M. Fátima G. F. da Silvaa,
Paulo C. Vieiraa, Odair C. Buenob, and Arlene G. Corrêaa,*

a Departamento de Quı́mica, Universidade Federal de São Carlos, C.P. 676, 13565-905,
São Carlos, SP, Brazil. E-mail: agcorrea@power.ufscar.br

b Instituto de Biociências, Centro de Estudos de Insetos Sociais, Universidade Estadual
Paulista, 13506-900, Rio Claro, SP, Brazil

* Author for correspondence and reprint requests

Z. Naturforsch. 61c, 749Ð755 (2006); received February 1/March 16, 2006

The leaf-cutting ant Atta sexdens rubropilosa Forel, 1908 is the most harmful of the Euca-
lyptus pests, causing severe losses in wood production through defoliation. Various strategies
have been tried and effort spent on the development of methods to control this pest, however
no practical and environmentally acceptable one currently exists. In this work the chemical
composition of the essential oil of seven Eucalyptus species was identified and the selectivity
and sensitivity of antennal receptors of A. sexdens rubropilosa workers to the volatile com-
pounds were determined using the electroantennographic technique (EAG and GC-EAD).
Analysis by GC-EAD showed in E. cloesiana and E. maculata, respectively, seventeen and
sixteen terpenes that elicited responses in ant workers’ antennae, indicating the potential
role of the essential oils as allelochemicals that determine the choice of the foraging material.

Key words: Leaf-Cutting Ant, Essential Oil, Electroantennography

Introduction

The leaf-cutting ants are the major insect prob-
lem to agriculture and forestry in Brazil and Atta
sexdens rubropilosa Forel, 1908 is one of the prin-
cipal species of this group. It is the most harmful
of the Eucalyptus pests, causing severe losses in
wood production through defoliation (Cherret,
1986; Fowler et al., 1989). Various strategies have
been tried and substantial effort spent on the de-
velopment of methods to control this pest, yet no
practical and environmentally acceptable one cur-
rently exists.

Several plant secondary metabolites exhibit bio-
logical activities that have the potential to exert
an effect on the physiology and/or behaviour of
insects, and it is usually considered that such com-
pounds are involved at some point in plant-insect
relationships (Herrera and Pellmyr, 2002). The
knowledge of the volatiles of essential oil of the
genus Eucalyptus could be used as the basis for a
control strategy of the leaf-cutting ants, given that
they have a well-developed olfactory system (Höll-
dobler and Wilson, 1990).

To understand the interactions mediated by
semiochemicals, it is necessary to study the factors

0939Ð5075/2006/0900Ð0749 $ 06.00 ” 2006 Verlag der Zeitschrift für Naturforschung, Tübingen · http://www.znaturforsch.com · D

involved in the olfactory perception of these com-
pounds in order to identify those that could attract
A. sexdens. To date, the study with electroanten-
nography (EAG) measurements applied to A. sex-
dens rubropilosa is restrict to two common trail
pheromone components, 4-methylpyrrol-2-carbo-
xylate and 2-ethyl-3,6-dimethylpyrazine (Kleinei-
dam et al., 2005).

In this work, we identified the chemical compo-
sition of the essential oils of seven Eucalyptus spe-
cies, which were selected on the basis of foliage
characteristics, oil content and resistance by not
preference (Anjos et al., 1987; Andrade et al.,
1989; Berti-Filho et al., 1991; Boland et al., 1991;
Anjos and Santana, 1994; Vendramin et al., 1995).
Besides, the selectivity and sensitivity of antennal
receptors of A. sexdens rubropilosa workers to
volatile compounds of these essential oils were de-
termined, using the electroantennographic tech-
nique (EAG). Gas chromatography linked on-line
to electroantennography detection (GC-EAD)
studies were also performed with volatiles sam-
pled from E. cloesiana and E. maculata essential
oils. This is the first report of the use of GC-EAD
technique for A. sexdens rubropilosa.



750 L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa

Materials and Methods

Eucalyptus essential oil extraction and analysis

Fresh leaves of E. grandis Hill ex Maiden, E.
citriodora Hook., E. camaldulensis Dehnh., E. sa-
ligna Sm., E. urophylla S. F. Blake and E. cloesiana
were collected in the forest species arboretum, lo-
cated in the experimental plantation of the Uni-
versidade Estadual Paulista, Botucatu, SP, Brazil.
E. maculata leaves were obtained from Univer-
sidade Federal de Viçosa, MG, Brazil. The leaves
were randomly collected from approx. six-year-old
Eucalyptus trees.

The fresh leaves (400 g) of each species were
submitted to steam distillation for 4 h, using a Cle-
venger apparatus. The essential oils in the distillate
were dried over anhydrous Na2SO4 and kept in
the freezer at 0 ∞C. Essential oils were analyzed
using a Shimadzu GC-17A gas chromatograph fit-
ted with a fused silica DB-5 capillary column
(30 m ¥ 0.25 mm ID, 0.25 μm film thickness; J and
W Scientific) and with helium as carrier gas at a
flow-rate of 1.6 ml minÐ1. The temperature was
held initially at 60 ∞C for 2 min and then increased
at a rate of 3 ∞C minÐ1 to 240 ∞C. The injector was
in the split mode at 225 ∞C. The interface tempera-
ture was 250 ∞C. The chromatograph was coupled
to a Shimadzu QP5000 mass selective detector
(EIMS) at 70 eV. Components were identified by
determination of their retention indices relative to
those of a homologous series of n-alkanes (Dool
and Kratz, 1963), by co-injection with authentic
samples, and by comparing fragmentation patterns
of mass spectra with those stored in the spectrom-
eter database and bibliography (Adams, 1995).

Collection of the insects

A. sexdens rubropilosa workers were collected
from six colonies located at the campus of the Uni-
versidade Federal de São Carlos, Central Region
of São Paulo State, Brazil. Only larger foragers
were collected, i.e., leaf-carrying workers walking
along trails.

Electrophysiology

Antennae of A. sexdens rubropilosa workers
were used for electroantennographic experiments
(EAG) and electroantennographic detection
(EAD). Each antenna was pulled from the head
with forceps and two segments were cut off at the
base (Bjostad, 1998). The antenna was then fixed

between two stainless steel electrodes by pushing
the base and tip into droplets of an electrically
conductive gel (Spectra 360“ electrode gel, Parker,
Orange, New Jersey) applied to the metal elec-
trodes. The antennal responses were amplified and
recorded with a data acquisition controller and
software EAG (Syntech, Hilversum, The Nether-
lands).

EAG experiments were performed in order to
elucidate the selectivity of the antennal receptors
of A. sexdens rubropilosa workers. The EAG re-
sponse was evaluated as follows: The volatile com-
pounds or control were released from Pasteur
pipettes containing a piece of filter paper (ca.
0.8 cm2) previously impregnated with 10 μl of a
freshly prepared solution of each test essential oil
in hexane, after the solvent had evaporated. The
puff containing the test essential oil (from seven
Eucalyptus species) was delivered into a continu-
ously humidified and purified air stream (1.2
l minÐ1) passing for 0.3 s through the impregnated
filter paper in the pipettes. Control stimulation
was made at the beginning and the end of each
series of EAG experiments. The essential oils were
then applied randomly at intervals of 90 s. EAG
amplitudes in response to the essential oils were
expressed in relation to the responses to the con-
trol (hexane), because of the large differences in
overall sensitivity between individual antennae,
and to compensate the decline in antennal sensi-
tivity during a measuring session. In this normali-
zation procedure, the responses to the control
were defined as 100%. The values obtained be-
tween two calibration references (controls) were
calculated by linear interpolation between those
references values. The Syntech EAG software cal-
culated the normalized values automatically. The
essential oils were tested on 20 antennae of A. sex-
dens rubropilosa workers. The mean normalized
responses of the different compounds were sub-
mitted to ANOVA for statistical analysis and com-
pared by the Tukey test (P � 0.05).

The GC-EAD equipment was used to identify
individual compounds in the E. cloesiana and E.
maculata oils capable of eliciting an electroanten-
nographic response. Ten replicates were recorded
for A. sexdens rubropilosa workers antennae, us-
ing essential oils from the two species. The GC-
EAD instrument consisted of a Shimadzu 17-A
gas chromatograph equipped with a flame ioniza-
tion detector (FID) coupled to an electroanten-
nography system and a DB-5 column (30 m ¥



L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa 751

0.25 mm ID, 0.25 μm film thickness), using the fol-
lowing temperature program: 60 ∞C for 1 min, ris-
ing at 3 ∞C minÐ1 to 150 ∞C, held for 10 min, and
finally rising at 1 ∞C minÐ1 to 280 ∞C and held for
16 min. The column effluent (carrier gas hydro-
gen) was mixed with a flow of nitrogen make-up
gas (12 ml minÐ1) before it was split to the FID
and to a heated transfer capillary leading to the
antennal preparation. From the transfer capillary
the compounds entered a humidified and purified
air stream (1.2 l minÐ1), which carried them di-
rectly over the antennal preparation on the steel
electrodes. The FID was kept at 280 ∞C, whereas
the temperature of the transfer capillary was main-
tained at 290 ∞C to avoid condensation.

Results and Discussion

By the electroantennographic technique a high
selectivity of the antennae of A. sexdens rubropi-
losa to the species E. cloesiana, E. camaldulensis,
E. urophylla, E. saligna, E. grandis, E. citriodora
and E. maculata was evident (Fig. 1). These results
indicate the potential role of the essential oils as
allelochemicals that determine the choice of the
foraging material. When we compared the results
obtained with the literature data we can infer that
E. cloesiana, followed by the species E. camaldu-
lensis, E. urophylla and E. saligna are the species
preferred for the forage in relation to E. grandis
and E. citriodora. These two last species are more
preferred than E. maculata. These results are in
agreement with the results of behavioral assays
obtained by Andrade et al. (1989) and Vendramin
et al. (1995), where the species E. camaldulensis,
E. urophylla and E. saligna were preferred in com-

Fig. 1. Mean EAG (ð SD) elicited from Atta
sexdens rubropilosa worker antennae at the
concentration of 100 mg mlÐ1 of Eucalyptus es-
sential oils. Mean values marked with the same
letter are not significantly different at P � 0.05
on the basis of the Tukey test (N = 20 anten-
nae).

parison to E. grandis and E. citriodora. These two
species were considered resistant by not prefer-
ence. Anjos et al. (1986) verified that among the
20 species of eucalyptus more planted in Brazil,
E. maculata was considered highly resistant to A.
sexdens rubropilosa. Anjos and Santana (1994)
demonstrated the negative effect in the behavior
and in the survival of A. sexdens caused by E.
maculata leaves; this effect was not evidenced for
the species E. grandis and E. citriodora.

Composition of each essential oil of Eucalyptus
is given in Tables I and II. Component relative
concentrations were calculated from GC peak
areas and arranged in order of GC elution.

E. cloesiana and E. maculata essential oils were
chosen for the coupled gas chromatography-elec-
troantennography detection (GC-EAD) experi-
ment because they caused, respectively, the
greater and the smaller depolarization in the A.
sexdens antennae by the EAG technique.

It was verified that E. cloesiana essential oil pre-
sented 32 compounds 15 of which were identified,
corresponding to 93.84% of the total oil mass (Ta-
bles I and II). The major component in the oil
was α-pinene corresponding to 76.08% of the total
mass. It was observed that the oil is constituted
basically by terpenes.

Thirty-seven compounds were found in the E.
maculata essential oil, 18 of which were identified,
equivalent to 74.2% of the total mass (Tables I
and II). Again, the major component was α-pinene
(39.4%) followed by �-caryophyllene (10.34%), �-
pinene (6.87%) and limonene (2.68%).

This research is pioneering in the employment
of the GC-EAD technique to verify the olfactory
sensitivity of A. sexdens rubropilosa workers. With



752 L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa

Table I. Essential oil composition (%) from E. cloesiana (ECL), E. saligna (ES), E. citriodora (EC), E. camaldulensis
(ECA), E. grandis (EG), E. urophylla (EU) and E. maculata (EM) compared to Adams (1995).

Compound Adams ECL (%) ES (%) EC (%) ECA (%) EG (%) EU (%) EM (%)
identified KI KI obs KI obs KI obs KI obs KI obs KI obs KI obs

Tricyclene 926 0.42 0.09 0.16
α-Pinene 932 76.08 25.91 0.28 6.12 40.55 8.03 39.4
α-Fenchene 946 0.17 0.09 0.3
�-Pinene 975 2.21 0.58 2.37 0.23 4.59 6.87
Myrcene 991 0.61 0.76
α-Phellandrene 1004 0.53 0.19 1.83 4.02
n-Pentyl isobutyrate 1012 0.42 0.55
Isosylvestrene 1014 0.15 0.16
o-Cymene 1021 0.16 24.38 2.13 13.13 2.5
Limonene 1024 2.66 1.57 14.22 2.66 7.13 2.68
1,8-Cineole 1028 6.86 52.82 0.45 53.11
(Z)-�-Ocimene 1033 0.2 2.92 0.93
(E)-�-Ocimene 1043 0.34 0.68
γ-Terpinene 1054 24.63 6.79 16.25 0.69
2,4(8)-p-Menthadiene 1080 0.43 0.7 1.22 0.78 0.65
Linalool 1093 0.28 0.14 0.84 0.19
exo-Fenchol 1106 0.13 0.42
α-Canpholenol 1119 1.27 2.29
trans-Pinocarveol 1132 0.27 0.27
neo-iso-3-Thujanol 1139 11.84
Citronellal 1150 75.99 1.09
trans-Pinocamphone 1156 0.11 0.17
Isoborneol 1161 0.21 0.76
Terpin-4-ol 1174 0.38 2.29 2.64 1.27 0.89
p-Cimen-8-ol 1183 0.09 2.9
α-Terpineol 1188 3.81 2.19 2.83 2.90 2.97 0.82
trans-Carveol 1218 0.46
Citronellol 1332 5.42
p-Cimen-7-ol 1287 0.15
Carvocrol 1302 0.12
α-Terpenyl acetate 1350 0.37 4.38
α-Gurjunene 1408 0.24 0.24 0.29 0.33 0.2
�-Caryophyllene 1417 2.33 0.25 0.52 1.19 10.34
Aromadendrene 1437 0.85 0.94 0.6 2.81
Seychellene 1460 0.17 0.38 0.88 0.73
Bicyclogermacrene 1497 0.98 0.55 2.2 0.2
(E,E)-α-Farnesene 1511 0.19
Geranyl isobutyrate 1517 0.28
α-Cadinene 1524 0.23 0.40
Elemol 1550 0.53
α-Cadinene 1559 0.30
Spathulenol 1576 1.9 0.32 0.95 0.55
Nerolidol 1583 0.98 0.37 0.88 0.57 0.29 3.78
γ-Eudesmol 1631 0.1 0.38
�-Eudesmol 1649 0.48 0.36 0.66
α-Eudesmol 1652 0.67 0.28 1.02
Guaiol epimer 1658 0.45

the GC-EAD technique it was possible to detect
17 bioactive compounds present in E. cloesiana es-
sential oil that stimulated the antennae of A. sex-
dens workers (Fig. 2). Of these bioactive com-
pounds 12 were identified as: α-pinene (peak 1),
�-pinene (2), limonene (3), 2,4(8)-p-menthadiene
(4), terpin-4-ol (7), α-terpineol (8), �-caryophyl-

lene (9), aromadendrene (10), spathulenol (13),
nerolidol (14), �-eudesmol (16), and α-eudesmol
(17).

In the E. maculata essential oil (Fig. 3), by GC-
EAD, it was possible to detect 16 bioactive com-
pounds. Efforts were made to identify all 16 of the
stimulatory compounds among the volatile frac-



L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa 753

Table II. Not identified compounds observed in the essential oil from E. cloesiana (ECL), E. saligna (ES), E.
citriodora (EC), E. camaldulensis (ECA), E. grandis (EG), E. urophylla (EU) and E. maculata (EM).

Compound not identified ECL (%) ES (%) EC (%) ECA (%) EG (%) EU (%) EM (%)
KI observed

906 0.23 0.08 6.23
913 0.27 0.86 0.31
921 0.19 0.13 4.1
931
963 0.4
965 0.06

1015 0.35 0.27
1033 0.16
1066 0.36 0.38
1083 0.1 0.18
1092 0.47
1099 0.1
1106 0.2 0.15
1140 0.21
1143 0.51
1164 0.22 0.88
1196 1.32
1255 0.22
1267 0.16
1332 0.7
1336 0.62
1342 0.17
1358 0.56
1366 0.8
1374 0.5
1386 0.25
1395 0.17
1413 0.27
1442 0.3 0.44
1453 0.32 0.17 1.12
1476 0.56
1479 0.84
1482 0.86
1488 0.19
1491 0.21
1495 1.37
1498 0.71 1.13 1.51
1504 0.16
1522 1.16
1545 0.39
1566 0.11 0.31
1590 0.21 0.34 0.27 0.24 0.5 0.96
1592 0.24 0.3 0.99
1597 3.99
1600 0.14 0.23 0.81
1620 1.82
1622 0.14 0.13 0.82
1627 0.34 0.67
1649 0.51
1652 0.48
1668 0.67 0.58
1972 0.42

tions eluted from the oil of eucalyptus; however,
only 11 could be identified: α-pinene (peak 1), �-
pinene (2), limonene (3), α-terpineol (5), �-caryo-

phyllene (8), aromadendrene (9), elemol (11), spa-
thulenol (12), nerolidol (13), �-eudesmol (15), and
α-eudesmol (16).



754 L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa

Fig. 2. GC-EAD response of Atta sexdens rubropilosa
worker antennae to E. cloesiana essential oil (0.1 mg
mlÐ1). The numbers indicate the peaks that elicited elec-
trophysiological responses (N = 10).

Ten bioactive compounds were almost chemi-
cally identical in E. cloesiana and E. maculata spe-
cies. The compounds 2,4(8)-p-menthadiene (4)
and terpin-4-ol (7) were present only in E. cloe-
siana essential oil while elemol (11) only in E.
maculata essential oil. Of these results it can be
inferred that the difference in the depolarizations,
obtained for EAG and EAD, could be related to

Fig. 3. GC-EAD response of Atta sexdens rubropilosa
worker antennae to E. maculata essential oil (0.1 mg
mlÐ1). The numbers indicate the peaks that elicited elec-
trophysiological responses (N = 10).

these compounds and mainly to the ratio of each
compound, since in each oil they presented a dif-
ferent percentage. Besides, the differences in de-
polarizations could also be attributed to bioactive
compounds not identified in each essential oil. Ac-
cording to Barnola et al. (1997) and Barnola and
Cedeño (2000), the concentration of some terpe-
nes, mainly �-pinene and �-caryophyllene, would
be responsible for the difference in the attack of
the A. laevigata leaf-cutting ant observed in the
forestry species Pinus caribaea. In accordance with
the same authors, �-caryophyllene, with occur-
rence in high percentage, would have a repellent



L. G. Batista-Pereira et al. · Electrophysiological Responses of Atta sexdens rubropilosa 755

effect to A. laevigata. Of this form, the difference
in the antennal response noticed for A. sexdens by
EAG in E. maculata can be related with the pres-
ence of blend in high percentages of �-pinene
(6.87%) with �-caryophyllene (10.34%); hence in
E. cloesiana (2.21% �-pinene and 2.33% �-caryo-
phyllene) and in other Eucalyptus species these
compounds were present in low concentrations.
Moreover, Marsaro et al. (2004) demonstrated, by
behavioral assays, that hexane extracts of E. macu-
lata interfered in recognition mechanism among
workers. The main active compounds identified
from this plant were elemol and �-eudesmol.
These compounds may be responsible for the re-
sistance of this species to ant attack.

In this study, it was verified that some terpenes,
found in essential oils of E. cloesiana and E. macu-
lata, are bioactive in the antennae of A. sexdens
rubropilosa workers. These results indicate that

Adams R. P. (1995), Identification of Essential Oil Com-
ponents by Gas Chromatography/Mass Spectroscopy.
Allured Publishing, New York, p. 469.

Andrade M. L., Forti L. C., Padovani C. R., and Pacheco
P. (1989), Preferência de Atta sexdens rubropilosa
Forel, 1908 (Hymenoptera: Formicidae) por diferen-
tes espécies de Eucalyptus. In: Encontro de Mirmeco-
logia, Vol. 9. Viçosa, Brazil, p. 19.

Anjos N. and Santana D. L. Q. (1994), Alterações dele-
térias no comportamento de Atta laevigata (F. Smith)
e Atta sexdens rubropilosa Forel (Hymenoptera: For-
micidae), causadas por folhas de Eucalyptus spp. An.
Soc. Entomol. Brasil 23, 25Ð30.

Anjos N., Santos G. P., and Zanuncio J. C. (1986), Resis-
tência de Eucalyptus spp. a saúva-limão, Atta sexdens
rubropilosa Forel, 1908 (Hymenoptera: Formicidae).
In: Congresso Brasileiro de Entomologia, Vol. 10. Rio
de Janeiro, Brazil, p. 404.

Anjos N., Santos G. P., and Zanúncio J. C. (1987), A la-
garta parda, Thyrinteina arnobia (Stoll, 1782) (Lepi-
doptera: Geometridae) desfolhadora de eucaliptos.
Bol. Tec. Epamig 25, 1Ð56.

Barnola L. F. and Cedeño A. (2000), Inter-population
differences in the essential oils of Pinus caribaea nee-
dles. Biochem. Syst. Ecol. 28, 923Ð931.

Barnola L. F., Cedeño A., and Hasegawa M. (1997), In-
traindividual variations of volatile terpene contents in
Pinus caribaea needles and its possible relationship
to Atta laevigata herbivory. Biochem. Syst. Ecol. 25,
707Ð716.

Berti-Filho E., Stape J. L., and Cerignoni J. A. (1991),
Surto de Thyrinteina arnobia (Stoll, 1782) (Lepidop-
tera: Geometridae) em Eucalyptus citriodora Hook
(Myrtaceae) no estado de São Paulo. Rev. Agric. 66,
47Ð51.

the GC-EAD analysis is a useful technique for the
selection of compounds present in plant extracts
that have importance in the behavior of insects.
The results obtained from the electroantenno-
graphic technique also suggest that A. sexdens ru-
bropilosa workers use some volatile compounds as
signals to find the host. But the relevance of these
allelochemicals remains to be determined, i.e, if
the bioactive compounds are satisfactorily attract-
ive and specific in the field.

Acknowledgements

The authors are grateful to MCT-PRONEX/
FINEP, CNPq, FAPESP and IFS/ OPCW for the
financial support. L. G. B.-P. thanks FAPESP for
the award of a fellowship. We are also grateful to
Dr. C. F. Wilcken, UNESP, Botucatu, Brazil, for
providing us the leaves of the seven Eucalyptus
species used for the extraction of essential oils.

Bjostad L. B. (1998), Electrophysiological methods. In:
Methods in Chemical Ecology: Chemical Methods,
Vol. 1 (Haynes K. F. and Millar J. G., eds.). Chapman
and Hall, London, p. 339Ð369.

Boland D. J., Brophy J. J., and House A. P. N. (1991), Eu-
calyptus Leaf Oils: Use, Chemistry, Distillation and
Marketing. Inkata Press, Melbourne, p. 252.

Cherret J. M. (1986), The biology pests status and con-
trol of leaf-cutting ants. Agric. Zool. Rev. 1, 1Ð37.

Dool H. V. D. and Kratz P. D. (1963), A generalization
of the retention index system including linear tem-
perature programmed gas-liquid partition chromato-
graphy. J. Chromatogr. 11, 463Ð471.

Fowler H. G., Pagani M. I., da Silva O. A., Forti L. C., da
Silva V. P., and Vasconcelos H. L. (1989), A pest is
a pest. The dilemma of Neotropical leaf-cutting ants:
keystone taxa of natural ecosystems. Environ. Manag.
13, 71Ð675.

Herrera M. C. and Pellmyr O. (2002), Plant-Animal In-
teractions: an Evolutionary Approach. Blackwell Pub-
lishing, Oxford.

Hölldobler B. and Wilson E. O. (1990), The Ants. The
Belknap Press of Harvard University Press, Cam-
bridge, p. 732.

Kleineidam C. J. Obermayer M., Halbich W., and Ross-
ler W. (2005), A macroglomerulus in the antennal
lobe of leaf-cutting ant workers and its possible func-
tional significance. Chem. Senses 30, 383Ð392.

Marsaro A. L., Souza R. C., Della Lucia T. M., Fernan-
des J. B., Silva M. F. G. F., and Vieira P. C. (2004), Be-
havioral changes in workers of the leaf-cutting ant
Atta sexdens rubropilosa induced by chemical compo-
nents of Eucalyptus maculata leaves. J. Chem. Ecol.
30, 1771Ð1780.

Vendramin J. D., Silveira Neto S., and Cerignoni J. A.
(1995), Não preferência de Atta sexdens Forel, 1908
por espécies de Eucalyptus. Ecossistema 20, 87Ð92.