Journal of Chemical Ecology, VoL 19, No. 4, 1993
SYNERGISM BETWEEN CHEMICAL ATTRACTANTS AND VISUAL CUES INFLUENCING OVIPOSITION O F T H E M O S Q U I T O , Culex quinquefasciatus (DIPTERA: CULICIDAE)
J.W. BEEHLER,* J.G. MILLAR, and M.S. MULLA Department of Entomology University of California Riverside, California 92521 (Received September 23, 1992; accepted November 19, 1992)
Abstract--Physical factors, such as the color of the oviposition substrate, have seldom been compared with chemical cues in their ability to elicit oviposition behavior in mosquitoes. The role of dyed oviposition waters in attracting ovipositing Culex quinquefasciatus was examined in laboratory experiments. Oviposition waters dyed with ink were found to be significantly attractive to ovipositing Cx. quinquefasciatus when compared to distilled water controls. Experiments demonstrated that the mosquitoes were responding to the increased optical density of the dyed oviposition water rather than volatile components of the dye. Ink was also considered in combination with chemical oviposition cues. No comparative data exist on the effect of physical and chemical factors presented together on the oviposition behavior of Cx. quinquefasciatus. Waters dyed with ink acted synergistically with a five-component chemical attractant mixture (3-methylindole, indole, 4-methylphenol, 4-ethylphenol, and phenol) in inducing oviposition in a 2 x 2 factorial experiment. Key Words--Insecta, Diptera, Culicidae, Culex quinquefasciatus, oviposition behavior, attractants, optical density, visual cues.
There has been great interest in describing and quantifying physicochemical and biotic factors involved in the oviposition site selection of gravid mosquitoes. A thorough knowledge of these factors and the interactions between them is crucial in acquiring an understanding of the oviposition process. The thrust of much of *To whom correspondence should be addressed. 635 0098-0331/93/0400-0635507.00/0 9 1993PlenumPublishingCorporation
BEEHLER ET AL.
the research into mosquito oviposition behavior has been toward the identification of chemical compounds that attract or stimulate gravid female mosquitoes (Bentley and Day, 1989). In natural situations these chemical compounds do not act independently. They interact with other biotic and abiotic factors at the oviposition site, and in sum they present an integrated picture of the quality and attractiveness of the oviposition site. Organic infusions have been commonly used in mosquito oviposition studies. Kramer and Mulla (1979) found that some organic infusions were attractive to ovipositing Culex mosquitoes while others were repellent. An alfalfa hay infusion (Reiter, 1986) is currently utilized in traps for gravid Culex quinquefasciatus Say. Miltar et al. (1992) replaced alfalfa hay with Bermuda grass (Cynodon dactylon L.) and isolated a five-component attractant mixture from an infusion that was strongly attractive to ovipositing Cx. quinquefasciatus. A single component of this fraction, 3-methylindole (skatole), was shown to be the most biologically active. Physical factors such as the color of the oviposition water have also been shown to influence oviposition activity in mosquitoes. In early experiments, Bates (1940) and Lund (1942) found that oviposition pans with darkened bottoms were attractive to ovipositing Anopheles labranchinae atroparvus van Thiel and Anopheles quadramaculatus Say, respectively. In laboratory experiments on Aedes triseriatus (Say), Williams (1962) found that dyed oviposition water was more attractive to ovipositing females than undyed water. In the field, Beehler and DeFoliart (1990) collected nearly four times as many Ae. triseriatus eggs in oviposition traps containing dyed tap water as compared to oviposition traps with only tap water. Interactions between chemical and physical oviposition attractants and cues have been documented primarily in container-bred mosquitoes. O'Gower (1963) found that oviposition activity in Aedes aegypti var. queenslandis Theobald is influenced by a number of physical and chemical factors including olfactory and visual cues. Wilton (1968) showed effects of olfactory and visual factors in laboratory experiments on Aedes triseriatus (Say). Also in laboratory studies with Ae. triseriatus, McDaniel et al. (1976) showed that visual factors interacted with factors produced by larval conspecifics. The objectives of our study were to investigate, in the laboratory, the importance of visual cues on the oviposition behavior of Culex quinquefasciatus, a mosquito species whose breeding habitats range from large containers to large polluted temporary ponds, and to determine if visual oviposition cues in the form of dyed waters interact with its chemical oviposition attractants. METHODS AND MATERIALS
The laboratory colony of Cx. quinquefasciatus was established using egg rafts collected from organic infusion-baited oviposition traps placed in Orange County, California in June through October of 1991. The colony was maintained
CHEMICAL AND VISUAL CUES IN MOSQUITO OVIPOSITION
as described by Kramer and Mulla (1979). Mosquitoes were allowed to mate and blood-feed on chicks seven days before laboratory experiments (University of California animal care protocol #A-9105063). The insectary had a photoperiod of 14 : 10 hr light-dark including a 2-hr dusk/dawn period. The day lighting was provided by two 40-W fluorescent tubes. The 1-hr dusk twilight period and the 1-hr dawn period were provided by a single 15-W incandescent bulb. Insectary temperatures were maintained at 25 + 3~ Six experiments were conducted to determine the effect of water dyed with ink and its interaction with a chemical attractant on the ovipositional response of Cx. quinquefasciatus. Experiment 1 was designed to determine the effect of water dyed with ink on oviposition behavior. Twenty gravid laboratory-reared Cx. quinquefasciatus were placed at 1700 hr in each of six cages (23 x 23 x 32 cm) in a laboratory insectary. Each of the six cages contained two randomly placed waxed paper oviposition cups (5 x 7.5 cm diam), one cup containing the test solution, which in this case was 25 #1 of India ink (Fount India, Pelikan, Hannover, Germany) in 80 ml of distilled water and the other a control, distilled water. Mosquitoes were left in the cages overnight, and the following morning the resulting egg rafts were counted and collected. Egg raft numbers were transformed (~x) and a paired t test was used to evaluate the attractancy of the dyed water. The square root transformation with a paired t test is appropriate in situations where the variance is proportional to the mean (Snedecor and Cochran, 1980). This six-cage experiment was repeated over three nights for a total of 18 replicates. Experiment 2 examined the oviposition response of gravid female Cx. quinquefasciatus to dyed oviposition waters under natural twilight conditions. Four screen cages (45 cm on a side) were placed on a window ledge where they were exposed to natural twilight conditions. Twenty gravid Cx. quinquefasciatus were placed in each cage with two waxed paper oviposition cups placed randomly in the rear comers of the cage containing either dyed oviposition water or a distilled water control. Mosquitoes were placed in cages at 1600 hr, approximately 2.5 hr before sunset, to be exposed to full crepuscular photoperiodic changes. Temperatures were maintained at 24 + 2~ The following morning the resulting egg rafts were removed, the number of egg rafts counted, and the data analyzed as in experiment 1. This experiment was repeated over three nights for a total of 12 replicates. Experiment 3 determined if the effect of the dyed oviposition water was due to a visual response to the dye or an olfactory response due to volatile compounds contained in the dye. Four cages were set under natural twilight conditions with 20 gravid female Cx. quinquefasciatus using the protocol of experiment 2. Two oviposition cups were again placed randomly in the rear comers of each cage, one containing 25 #1 of India ink placed in 80 ml of
BEEHLER ET AL.
distilled water and the other containing 25/zl of India ink from which the volatile fraction had been removed in 80 ml of distilled water. The experiment was also repeated over three nights for a total of 12 replicates. Data were analyzed as in experiment 1. To obtain a profile of the India ink volatiles, purge and trap analysis using a Tekmar LSC 2000 purge and trap device (Tekmar Co., Cincinnati, Ohio) was interfaced to a cryocooled (liquid CO2) Hewlett-Packard 5890 gas chromatograph (GC) (Hewlett-Packard, Avondale, Pennsylvania) coupled to a HewlettPackard 5970 mass selective detector. Ink (5 ml) was loaded into a 25-ml fritted sparge sampler and purged for 12 min at ambient temperature with He (20 ml/min). Purged volatiles were collected on a Tekmar Carbopak B adsorbant cartridge. The adsorbent cartridge was then flushed with dry He for 4 min to drive off water and desorbed for 4 min at 225~ directly into the GC injector. The transfer line temperature was 100~ The GC was programmed from 0~ for 2 rain, 10~ to 250~ The injector and detector temperatures were 250~ Analyses were replicated twice with a blank run between each replicate. Volatiles were removed from a 20-ml ink sample by concentrating the ink to dryness on a rotary evaporator under reduced pressure at 40~ followed by pumping the solid residue at 0.02 mm Hg overnight. The resulting black solid was reconstituted with 20 ml of distilled water. The resulting mixture was sonicated until it became homogenous and was then compared with India ink for oviposition attractancy under natural twilight conditions. In experiments 4 and 5, an oviposition choice test protocol similar to that used in the previous insectary experiment (experiment 1) compared experimental waters treated with a chemical oviposition attractant to those treated with a physical attractant, the India ink. A test oviposition cup containing 80 ml of distilled water was treated with 7.6/xg phenol, 0.65 t~g 4-methylphenol, 3.2/zg 4-ethylphenol, 0.4/zg indole, and 2.4/zg 3-methylindole, which was shown by Millar et al. (1992) to be attractive to ovipositing Cx. quinquefasciatus. The other cup, also containing 80 ml of distilled water, was treated with 25 ~1 of India ink. These cups were placed randomly in the rear comers of an oviposition cage containing 20 gravid female Cx. quinquefasciatus for one night. This experiment was repeated over three nights for a total of 18 replicates. Experiment 5 was conducted similarly using paired oviposition cups, each containing 80 ml of distilled water treated with 25 /zl of India ink. One cup of the pair was randomly chosen and treated with the chemical attractant mixture described above. Data were analyzed as in experiment 1. In experiment 6, 2 • 2 factorials were performed under natural photoperiodic conditions to determine if there were interactive effects of the physical factor (India ink) and the chemical mixture on the oviposition behavior of Cx. quinquefasciatus. These experiments were conducted in cages (45 cm per side) using four oviposition cups (80 ml) placed randomly in each of the four comers.
CHEMICAL AND VISUAL CUES IN MOSQUITO OVIPOSITION
The cups contained: (1) distilled water (control), (2) water and dye (25/zl), (3) water and chemical attractant mixture (at the above concentrations), and (4) water and chemical attractant mixture and dye. Thirty gravid female Cx. quinquefasciatus were placed in each cage and allowed to oviposit overnight. This experiment was repeated six nights with four cages per night for a total of 24 replications. Egg rafts were collected in the morning, data were transformed to square roots, and analyzed using multiple regression methods standard for factorial experiments (Box et al., 1978).
RESULTS AND DISCUSSION The results of experiments 1 and 2 are summarized in Table 1. In experiments conducted under insectary lighting, cups that contained distilled water dyed with ink collected means o f 4.3-6.5 egg rafts/cup, while distilled water controls collected means of 3.0-8.0 egg rafts/cup. The mean number of egg rafts collected in the dyed cups was not significantly higher than the control cups (paired t, P > 0.05). In the experiments conducted under natural twilight conditions, the cups that contained distilled water treated with ink collected 14.0-16.5 egg rafts/cup, while the distilled water controls collected means of 2.8-4.8 egg rafts/cup. In all cases the cups treated with ink collected significantly more egg rafts than did the undyed controls (paired t, P < 0.025),
TABLE 1. OVIPOSITIONALRESPONSEOF Culex quinquefasciatus TO DYED OVIPOSIT1ONWATER Mean N egg rafts/cup Night"
6.5 11.5 4.3
5,0 NS 8.7 NS 3,0 NS
Insectary lighting 1
2 3 Natural twilight
aEach experiment consisted of six replicated cages with 20 gravid females per cage repeated over three nights in insectary experiments and four replicated cages also with 20 females per cage in natural twilight conditions. h**p < 0.025, paired t; ***P < 0.01, paired t; NS, not significantly different (P > 0.05, paired t).
BEEHLER ET AL.
outnumbering the controls by more than 3 : 1. Twilight in the insectary experiments was supplied by a single 15-W incandescent bulb, which provided illumination for 1 hr after fluorescent bulbs were switched off. Twilight under natural conditions was significantly longer and the decline in light levels was moJ~ gradual. When Cx. quinquefasciatus females were exposed to natural twilight conditions, dyed water was clearly attractive to females. Dyed oviposition water has been shown to be attractive to ovipositing female mosquitoes in a number of field and laboratory studies. The containerbreeding mosquito, Aedes triseriatus, has been shown to oviposit preferentially in dyed oviposition waters in both field and laboratory studies (Williams, 1962; Wilton, 1968; Beehler and DeFoliart, 1990; Beehler, 1992). Both Anopheles atroparvus and An. quadramaculatus oviposited preferentially in oviposition waters that appeared darker than controls (Bates, 1940; Lund, 1942). In field studies conducted by the Orange County, (California) Vector Control District, Cx. quinquefasciatus oviposited preferentially in traps containing darkened oviposition water (J.P. Webb, personal communication). Our studies provide further evidence that darkened oviposition water attracts ovipositing Cx. quinquefasciatus females. Experiment 3 was designed to study whether the response of gravid Cx. quinquefasciatus to dyed water was due to the increased optical density of dyed oviposition water or volatile components of the dye itself. Purge and trap analysis revealed that the India ink used in these experiments contained volatile fractions, including ethanol, isopropanol, hexane, ethyl acetate, dimethoxymethane (methylal), and traces of naphthalene and methyl naphthalene. Headspace chromatography of 29 inks has shown that alcohols are among the most common solvents in ink samples (Rastogi, 1991). Concentration and pumping under vacuum removed all of these volatiles from the India ink sample. The evaporated and reconstituted ink had a purge and trap volatiles profile equivalent to that of the distilled water with which the ink had been reconstituted. Paired tests comparing the reconstituted ink that had the volatiles removed with the unpurged India ink showed no difference in the number of egg rafts collected in the paired cups (paired t, P > 0.05). Cups containing distilled water and ink with the volatile components removed collected means of 7.7-13.7 egg rafts/cup and those that contained distilled water and unreconstituted ink collected means of 9.2-12.7 egg rafts/cup (Table 2). These data strongly suggest that the ovipositional responses of the gravid female Cx. quinquefasciatus were due to the change in optical density of the experimental waters produced by the ink, not by the volatiles present in the ink itself. Experiment 4 was conducted under insectary conditions to further determine if the ovipositional response of Cx. quinquefasciatus was a response to the optical density of dyed water or to chemical components found in the ink. Experiments 1 and 2 showed that the environment provided by the insectary
CHEMICAL AND VISUAL CUES IN MOSQUITO OVIPOSITION
TABLE 2. INFLUENCE OF INDIA INK CONTAINING VOLATILE SUBSTANCES AND INDIA INK WITHOUT VOLATILES ON OVIPOSITION BY
Mean N egg rafts/cup No
9.2 12.7 12.5
13.7 NS 7.7 NS 17.5 NS
aThe experiment consisted of six replicated cages with 20 gravid females per cage repeated over three nights. Experiments were conducted in natural twilight. ~ not significantly different (paired t, P > 0.05).
cages is only marginally suitable for the use o f visual cues in the selection of an oviposition site. Dyed water was not shown to be an oviposition attractant in the insectary, but in cages subjected to natural twilight the ink proved to be attractive to Cx. quinquefasciatus females. Therefore, insectary experiments were conducted in conditions in which visual cues were minimized and chemical cues were maximized. In experiment 4, water treated with the oviposition attractant mixture at the concentration used in the earlier experiments collected means of 6.3-10.3 egg rafts compared to 3.0-5.3 egg rafts collected in the dyed water (Table 3). The mean values in the attractant-treated cups were significantly higher (paired t, P < 0.01). In experiment 5, waters that contained India ink alone collected significantly fewer (paired t, P < 0.01) rafts than did oviposition cups containing ink and attractant mixture (Table 3). The 2 • 2 factorial experiment is summarized in Figure 1. Control cups containing only distilled water collected a mean of 0.5 egg rafts/cage; cups containing dyed water collected a mean of 3.8 egg rafts/cage; cups containing water plus chemical attractant collected 3.0 egg rafts/cage; and cups containing both attractant and dye collected 19.8 egg rafts/cup. Regression analysis of the 48 replicates over 12 nights shows that both the ink (F = 4.31, P < 0.001) and the chemical mixture (F = 4.50, P < 0.001) were significantly more attractive to ovipositing Cx. quinquefasciatus than were the distilled water control oviposition cups. Cups containing a combination of both ink and attractant mixture were significantly more attractive (F = 3.70, P < 0.001) than either of the test substances presented singly or the untreated control. The R 2 for the regression was 0.95. The ovipositional response to the ink presented in tandem with the chemical attractant mixture appeared to be synergistic. Synergism has been defined as a response that is more than the mean additive effects of the compounds presented
BEEHLER ET AL.
TABLE 3. LABORATORYOVIPOSITIONCHOICE EXPERIMENTSWITH Culex quinquefasciatus USING DYED OVIPOSITIONWATER AND A CHEMICALATTRACTANT BLEND. 1 Mean N egg rafts/cup Nighta
10.3 6.3 Dye + Attractants
5.3*** 3.0*** Dye alone
~ 5-component mixture (Millar et al. 1992) bEach experiment consisted of six replicated cages with 20 gravid females per cage. Experiments were conducted under insectary conditions. ***P < 0.01, paired-t.
Control "l Chemicals -
10 20 Mean no. of egg rafts/cup
FIG. 1. Synergism between a physical factor (dyed oviposition water) and a chemical factor (five-component oviposition attractant mixture) in a replicated 2 • 2 factorial experiment. Bars followed by different letters are significantly different (P < 0.05).
in t a n d e m ( W a d l e y , 1967). T h e effect o f ink and the c h e m i c a l attractant presented together was tested using the f o l l o w i n g contrast: (P'ink) -[- (//'chemical) ~--- (]~ink+chemicat) The m e a n response to the combination o f ink and chemical mixture (/zin k + chemical)
CHEMICALAND VISUALCUES IN MOSQUITOOVIPOSITION
presented together was significantly greater than the s u m o f the m e a n effects o f the single c o m p o n e n t s [(/~i,k) and (/Xchemical)] (P < 0.01), indicating that the chemical and visual attractants acted synergistically. In conclusion, these laboratory studies show that gravid female Cx. quinq u e f a s c i a t u s mosquitoes use both visual and chemical cues in the selection of an oviposition site. The experiments using India ink show that the mosquitoes readily recognize and oviposit preferentially in natural twilight, but not in the insectary, in waters which have increased optical density. In field conditions, darkened waters are usually the most organically enriched and provide a highquality nutritional substrate for the larvae. The chemical c o m p o n e n t s o f the attractant mixture used are also indicative o f a larval habitat that is nutritionally enriched. The chemical and physical cues appear to act synergistically in influe n c i n g the ovipositional site selection behavior o f gravid female mosquitoes in a way that optimizes the quality o f the larval habitat. Acknowledgments--The authors thank John Chaney and Lori Reuter for their technical assistance. Fred Chun maintained the mosquito colony. This research was supported in part by the University of California Special Funds for Mosquito Research, the Orange County Vector Control District, and the South East Mosquito Abatement District.
REFERENCES BATES,M. 1940. Oviposition experiments with anopheline mosquitoes. Am. J. Trop. Med~ 20:569583. BEEHLER, J.W. 1992. The oviposition behavior of Aedes triseriatus. Proc. Calif. Mosq. Vector Control Assoc. 59:108-113. BEEHLER,J.W., and DEFOLIART,G.R. 1990. A field evaluation of two suggested Aedes triseriatus oviposition attractants. J. Am. Mosq. Control Assoc. 6:720-722. BENTLEY, M.D., and DAY, J.F. 1989. Chemical ecology and behavioral aspects of mosquito oviposition. Annu. Rev. Entomol. 34:401-421. Box, G.E.P., HUNTER,W.G., and HUNTER,J.S. 1978. Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building. John Wiley & Sons, New York. KRAMER, W.L., and MULLA, M.S. 1979. Oviposition attractants and repellents of mosquitoes: Oviposition responses of Culex mosquitoes to organic infusions. Environ. Entomol. 8:11111117. LUND,H.O. 1942. Studies on the choice of a medium for oviposition by Anopheles quadrimaculatus. J. Nat. Malaria Soc. 1:101-111. MCDANIEL,I.N., BENTLEY,M.D., LEE, H.P., and YATAGAI,M. 1976. Effects of color and larvalproduced oviposition attractants on oviposition ofAedes triseriatus. Environ. Entomol. 5:553556. MILLAR, J.G., CHANEY,J.D., and MULLA,M.S. 1992. Identification of oviposition attractants for Culex quinquefasciatus from fermented Bermuda grass infusions. J. Am. Mosq. Control Assoc. 8:11-17. O'GOWER, A.K. 1963 Environmental stimuli and the oviposition behaviour of Aedes aegypti var. queenslandis Theobald (Diptera, Culicidae). Anita. Behav. 11:189-197. RASTO~I, S.C. 1991. Analysis of organic solvents in printing inks by headspace gas chromatography-mass spectrometry. J. High Resol. Chromatogr. 14:587-589.
BEEHLER ET AL.
REITER, P. 1986. A standardized procedure for the quantitative surveillance of certain Culex mosquitoes by egg raft collection. Mosq. News 2:219-221. SNEDECO~, G.W., and COCI-IRAN,W.G. 1980. Statistical Methods. Iowa State University Press, Ames, iowa. WADLEV,F.M. 1967. Experimental Statistics in Entomology. Graduate School Press, USDA, Washington, D.C. WILLIAMS, R.E. 1962. The effect of coloring oviposition media with regard to the mosquito Aeries triseriatus (Say) J. Parasitol. 48:919-925. W1LTON~ D.P. 1968. Oviposition site selection by the tree-hole mosquito, Aedes triseriatus (Say). J. Med. Entomol. 5:189-194.