Development and laboratory evaluation of chemically-based baited ovitrap for the monitoring of Aedes aegypti



Aedes (Stegomyia) aegypti is considered to be the most important dengue vector worldwide. Studies were conducted to design and evaluate a chemically-based baited ovitrap for monitoring Ae. aegypti under laboratory conditions. Several known chemical attractants and three types of ovitraps (ovitraps A, B, and C) were evaluated throughout the oviposition bioassays. Oviposition responses of gravid female Ae. aegypti were evaluated to n-heneicosane, 3-methylindole (skatole), 4-methylphenol (p-cresol), and phenol. Female Ae. aegypti were attracted to all the evaluated compounds. Among them, n-heneicosane at a concentration of 10 ppm (mg/l), skatole from 50 to 1000 ppm, p-cresol at 100 ppm, and phenol at 50 ppm showed a significant positive oviposition response. A blend of the four chemical attractants increased the oviposition response; 67% of the eggs were deposited in the treatment compared to the control. Female Ae. aegypti were significantly more attracted to ovitrap A loaded with the four-component synthetic blend compared to the standard ovitrap in the oviposition bioassays. The compound used in ovitrap A retained its attractant property for up to three days. The chemically-based baited ovitrap may be considered as an option to be integrated during the monitoring of dengue virus vectors in México.


Dengue is the most rapidly spreading mosquito-borne viral disease in the world, increasing its geographic expansion in the last 50 years. An estimated 50 million dengue infections occur every year worldwide (Weaver and Reisen 2010). Various serotypes of the dengue virus are transmitted to humans through the bite of infected Aedes mosquitoes, principally Aedes (Stegomyia) aegypti Linnaeus (Diptera: Culicidae). The immature stages are found mostly in artificial water-filled containers such as vases, flower pots, bottles, cans, buckets, tires, cisterns, and water storage tanks which are closely associated with human dwellings and are often indoors (García-Rejón et al. 2011). Mexico is considered an endemic region for dengue since the mosquito vector Ae. aegypti and dengue transmission is present in more than 85% of the country (Moreno-Altamirano et al. 2009).

There are no antivirals or vaccines currently available for dengue treatment, with virus control strictly limited to mosquito eradication. In México, chemical control of the vector is the common strategy used to control the spread of the disease, but this has an adverse effect on the environment and non-targeted organisms. Resistance is also generated in the vector populations to the main ingredients used in insecticides. Therefore, the implementation of new strategies is needed, especially focusing on those that are within the ecological and community context. A broad knowledge of the vector's behavior can direct the site-specific measures, especially in those areas where they are most susceptible. Epidemiologically, the most important behaviors of the vectors are feeding and oviposition, and the repetition of these cycles ensures the presence of the vector and the cycles of transmission (Bentley and Day 1989). Site selection and subsequent oviposition behavior in mosquitoes is influenced by visual, tactile, and olfactory factors, with the latter considered of primary importance (Ponnusamy et al. 2008). Knowledge of these cues has led to the identification and incorporation of site-specific attractants that can function as ovitraps, which are used in dengue epidemiologic surveillance (Reiter et al. 1991, Chadee 2009).

The ovitrap has been shown to be an inexpensive, yet rapid and sensitive means of detecting Ae. aegypti in a given area, even in low density population areas (Fay and Perry 1965, Chadee 2009). Past studies have focused on texture, color, shape, and odors that must be included in an ovitrap in order to increase its effectiveness (Reiter et al. 1991, Ritchie et al. 2003, Lenhart et al. 2005, Chadee 2009). Research has also identified chemical attractants present in the Ae. aegypti breeding sites, whose origins are thought to come from eggs, larvae, bacteria, and predators present in oviposition sites (Hasselschwert and Rockett 1998, Mendki et al. 2000, Torres-Estrada et al. 2001, Ganesan et al. 2006, Ponnusamy et al. 2010). Baited oviposition traps with fermented organic infusions are attractive to gravid female mosquitoes (Reiter et al. 1991). In this regard, Millar et al. (1992) successfully isolated and identified five compounds found in areas associated with mosquito oviposition. These compounds are 3-methylindole, 4-methylphenol, 4-ethylphenol, phenol, and indole, which are the result of fermented bermuda grass (Cynodon dactylon). On the other hand, n-heneicosane was identified from the larval cuticle of Ae. aegypti (Mendki et al. 2000). Although these semiochemicals have been described as attractant to gravid mosquitoes, there are few studies that have systematically evaluated the effect of these synthetic compounds when placed in ovitraps as a single compound or as a mix of all of them. The objective of this study was to develop an ovitrap impregnated with chemical attractants to identify Ae. aegypti oviposition preferences and also to evaluate, under laboratory conditions, the effectiveness of the compound blends when added to the traps.



Aedes aegypti colonies were established from field-collected larvae from Tapachula, Chiapas, in 2010. Larvae were reared as described by Torres-Estrada et al. (2001) at 28 ± 1° C water temperature and a photoperiod of 12:12 (L:D). Bioassays were performed with the F2–F6 of Ae. aegypti. Female wing length averaged 2.6 ± 0.1 mm. Before the experiments, adults were kept in 30 cm3 screened cages at 27 ± 1° C, 70% RH, and fed 10% sugar soaked cotton pads. Mated females were blood-fed on rabbits and then allowed to develop their eggs before experimentation. Different batches of 30 gravid females Ae. aegypti were used in each bioassay and were between six to seven days old (96 h post-feeding).

Chemical test

Synthetic compounds were purchased from Sigma-Aldrich and J. T. Baker (Toluca, state of México, México). The purities were >98%. The compounds tested at different concentrations were diluted in dichloromethane (Golden Bell Reactive, Toluca, state of México). The experimental concentrations of each compound are described in Table 1; some of them have been previously reported (Allan and Kline 1995, Seenivasagan et al. 2009).

Table 1.  Oviposition response of females Ae. aegypti to different concentrations of chemical attractants. Thumbnail image of

Bioassay using oviposition attractants

Dual choice bioassays for oviposition were carried out in the laboratory as described by Sharma et al. (2008). Assays were performed in duplicate in two separate wire mesh-fitted cages of size 60 cm3. A plastic container (7×7×5 cm) filled with 50 ml of de-chlorinated water (Hycel®) was used as the oviposition site. To prepare the oviposition substrate, a piece of paper filter Whatman® No. 2 (27×5 cm) was impregnated with 20 μl of the test solution and smoothly adhered to the inner wall of the container with the paper half submerged and the lines on the paper horizontal. The oviposition substrate used as a control was Whatman® No. 2 filter paper impregnated with 20 μl of the dichloromethane solvent. To ensure that the filter paper was clean and free of extraneous odor, it was washed with dichloromethane and allowed to fully dry in a chemical extraction hood.

During evaluation of n-heneicosane attractant, filter paper without attractant was used as oviposition substrate in the containers. The chemical n-heneicosane functions as a tactile signal for Ae. aegypti and was added directly to the water (Seenivasagan et al. 2009). At each dose tested, 20 μl of the n-heneicosane solution was used in ten independent repetitions. The internal control for this treatment was 20 μl of the dichloromethane directly added to the water. Bioassays were conducted using 30 gravid females added to the cage.

Dual choice bioassays allowed the gravid females to choose between treated and untreated (control) oviposition sites. The oviposition sites were placed separately in each cage, diagonally in each corner at equal distance. In each experimental day, the containers were rinsed with tap water followed by dichloromethane. Each dose of compound was evaluated separately in duplicate (two different cages), and the experiment was repeated for five days with different batches of mosquitoes (ten replicates). During the evaluation of attractants, each was tested one dose at a time, in two different cages for five days. The number of eggs laid in control and treated oviposition substrates in a day were counted manually to assess the oviposition preference.

Oviposition activity was expressed using the oviposition activity index (OAI) of Kramer and Mulla (1979), which is calculated as follows:


where Nt= number of eggs deposited in treatment container and Nc = number of eggs deposited in control container. Responses using this system, range from 1 (attraction) to −1 (repellency) with zero denoting no preference. In this context, attraction refers to the deposition of eggs as the end result of a complex sequence of behaviors including orientation towards the source and stimulation to oviposit (Allan and Kline 1995).

Ovitrap design

Three types of ovitraps were designed (labeled as ovitraps A, B, and C in Figure 1) using different sizes of disposable plastic soda bottles that have been modified from previous studies (Fay and Perry 1965, Lenhart et al. 2005). Ovitrap A is a black-colored cylindrical container 11 cm in diameter with a holding capacity of 2.5 liters; it has a wide slot of 23.5 cm wide by 10 cm long. Ovitrap B is a black-colored cylindrical container with a diameter of 13 cm and a capacity of 3 liters, and three slots each 11 cm wide by 10 cm long. Ovitrap C is a black-colored cylindrical container with a diameter of 11 cm and holding capacity of 2.5 liters and three slots of 9 cm wide by 11 cm long. A standard ovitrap control was a 1 liter cup painted in black (Service 1993). Ovitraps filled with 500 ml of dechlorinated water were used as the oviposition site. Filter paper were used as oviposition substrate (without attractant). Paper was placed directly in the internal wall of the ovitraps. Before starting the study, ovitraps were painted black and kept at room temperature for several days to dry out and eliminate any paint-related odors. Bioassays were conducted using 30 gravid females added to the cage. Laboratory bioassays were performed in duplicates using two separate wire mesh-fitted cages 60×60×60 cm in size.

Figure 1.

Prototypes of ovitraps evaluated in the study.

Multiple choice bioassays were conducted. Females were allowed to choose between standard ovitrap and ovitraps A, B, and C. The four ovitraps were placed diagonally equidistant in each corner of the cages. The position of ovitraps was shifted in a clockwise rotation every day. Each experiment was evaluated separately in duplicate (two different cages) and was repeated for four days with different batches of mosquitoes (eight replicates total) (Sharma et al. 2008).

Oviposition response to attractant mixtures

Ovitrap A and compounds n-heneicosane at 10 ppm, skatole at 500 ppm, p-cresol at 100 ppm, and phenol at 50 ppm produced positive oviposition responses in previous bioassays. The dose of 500 ppm of skatole was chosen as the attractant mixture because the oviposition index was higher than with 50 ppm, 100 ppm, and 1,000 ppm (Table 1). The standard ovitrap was used as a control because the aim of the project was to demonstrate that our ovitrap A is the most effective and logistically easy to use ovitrap compared to the standard ovitraps currently used for the active surveillance of Ae. aegypti by Mexican health authorities. The attractive effect of the blend of synthetic chemicals baited in ovitrap A was compared against a standard ovitrap in dual choice bioassays (Sharma et al. 2008). Laboratory bioassays were performed in 60×60×60 cm wire mesh-fitted cages for the mosquito tests. Ovitraps filled with 500 ml of de-chlorinated water were used as the oviposition sites. Filter paper was used as the substrate for oviposition.

To prepare the dispenser semiochemical, a piece of filter paper was impregnated with 200 μl of the attractant mixture and was attached with paper clips to the inner wall of ovitrap A without touching the water (Figure 1). Each treatment consisted of an ovitrap A loaded with a dispenser (skatole at 500 ppm, p-cresol at 100 ppm, phenol at 50 ppm) and 1 ml of n-heneicosane at 10 ppm directly added to the water. The controls consisted of the standard ovitrap with a piece of filter paper impregnated with 200 μl of dichloromethane and 1 ml of the same solvent added to the water.

Bioassays were conducted with 30 gravid females added to the cage. The experimental design was completely randomized, each experiment was evaluated separately in duplicate (two different cages), and each was repeated for five days with a different batch of mosquitoes (ten replicates).

Residual effect of synthetic compounds

During the evaluation of the attractants mixture it was determined that ovitrap A was more attractive compared to the standard ovitrap. The test of the residual effect was performed in order to evaluate the lasting-time effectiveness to attract Ae. aegypti. Residuality was evaluated during every five days because it is the time interval that the Mexican health authorities checked the standard ovitraps in the field.

The residual effect of the attractant mixture was evaluated based on the methodology described by Seenivasagan et al. (2010). Experiments were performed simultaneously in two different 60×60×60 cm cages. For one cage, ten sets of ovitraps were filled with 500 ml of dechlorinated water. In the other cage, ten sets of ovitraps were also prepared. One set comprised a standard ovitrap control and ovitrap A (treatment) marked as day 1 and kept inside the cage, and the water level in each ovitrap was marked to observe any evaporation loss in subsequent days. A standard ovitrap was loaded with filter paper impregnated with 200 μl of dichloromethane and 1 ml of the same solvent added to the water. Ovitrap A was loaded with a dispenser with the attractant mix and 1 ml of n-heneicosane at 10 ppm directly added to the water. To prepare the semiochemical dispenser, a piece of filter paper was impregnated with 200 μl of the attractant mixture (skatole at 500 ppm, p-cresol at 100 ppm, phenol at 50 ppm) and was attached with paper clips to the inner wall of ovitrap A without touching the water. In total (2 × 10 = 20 sets), 20 sets of ovitraps were used. Treatments were done on the first day (day 1) and other sets of treatments were marked accordingly until the fifth day. In each experimental, the day one set of treated ovitraps were offered to a new batch of mosquitoes (30 gravid Ae. aegypti) after compensating for even minimum evaporative water loss in the substrates. The two ovitraps were placed diagonally equidistant in each corner of the cages. The experimental design was completely randomized in each bioassay; females were allowed to choose between a standard ovitrap vs ovitrap A.

Statistical analysis

All statistical analyses were performed using the PASW 18 software. Data was tested for normality using the Shapiro-Wilk test and homoscedasticity with test of Levene. The percentage of eggs registered in oviposition bioassays of dose-response for each chemical compound and blend were analyzed by a paired t-test, while those without normal distribution were analyzed with the Wilcoxon test. The percentages of eggs registered in the evaluation of the types of ovitraps were analyzed by one-way analysis of variance (ANOVA, p < 0.05). The proportion of registered eggs in the evaluation of residual effect was arcsine square root transformed and analyzed by two-way ANOVA in order to show that there is no statistical difference in the mean number of eggs collected per ovitrap per day (ovitrap A vs standard ovitrap). If p < 0.05, the null hypothesis was rejected.


Bioassay using oviposition attractants

The oviposition responses of Ae. aegypti gravid females exposed to different synthetic compounds are shown in Table 1. Ten ppm (mg/L) of n-heneicosane was attractive for oviposition and induced 55% of eggs laid in the containers with treatment (t=−2.455, df= 9, p= 0.036). The oviposition activity index (OAI) was 0.09, indicating that n-heneicosane is slightly attractive (Table 1). Skatole to 100 ppm (t=−3.281, df= 9, p= 0.010) and 500 ppm (z=−2.803, df= 9, p= 0.005) had the greatest effect on attraction (Table 1). The concentration of 500 ppm had the highest OAI (0.41). p-cresol was attractive for oviposition to 100 ppm, (t=−2.777, df= 9, p= 0.022). At this concentration, 61.5% of the eggs were deposited in the container treatment and OAI was 0.23 (Table 1). A concentration of 1,000 ppm of p-cresol acted as a repellent for oviposition. Finally, phenol was an oviposition attractant at 50 ppm (t=– 2.401, df= 9; p= 0. 040), with 58% of the eggs deposited in the container with the treatment (Table 1); the OAI was 0.16.

Ovitrap design

Although there was no statistical difference in the number of eggs recorded among the different sets of traps (F= 1.515, df= 31, p= 0.232), it was observed that Ae. aegypti laid the highest percentage of eggs in ovitrap A (28.62%) when compared with the rest of the ovitraps (B, C, and control). Ovitrap A and the standard ovitrap were selected to evaluate the effect of the chemical compounds because our hypothesis was that the addition of an attractant mixture in ovitrap A would increase the attraction of Ae. aegypti to the treatment ovitrap.

Aedes aegypti oviposition response to attractant mixture

Ovitrap A, baited with a mix of n-heneicosane, skatole, p-cresol, and phenol, elicited a significant response by Ae. aegypti compared to a standard ovitrap (t=−11.650; df= 9; p= 0.000). In ovitrap A baited with the blended compounds, 67% of the eggs were deposited (Figure 2). The OAI of 0.50 suggested that this combination of effects increased the attraction effect.

Figure 2.

Aedes aegypti oviposition response to attractant mixtures under laboratory conditions. Error bars (95% CI) on the number of eggs (%) with a different letter are significantly different (t-test paired, P<0.05, n= 10).

Residual effect of synthetic compounds

The residual effect of the synthetic blend of compounds was evaluated in ovitrap A (loaded with n-heneicosane, skatole, p-cresol, and phenol) and was compared against the standard ovitrap during a five-day trial. Two-way analysis (ANOVA) of the residual effect revealed that there was a statistical difference between traps (F= 172.325, df= 1, p= 0.000). There was no significant statistical difference in the number of eggs laid between days (F= 0.000, df=4, p=1.000). The interaction between the factors residual effect*traps was statistically significant (F=38.798, df=4, p=0.000) (Table 2). In this particular study we were interested to know the longest number of days up to which there is still attraction to the compounds. Using the paired t test, it was determined that the attractant effect of the chemical blend has a three-day residual effect (day 1, t=−14.373, df= 1, p=0.044; day 2, t=−16.372, df=1, p=0.039; day 3, t=−13.231, df=1, p=0.048), but this effect is lost by days 4 and 5 (t=1.802, df=1, p= 0.323; t=−1.645, df=1, p=0.348, respectively).

Table 2.  Result of two-way ANOVA (n= 5) for the residual effect of the blend of chemical attractants on oviposition response of female Ae. aegypti under laboratory conditions (proportion of eggs were transformed to arcsine square root).
Source SS df F P
Residual effect0.00040.0001.00
Residual effect * Traps0.078438.7980.00


It was determined that the ovitraps with chemical attractants evaluated in this study elicited higher oviposition responses of Ae. aegypti compared with the standard ovitrap and indicates that chemically-attractive ovitraps may be an option to be considered in the epidemiological surveillance of the dengue vector. Ovitrap A incorporates physical and chemical cues and in the future could serve as a ‘pull’ component in a strategy for mosquito ‘attract and kill,’ similar to what has been developed and applied mainly to control pests of agricultural importance such as sticky ovitraps (Ritchie et al. 2003).

In nature, gravid females of Ae. aegypti are capable of detecting chemical cues produced in breeding sites to ensure the successful development of their progeny (Bentley and Day 1989, Ponnusamy et al. 2008). Knowledge of chemical compounds that serve as these cues has been important in the development of effective ovitraps. The compounds tested were selected from previous studies (Mendki et al. 2000, Seenivasagan et al. 2009, Allan and Kline 1995) and have been reported in the literature as oviposition attractants for this mosquito species. The compound n-heneicosane has been identified and isolated from the larval cuticle of Ae. aegypti and was evaluated to 10 and 69 ppm with attractant effect to Ae. aegypti (Mendki et al. 2000, Seenivasagan et al. 2009). In this study, tests were conducted with concentrations of compounds ranging from 1 to 100 ppm, of which only the 10 ppm of concentration presented an attractive effect. Seenivasagan et al. (2009) reported a repellent effect at 50 ppm or greater of n-heneicosane. The dose-dependent effect of this compound has been observed in different mosquito species, including Ae. aegypti, Ae. albopictus, and Anopheles albimanus, with previous reports observing a significant reduction in the deposition of eggs to high concentrations of different chemical compounds (Ikeshoji 1975, Knight and Corbet 1991, Allan and Kline 1995, Torres-Estrada et al. 2005, Ganesan et al. 2006, Sharma et al. 2008).

Some chemical attractants involved in oviposition have been as well identified in plants, defense secretions of insects, and feces (Bentley et al. 1979, Millar et al. 1992). We showed that gravid females of Ae. aegypti responded to the attractants skatole, p-cresol, and phenol. Allan and Kline (1995) reported attraction to phenol for this species of mosquito and p-cresol is an oviposition attractant for Aedes triseriatus (Bentley et al. 1979, 1981), Stomoxys calcitrans (Jeanbourquin and Guerin 2007) and Ae. albopictus (Allan and Kline 1995). Finally, skatole has been reported as an oviposition attractant for Culex quinquefasciatus Say (Millar et al. 1992), Ae. albopictus (Allan and Kline 1995), Anopheles gambiae s.s. (Blackwell and Johnson 2000), Toxorhynchites moctezuma, Toxorhynchites amboinensis (Collins and Blackwell 2002), and S. calcitrans (Jeanbourquin and Guerin 2007).

In previous work related to oviposition attractants, there is no mention of the effect of high concentration of those compounds in the oviposition behavior of the mosquitoes. Here, Ae. aegypti responded to a wide range of concentrations of skatole (50–1000 ppm), similar to that reported in the traps used to collect unfed females of Ae. aegypti that release CO2 odor plumes up to 8,700 ppm (Cooperband and Cardé 2006).

This study also showed that the attractive effect of the blend of compounds is greater in Ae. aegypti. In another study, Ae. albopictus and Ae. aegypti did not respond to the blend of skatole, p-cresol, phenol, 4-ethylphenol, and indol (Allan and Kline 1995). Millar et al. (1992) demonstrated that Cx. quinquefasciatus is stimulated to lay eggs in the blend of phenol, p-cresol, 4-ethylphenol, indole, and skatole. Mboera et al. (1999) also determined that Cx. quinquefasciatus had the greatest attraction to the blend of skatole plus the oviposition pheromone of Cx. quinquefasciatus ((5R, 6S)-6-acetoxy-5-hexadecanolide).

Ovitraps have been widely used to detect the presence of Ae. aegypti (Chadee 2009) and it has been observed that the addition of hay infusion in the water used in the ovitrap significantly increased the number of eggs counted (Polson et al. 2002, Santana et al. 2006, Ponnusamy et al. 2010). However, it has been observed that infusions are unstable dynamic systems that can switch from attractive to repellent effects. Operationally, it is considered more appropriate to apply synthetic chemical attractants because of their stability in the environment and the easy formulations that are accurate and reproducible (Millar et al. 1992).

When working with attractants, it is important to determine the optimal concentration that evokes the desired behavior in the insect and still remains in the environment. Therefore, the dispenser of the semiochemical is important to ensure the continuous release of the attractant (Byers 1998). The rate at which chemical compounds are released is influenced by temperature, rain, wind, and dust, therefore having an impact on the residual effect of the attractant (Barrera et al. 2006). In this study, we were able to design an ovitrap with the special feature of holding an attractant dispenser. Our design has the capability to avoid water and dust contamination, minimizing the effect of these external environment factors on the attractants placed in the ovitrap. In this particular study, the attractant residual effect lasted for three days. It is important to keep evaluating suitable dispensers that could improve residual effect of the attractants. On the other hand, Mboera et al. (1999) reported a residual effect of a week in the blend of skatole plus the oviposition pheromone of Cx. quinquefasciatus ((5R, 6S)-6-acetoxy-5-hexadecanolide). Similarly, the oviposition repellent of Ae. aegypti (hexadecyl pentanoate) retained its effect for up to one week (Seenivasagan et al. 2010).

Ovitrap A has the following advantages: it is recycled and removed from the natural environment, it is inexpensive (cost of 100 ovitraps A is US$12.10) and easy to use, it requires a short time to set the synthetic attractants, and the compounds are prepared with a consistent formulation. One of the limitations of the system is the short lasting effect of the attractants which is only three days. It is required to identify a suitable way to generate a low-release mechanism for the attractants.

In conclusion, these results showed that the ovitrap evaluated in the laboratory and baited with a blend of chemical attractants has the potential to be used in monitoring and control populations of Ae. aegypti. In future studies it is important to determine a better dispenser and the optimal release rate of the chemical compounds to make the designed ovitrap more effective.


We thank Francisco Santoyo, Farah Vera, Octavia Medina, Gabriel Fuentes, Laura Noriega, and Josué Espinoza for technical assistance. The study was supported by CONACYT under the project COCYTECH FOMIX-78917 “Development and evaluation of an ovitrap baited with chemical attractants for monitoring and control of dengue vector a medium-scale.”