J Chem Ecol (2006) 32: 1625–1634

DOI 10.1007/s10886-006-9097-9

Geographic Variation in Attraction to Human Odor

Compounds by Aedes aegypti Mosquitoes (Diptera:
Culicidae): A Laboratory Study

Craig R. Williams & Scott A. Ritchie &

Richard C. Russell & Alvaro E. Eiras &
Daniel L. Kline & Martin Geier

Received: 26 July 2005 / Revised: 4 January 2006 /

Accepted: 11 January 2006 / Published online: 26 July 2006
# Springer Science + Business Media, Inc. 2006

Abstract Previous investigations of Aedes aegypti response to human odor components

have revealed a number of compounds that attract host-seeking females. However, such
studies have utilized only a small number of long-term laboratory Ae. aegypti colonies.
Using laboratory y-olfactometers, we studied the attraction of four different Ae. aegypti
populations (North Queensland, Australia; Florida, USA; Singapore; and Minas Gerais,
Brazil) to a key attractant compound from human skin, lactic acid. Combinations of lactic
acid with ammonia and a fatty acid (caproic acid) were also investigated. The aims were to
determine the extent of variation in lactic acid dose response among populations and to see
whether all four populations responded equally to combinations of human odor
components. Although all Ae. aegypti populations were attracted to lactic acid, there were
differences in the threshold dose: Florida 0.03 μg/min, Singapore 0.17 μg/min, North
Queensland 1.92 μg/min, and Brazil 10.27 μg/min. Attraction to lactic acid alone

C. R. Williams (*) : S. A. Ritchie

School of Public Health and Tropical Medicine, James Cook University, P.O. Box 6811, Cairns,
Queensland 4870, Australia
e-mail: craig.williams1@jcu.edu.au

S. A. Ritchie
Tropical Public Health Unit, Queensland Health, P.O. Box 1103, Cairns, Queensland 4870, Australia

R. C. Russell
Department of Medical Entomology, University of Sydney, ICPMR, Westmead Hospital, Westmead,
NSW 2145, Australia

A. E. Eiras
Department of Parasitology, Federal University of Minas Gerais (UFMG), Belo Horizonte, Brazil

D. L. Kline
United States Department of Agriculture, ARS-CMAVE, 1600 SW 23rd, Gainesville, FL, USA

M. Geier
Institut für Zoologie, University of Regensburg, Universitätstrasse 31, 93040 Regensburg, Germany
1626 J Chem Ecol (2006) 32: 1625–1634

(maximum <40%) was significantly lower than for human odor (>87% for all populations).
Significant increases in attraction were observed when lactic acid was combined with
ammonia or caproic acid, although not for all populations. In addition, the highest doses of
caproic acid tested decreased attraction when combined with lactic acid. The divergent
responses to host kairomones seen here may be evidence of adaptation to locally available
hosts in different parts of the geographic range of Ae. aegypti.

Keywords Aedes aegypti . Ammonia . Caproic acid . Host-seeking . Lactic acid .

Y-olfactometer

Introduction

The yellow fever mosquito Aedes aegypti (L.) is a prominent vector of dengue viruses
throughout tropical regions, resulting in human illnesses ranging from mild fever to fatal
hemorrhagic disease (Gubler and Kuno, 1997). The role of Ae. aegypti as a dengue vector is
related to its ability to breed in water-filled habitat associated with human habitation and its
preference for human blood hosts (Gubler and Kuno, 1997). This host preference is
facilitated by strong olfactory responses to human skin emanations (Skinner et al., 1965;
Schreck et al., 1990; Geier and Boeckh, 1999).
The surveillance of Ae. aegypti populations has traditionally relied on labor-intensive
methods, such as the sampling of immature stages from breeding sites, and on the use of
ovitraps that sample the egg-laying portion of the population. However, some of the indices
obtained lack epidemiological sensitivity (Focks, 2003). Sampling techniques for adult Ae.
aegypti, such as human-biting catch and aspiration samples, can be unsafe and labor
intensive for operators (Focks, 2003). Hence, there is a need to develop efficient sampling
methods for adult Ae. aegypti that can be readily related to dengue epidemiology. The
development of traps that specifically attract host-seeking Ae. aegypti through the use of
kairomone lures would provide a more efficient and epidemiologically relevant alternative,
and possibly augment current population reduction methods.
A number of compounds identified from human skin emanations attract host-seeking Ae.
aegypti. Lactic acid was identified as a key kairomone for Ae. aegypti host finding (Acree
et al., 1968; Geier and Boeckh, 1999), given that the addition and removal of this
compound from human skin extracts respectively increases and decreases attraction in
laboratory olfactometers (Geier et al., 1996; Steib et al., 2001). However, it is much less
attractive alone than whole human odor extracts (Eiras and Jepson, 1991; Geier et al.,
1996), indicating that other kairomones are important in Ae. aegypti host finding. The
combination of lactic acid with carbon dioxide creates a powerful attractive synergism
(Acree et al., 1968; Smith et al., 1970; Eiras and Jepson, 1991) as do other compounds from
human skin, namely ammonia (Geier et al., 1999a), fatty acids of various chain length
(Bosch et al., 2000), dimethyl disulfide, and acetone (Bernier et al., 2003). The response to
lactic acid alone or when combined with other kairomones is dose dependent (Geier et al.,
1996, 1999a; Bosch et al., 2000). However, the importance of such compounds in nature is
not as certain, so extrapolation from laboratory studies must be performed with caution,
especially as all research to date has been carried out with a single Ae. aegypti strain that
has been reared in the laboratory for several decades.
To develop attractive kairomone blends for global use in Ae. aegypti trapping, the
response of different populations to key kairomones should be assessed to evaluate
J Chem Ecol (2006) 32: 1625–1634 1627

geographic variability in responses and to determine whether specific blends are needed for
different populations. Furthermore, such a study may detect differences in the sensory
physiology of Ae. aegypti throughout its geographic range. Given that genetic differenti-
ation may exist between Ae. aegypti populations only a few kilometers apart (Tabachnick
and Powell, 1978; Paupy et al., 2004), and that other mosquito species show heterogeneity
in odor-mediated host preferences throughout their geographic range (e.g., Williams et al.,
2003), it seems likely that Ae. aegypti populations could become adapted to particular
kairomones exuded by locally available resources (i.e., humans). Any geographic
variability in human odor profiles could facilitate such adaptation, which would be
manifested in divergent responses to certain kairomones.
The aims of this study were to describe the response of host-seeking Ae. aegypti from
four populations of distinct geographic origin to host kairomones, using laboratory
y-olfactometers. The response of each population to increasing doses of lactic acid alone as
well as in combination with ammonia or a fatty acid (caproic acid) was to be established. A
combination of all three compounds forms a proprietary blend (Geier and Eiras, 2003;
BioGents GmbH, Regensburg, Germany) and is the focus of a subsequent study.

Methods and Materials

Mosquito Rearing Methodology

Four populations of Ae. aegypti mosquitoes, which varied in their laboratory colonization
history, were used in these studies at the University of Regensburg (September–November
2004; Table 1). Although it would have been ideal for all populations to be tested as field F1
generations, this was not practical. Larvae were fed with Tetramin fish food. Adults were
maintained in plastic containers (50 × 40 × 25 cm) at 26–28°C, 60–70% relative humidity,
light/dark 12:12 hr, and with continuous access to a 10% glucose solution.

Olfactometers

A series of four identical, Plexiglas®, y-olfactometers (Geier and Boeckh, 1999) on white
tables was used (Fig. 1). Eighty l/min of charcoal-filtered, heated (27.6 ± 0.1°C), and
humidified (63.7 ± 0.6% relative humidity) air was passed through two parallel stimulus
chambers, in which test compounds mixed with the air. These two parallel airstreams
(0.2 m/sec) then passed through the common rectangular branch of the olfactometer and
into the stem (0.4 m/sec) that was attached to the release chamber holding the mosquitoes.
Rotating screens in the release chamber and in both arms of the upwind end enabled the
release and recapture of mosquitoes.

Table 1 Summary data for Ae. aegypti populations used in this study

Population Source Length of time in laboratory

North Queensland, Australia Cairns Nil (F1 generation)

Florida USA Orlando Since ca. 1960 (regularly supplemented
with field material until ca. 1992)
Minas Gerais, Brazil Belo Horizonte Nil (F1 generation)
Singapore Singapore ca. 20 generations
1628 J Chem Ecol (2006) 32: 1625–1634

Fig. 1 Schematic drawing of the y-olfactometer design used to test geographically disparate populations of
Ae. aegypti (Geier and Boeckh, 1999). Dimensions are given in millimeters

Volatile Stimulus Delivery

For each test, volatile stimuli were introduced into one of the two parallel stimulus
chambers, with the other remaining as a blank control. Air was passed through a 200-ml
Erlenmeyer flask containing 20 ml of the compound under study, and carried the headspace
volatiles to ports immediately proximal to the treatment arms of the olfactometers by Teflon
tubing (Geier et al., 1999a; Bosch et al., 2000). The delivery rate of volatile compounds to
the olfactometers was determined by the airflow rate through the flask. When combinations
of compounds were tested, each was delivered through separate flasks and tubing that met
at a single release point in the olfactometer.

Bioassay and Statistical Analyses

Previously published methods (Geier et al., 1999a; Bosch et al., 2000) were used. Groups
of 17–25 avid female Ae. aegypti (10–20 d postemergence, not blood fed, nulliparous) were
lured from their holding cages by human odor, thereby selecting host-seeking mosquitoes
for tests. This was achieved by fitting the y-olfactometer release cage onto a specially
designed fitting on the holding cage, and creating airflow with a small fan that carried
human odor from a hand through the release cage into the holding cage. Attracted
mosquitoes then entered the release cage, which was then removed.
Before tests, the release cage containing mosquitoes was attached to the olfactometer.
Mosquitoes were allowed 20 min to acclimatize, with clean air flushing through. When
odor stimuli were introduced, mosquitoes were released from their chamber and allowed
30 sec to respond, after which the rotating screens were closed, trapping those in the
treatment and control arms, and in the release cage. Odor stimuli were tested in balanced,
randomized block designs. Each experiment involved the use of all four Ae. aegypti strains,
J Chem Ecol (2006) 32: 1625–1634 1629

allocated to y-olfactometers in a randomized block design. Those captured in the treatment

arm of the olfactometer were deemed to have been attracted to the stimulus. The small
numbers captured in the control arm were also recorded. Mosquitoes were lured back into
their release cage in response to human odor (hand of CRW) by reversing the air flow.
Groups of mosquitoes were retested up to five times in this manner, with a 30-min “rest”
period between tests, during which clean air continued to flow through the olfactometer.
This method verified that mosquitoes would readily respond to natural human odors in
between each trial. Previous studies with this apparatus have demonstrated that there are no
carryover effects of contamination or behavioral conditioning in between replicate trials
(Geier 1995; Geier and Boeckh, 1999).
Furthermore, previous studies have shown that a single laboratory strain of Ae. aegypti
will respond in a consistent manner to a variety of kairomones (M. Geier and C. Aigner,
unpublished data, University of Regensburg, 2001). This was demonstrated by testing
cohorts from the same colony repeatedly over 5 mo in the same y-olfactometers used here.
Percentage data (number of mosquitoes in the test arm as a percentage of those released)
were arcsine-transformed to normality before ANOVA with least significant difference
(LSD) post hoc tests performed using SPSS statistical software (Release 11.0.1, 2003,
SPSS, Inc., Chicago, IL, USA).

Experiment 1: Lactic Acid

Lactic acid (analytical reagent grade, Merck, Germany) dose was manipulated by adjusting
flow rate through the Erlenmeyer flask. Five flow rates ranging from 0.2 to 300 ml/min
were used. The size of the lactic acid dose (in μg/min) was determined by using calibration
values taken from Geier et al. (1999b).
Human odor was tested as a gold standard control to determine the optimal behavioral
response of each population in the y-olfactometers. This was introduced into stimulus
chambers by inserting an index finger (always that of CRW) through a snug-fitting hole. A
20 ml water control (air flow: 10 ml/min) was also used to ascertain the baseline level of
upwind flight. A minimum of 10 replicate trials was performed for each population for each
treatment. Human odor was used despite the risk of subtle variations in kairomone output.
The balanced, randomized block design used here ensured that all four Ae. aegypti
populations would be subject to the same variations in output from the human hand,
thereby controlling for this factor. Human hands have been previously used as an odor
source in similar experiments (e.g., Geier and Boeckh 1999; Geier et al. 1999b).
The lowest lactic acid dose required to cause attraction significantly greater than the
water control is hereafter termed the “attraction response threshold” dose. Means were
separated using LSD post hoc tests.

Experiment 2: Lactic Acid + Ammonia, and Experiment 3: Lactic Acid + Caproic Acid

The attraction of a fixed dose of lactic acid (determined from experiment 1) with four 10-
fold increasing flow rates of ammonia (experiment 2) or caproic acid (experiment 3)
ranging from 0.3 to 300 ml/min was investigated. Unlike for lactic acid, we could not
determine the ammonia and caproic acid doses in micrograms per minute because of a lack
of calibration data. Ammonia (analytical reagent grade, Merck, Germany) was a 25%
solution of ammonium hydroxide further diluted 1:100 in water. Caproic acid (Fluka
Chemika, Germany) was also of analytical reagent quality.
1630 J Chem Ecol (2006) 32: 1625–1634

Results

Experiment 1

All four mosquito populations responded to both lactic acid and human odor (Fig. 2). In all
cases, the responses to human odor were significantly higher than to any concentration of
lactic acid tested. Mean percentage capture in the control arm was ≤6.4% for the lactic acid
treatments.
Each of the populations demonstrated different attraction response thresholds. These
varied from 0.03 μg/min for Florida USA (F = 68.86, d.f. = 6, P < 0.001) to 0.17 μg/min for
Singapore (F = 35.964, d.f. = 6, P < 0.001), 1.92 μg/min for North Queensland Australia
(F = 57.04, d.f. = 6, P < 0.001), and 10.27 μg/min for MG Brazil Ae. aegypti (F = 66.42,
d.f. = 6, P < 0.00; Fig. 2).
There is no one common flow rate that is more attractive than others for all populations.
Therefore, a dose of 50 ml/min (10.27 μg/min) was chosen for experiments 2 and 3,
because it was greater than or equal to the attraction response threshold dose for all groups.

Experiment 2

For the North Queensland Australia population, there was no statistically significant
increase in attraction when ammonia was added to lactic acid (F = 1.86, d.f. = 4, P = 0.131;
Fig. 3), in contrast to the marked increase in attraction seen with the Florida USA
population (F = 12.97, d.f. = 4, P < 0.001). For the MG Brazil and Singapore populations,
lactic acid and ammonia combined provided greater attraction than lactic acid alone (F =
2.59, d.f. = 4, P = 0.048 for both) although only at the lower ammonia concentrations
(Fig. 3).

Fig. 2 Mean percentage attraction (±SE) of four populations of Ae. aegypti to various doses of lactic acid, as
well as human odor and water. Different letters indicate significant differences within a population
J Chem Ecol (2006) 32: 1625–1634 1631

Fig. 3 Mean percentage attraction (±SE) of four Ae. aegypti populations to mixtures of lactic acid (LA) at
10.3 μg/min and ammonia (NH3) in y-olfactometers. Different letters indicate statistically significant
differences within a population

When tested alone (0.3 ml/min), ammonia proved slightly attractive to some
populations: North Queensland Australia 28.3 ± 3.9%, Florida USA 19.7 ± 6.2%,
Singapore 23.1 ± 3.9%, and MG Brazil 8.1 ± 2.7%. These data enabled calculation of
the expected additive attraction (E ) for lactic acid and ammonia (Colby 1967). These
values were then compared with the observed attraction percentage (O ± 95% confidence
interval) to determine the presence of synergistic attraction. Comparisons were not
performed for North Queensland Australia Ae. aegypti, as ammonia did not improve
attraction for this group (Fig. 3). Singapore (E = 51.1%, O = 54.8 ± 11.7%), Florida USA
(E = 41.3%, O = 48.7 ± 8.1%), and MG Brazil (E = 17.1%, O = 23.4 ± 9.6%) Ae. aegypti
all showed slightly greater attraction (O) than the expected additive effect (E ). However,
overlapping 95% confidence intervals were suggestive of an additive rather than synergistic
effect.

Fig. 4 Mean percentage attraction (±SE) of four populations of Ae. aegypti to lactic acid (LA) at 10.3 μg/min
and caproic acid (CA) in y-olfactometers. Different letters indicate significant differences within a
population
1632 J Chem Ecol (2006) 32: 1625–1634

Experiment 3

Lactic acid plus caproic acid (30 ml/min) significantly increased attraction over lactic acid
alone or with caproic acid at 0.3 or 3 ml/min, in all populations except Singapore (Fig. 4).
Increasing the concentration of caproic acid to 300 ml/min caused a significant decline in
attraction, compared with lactic acid plus caproic acid (30 ml/min) in all populations, and
was less attractive than lactic acid alone for Florida USA and Singapore populations. (F =
11.19, d.f. = 4, P < 0.001).
Caproic acid alone (0.3 ml/min) showed little or no attraction for Ae. aegypti: North
Queensland Australia 16.2 ± 7.8%, Florida USA 3.2 ± 1.9%, Singapore 7.8 ± 5.9%, and
MG Brazil 6.5 ± 2.6%. Furthermore, caproic acid did not improve the attraction of lactic
acid at this flow rate (0.3 ml/min, Fig. 4) and no synergism calculations were possible.

Discussion

Geographically disparate populations of Ae. aegypti responded to lactic acid but differed in
their sensitivities. The attraction response threshold dose for Florida USA and Singapore
strains was below the lowest rate of lactic acid emission from human hands (0.38 μg/min;
Smith et al., 1970). Human individuals differ in their level of emission of L-lactic acid from
their skin by at least a factor of 5 (Acree et al., 1968; Smith et al., 1970; Steib et al., 2001).
However, we do not know how big the differences in the lactic acid emission rates are
between different geographic subpopulations of human hosts. MG Brazil mosquitoes
showed a much lower response to lactic acid at any dose when compared with other
populations. Given the avid response to human odor by this population (Fig. 4), the low
response to lactic acid cannot be attributed to some behavioral anomaly hampering
y-olfactometer flight. Rather, it indicates that other kairomones may be more important in
host finding for this population.
The results presented here concur with those of Geier et al. (1996), who demonstrated
that increasing the dose of lactic acid past the attraction response threshold does not
significantly increase the attraction of Ae. aegypti (Bayer strain). The level of attraction (ca.
25%) reported in Geier et al. (1996) was similar to the level reported for three of the four
populations studied here. Also, a similar mean level of attraction (20.5%) was reported by
Bernier et al. (2003) for Ae. aegypti from the same Florida USA laboratory population by
using similar apparatus. This finding demonstrates consistency in response of this Ae.
aegypti strain to lactic acid when tested in different laboratories. This consistency is further
illustrated by similar responses to a human hand in both laboratories (this study; Kline
1998).
The greater attraction of the lactic acid–ammonia combination compared with lactic acid
alone, first described for the Bayer strain of Ae. aegypti by Geier et al. (1999a), has been
confirmed here. An exception is the North Queensland Australia population in which the
lack of effect may reflect the good response of North Queensland Australia Ae. aegypti to
lactic acid alone, or that ammonia is not an important attractant for this population. A
significant combination effect was demonstrated for the MG Brazil population, although its
attraction both to lactic acid and the lactic acid–ammonia combination was poor. This
possibly reflects a poor response to lactic acid in general. The attraction to ammonia when
presented alone, up to ca. 28% (North Queensland Australia), is much greater than the ca.
5% presented by Geier et al. (1999a). This may be attributed to behavioral variability
between Ae. aegypti populations. The increased attraction of lactic acid with the addition of
J Chem Ecol (2006) 32: 1625–1634 1633

caproic acid, first described by Bosch et al. (2000), has been confirmed here for at least
three Ae. aegypti populations, although too much caproic acid may cause reduced
attraction.
Because of the varied colonization history of the populations studied here, and the
significant genetic differentiation that exists even between Ae. aegypti populations
separated by just a few kilometers (Tabachnick and Powell, 1978; Paupy et al., 2004),
conclusions about naturally occurring heterogeneity in response to attractants throughout
the geographic range of Ae. aegypti must be made with caution. Nor can generalizations be
made about Ae. aegypti response to kairomones from particular regions, as population
genetic studies would be needed to determine how representative these laboratory
populations are for each region.
Groups of Ae. aegypti may respond differently to the same volatile compounds, and
statements concerning responses to kairomone combinations must be specific to the
population in question. The divergent responses to host kairomones seen here may be
evidence of adaptation to locally available hosts in different parts of the geographic range of
Ae. aegypti. However, a study of human odor components in different regions (particularly
lactic acid emission levels), paired with population genetic studies and behavioral responses
of mosquito strains from different regions, is needed to address such a hypothesis. The
results presented here indicate that the development of trapping lures for host-seeking Ae.
aegypti may require some specialization for the region of use. The validity of laboratory
olfactory studies in terms of field behavior has been demonstrated directly for at least one
mosquito species (Culex annulirostris in Australia; Williams et al., 2003). Nonetheless, we
report here the response of Ae. aegypti to human odorants over a very short range in an
artificial environment. It follows that field trials are required to validate the attraction of Ae.
aegypti to these odorants. Last, it is clear that human odor is significantly more attractive to
Ae. aegypti than any of the odorants tested here, confirming that a complex suite of
kairomones is required to maximize the attraction of this species.

Acknowledgments Ulla Kroeckel, Thomas Hoerbrand, Sebastian Haas, and Margit Lindner provided
technical assistance at the University of Regensburg. Carl Baptista from the Origin Exterminators PTE, Ltd.,
Singapore provided the Singapore mosquito colony. These experiments adhered to James Cook University
guidelines for the use of human subjects in experimentation. This work was funded by Australian National
Health and Medical Research Council grant 279401 (to SAR and RCR).

References

ACREE, F., JR., TURNER, R. B., GOUCK, H. K., BEROZA, M., and SMITH, N. 1968. L-Lactic acid: A mosquito
attractant isolated from humans. Science 161:1346–1347.
BERNIER, U. R., KLINE, D. L., POSEY, K. H., BOOTH, M. M., YOST, R. A., and BARNARD, D. R. 2003.
Synergistic attraction of Aedes aegypti (L.) to binary blends of L-lactic acid and acetone, dichloro-
methane, or dimethyl disulfide. J. Med. Entomol. 40:653–656.
BOSCH, O. J., GEIER, M., and BOECKH, J. 2000. Contribution of fatty acids to olfactory host finding of
female Aedes aegypti. Chem. Senses 25:323–330.
COLBY, S. R. 1967. Calculating synergistic and antagonistic responses of herbicide combinations. Weeds
15:20–22.
EIRAS, A. E. and JEPSON, P. C. 1991. Host location by Aedes aegypti (Diptera: Culicidae): a wind tunnel
study of chemical cues. Bull. Entomol. Res. 81:151–160.
FOCKS, D. A. 2003. A Review of Entomological Sampling Methods and Indicators for Dengue Vectors.
TDR, WHO, Geneva, Switzerland.
GEIER, M. 1995. Verhaltensversuche mit Gelbfiebermücken Aedes aegypti zur Aufklärung des attraktiven
Reizmusters bei der olfaktorischen Wirtsfindung. PhD thesis, University of Regensburg.
1634 J Chem Ecol (2006) 32: 1625–1634

GEIER, M. and BOECKH, J. 1999. A new Y-tube olfactometer for mosquitoes to measure the attractiveness of
host odours. Entomol. Exp. Appl. 92:9–19.
GEIER, M., BOSCH, O. J., and BOECKH, J. 1999a. Ammonia as an attractive component of host odour for the
yellow fever mosquito, Aedes aegypti. Chem. Senses 24:647–653.
GEIER, M. BOSCH, O. J. and BOECKH, J. 1999b. Influence of odour plume structure on upwind flight of
mosquitoes towards hosts. J. Exp. Biol. 202:1639–1648.
GEIER, M. and EIRAS A. E. 2003. Composition for attracting blood sucking arthropods. Worldwide patent no.
WO 03/103395 A1. Applicant: University of Regensburg.
GEIER, M. SASS, H., and BOECKH, J. 1996. A search for components in human body odour that attract
females of Aedes aegypti, pp 132–148, in G. R. Bock and G. Cardew (eds.), Olfaction in Mosquito–Host
Interactions. Wiley, Chichester, United Kingdom.
GUBLER, D. and KUNO, G. 1997. Dengue and Dengue Hemorrhagic Fever. CAB International, Fort Collins,
CO, USA.
KLINE, D. L. 1998. Olfactory responses and field attraction of mosquitoes to volatiles from limburger cheese
and human foot odor. J. Vector Ecol. 23:186–194.
PAUPY, C., CHANTHA, N., HUBER, K., LECOZ, N., REYNES, J.-M., RODHAIN, F., and FAILLOUX, A.-B. 2004.
Influence of breeding sites features on genetic differentiation of Aedes aegypti populations analyzed on a
local scale in Phnom Penh municipality of Cambodia. Am. J. Trop. Med. Hyg. 71:73–81.
SCHRECK, C. E., KLINE, D. L., and CARLSON, D. A. 1990. Mosquito attraction to substances from the skin of
different humans. J. Am. Mosq. Control Assoc. 6:406–410.
SKINNER, W. A., TONG, H., PEARSON, T., STRAUSS, W., and MAIBACH, H. 1965. Human sweat components
attractive to mosquitoes. Nature 207:661–662.
SMITH, C. N., SMITH, N., GOUCK, H. K., WEIDHASS, D. E., GILBERT, I. H., MAYER, M. S., SMITTLE, B. J.,
and HOFBAUER, A. 1970. L-Lactic acid as a factor in the attraction of Aedes aegypti (Diptera: Culicidae)
to human hosts. Ann. Entomol. Soc. Am. 63:760–770.
STEIB, B. M., GEIER, M., and BOECKH, J. 2001. The effect of lactic acid on odour-related host preference of
yellow fever mosquitoes. Chem. Senses 26:523–528.
TABACHNICK, W. J. and POWELL, J. R. 1978. Genetic structure of the East African domestic populations of
Aedes aegypti. Nature 272:535–537.
WILLIAMS, C. R., KOKKINN, M. J., and SMITH, B. P. 2003. Intraspecific variation in odour-mediated host
preference of the mosquito Culex annulirostris. J. Chem. Ecol. 29:1861–1879.