Comparison of the attractiveness of organic infusions to the standard CDC gravid mosquito trap.
Authors: McPhatter, Lee P.
Olsen, Cara H.
Debboun, Mustapha
Pub Date: 07/01/2009
Publication: Name: U.S. Army Medical Department Journal Publisher: U.S. Army Medical Department Center & School Audience: Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 U.S. Army Medical Department Center & School ISSN: 1524-0436
Issue: Date: July-Sept, 2009
Accession Number: 242963594
Full Text: Introduction

Mosquito-borne pathogens are generally acquired from vertebrate hosts and transmitted by blood-feeding female mosquitoes. Since pathogen acquisition and subsequent disease transmission require at least 2 blood meals and one oviposition cycle to be completed, screening gravid female mosquitoes for virus infection provides the best estimate of the infected population. (1) Multiple techniques and methods are used to conduct mosquito surveillance, such as incandescent light traps, ultraviolet light traps, electric nets, C[O.sub.2] traps, counterflow traps, human bait counts, oviposition traps, and Centers for Disease Control and Prevention (CDC) gravid traps. Of all the mosquito surveillance traps and methods, the CDC gravid trap is the only one that specifically targets and attracts gravid females. (2) Therefore, the likelihood of arbovirus detection significantly increases when using this trap. According to some surveillance studies, collections from gravid traps were estimated to contain from 57%3 to 95% (4) gravid females. The CDC gravid trap, shown in Figure 1, consists of a plastic rectangular basin that holds an infusion of organic matter that is attractive to some species of ovipositing mosquitoes. A vertical hollow cylinder with a battery-powered fan is mounted approximately 2.5 cm over the infusion. The action of the fan creates an upward, unidirectional air flow that vacuums the mosquitoes to the top of the tube and into a collection net.


Both laboratory and field studies have shown that gravid mosquitoes exhibit significant preferences when selecting aquatic sites for oviposition. (5,6) The physical characteristics such as color, substrate texture, odorants, and chemical cues were some of the factors identified to influence oviposition. (7) Additional studies have indicated that gravid females of different species used different factors for acquisition and selection of oviposition sites. For example, Aedes albopictus (Skuse) females exhibited a strong preference to oviposit in artificial containers such as tires. Organic matter dissolved in water has been found to attract different mosquito species. (8) Aedes aegypti (L) females were attracted to horse manure infusions (8) and Aedes triseriatus (Say) females to tree-hole water. (9) Infusions made with materials such as hay, (10) grass, (11) alfalfa pellets, bulrush, (12) manure, (8) and oak leaves were reported as attractive substrates for Culex species. The purpose of this study was to compare the relative attractiveness of several organic infusions to different mosquito species in southeastern Texas.


Materials and Methods

Study Site

The field study was conducted at six locations on Fort Sam Houston military reservation (29.42[degrees]N 98.49[degrees]W, 195 m elevation) (Figure 2). Fort Sam Houston is a US Army base located in the northeast of San Antonio, Texas. The installation comprises approximately 3,000 acres and has a population of 35,000 people. Mosquito activity was prevalent throughout the installation from July to October 2008.

Trapping Techniques

A total of 6 mosquito gravid trapping sites were selected for this study (Figure 2). Trap site one was placed on the west side of the installation near the military horse stables adjacent to an active creek that is surrounded by urban terrain (residential). Trap two was placed in the northeast side of the installation near a low lying drainage channel that divides two urban communities. Trap three was placed near a large creek located on the edge of a recreational picnic area. Trap four was placed next to a house in a populated neighborhood, which had no immediate proximity to water. Trap five was placed at the edge of the wood line near an active athletic complex. Trap six was placed next to a tree near the golf course. The location of each trap was not less than 900 m from other traps. The traps used for this study were CDC gravid traps (model 1712, John W. Hock Company, Gainesville, FL). The tray of each gravid trap was filled 1/4 full with a different infusion bait (rabbit chow, Bermuda grass, live oak leaf Quercus virginiana (Mill), acacia leaf Acacia schaffneri (Wats), and filamentous green algae). All infusions except the green algae were prepared using 227 g of substrate added to 4 L of tap water. Each mixture was placed in closed Rubbermaid (Newell Rubbermaid Inc, Atlanta, GA) plastic containers and left outside in the sun to incubate for 5 days prior to each trap week. The algae was collected from fresh water ponds and was incubated at outdoor temperatures (27[degrees]C--29[degrees]C) for 7 days prior to each trapping week. The green algae infusion consisted of 133 g of wet algae mixed with 4 L of pond water. Trapping was conducted 3 days per week from September 3 to November 22, 2008. The trapping cycle was divided into 2 experiments. Experiment one was conducted at locations one, two, and three during the first 6 weeks and experiment two was conducted at locations four, five, and six during the last 6 weeks. Trapping sessions commenced at 8 am on Wednesday and ended 8 am on Saturday. After the completion of each trap week, each infusion was rotated to a different trap site to ensure each infusion was tested once at each trap location. All of the mosquitoes collected were counted and identified to species using the morphological keys by Darsie and Ward. (13)

Statistical Analysis

All data were analyzed using Stata release 10 for Windows (StataCorp LP, College Station, TX). Collection data was converted using the logarithmic scale. Counts of zero were replaced with a value of 0.5 before taking the natural log. Histograms and normal quantile plots of the log counts were examined and no substantial deviations from normality were observed. Log-transformed counts of female mosquitoes collected by gravid traps were analyzed by 2-way analysis of variance (ANOVA) for variation among the most significant infusions, after adjusting for species. Multiple comparisons among infusions were made using Tukey HSD (honest significant difference) test at a 95% confidence level. A 3-factor ANOVA was also conducted to determine if the different trap sites affected the species of mosquitoes and the different infusions.


A total of 2,003 females of 14 species of mosquitoes in 3 genera (Aedes, Culex, Orthopodomyia) were collected. The Table shows the mosquitoes that were attracted to the 5 different infusion types. The most common mosquitoes collected were Culex quinquefasciatus Say (35%), Cx erraticus Dyar and Knab (34%), Cx nigripalpus Theobald (20%), Cx interrogator Dyar and Knab (6%), and Aedes albopictus (3%). The average numbers of Ae albopictus and Cx interrogator collected were significantly lower than those for the other species. Most of the mosquitoes collected (51%) were from traps containing the Bermuda grass infusion. Based on Tukey's multiple comparison procedure, traps baited with Bermuda grass (F = 23.57, df=3, P<0.0010) had a significantly higher number of mosquitoes than the other infusion types tested. Following Bermuda grass (51%), oak leaves (18%), algae (15%), and acacia leaves (12%) attracted the bulk of the remaining mosquito population. However, these 3 infusion types were not significantly different from each other. Of the 3 genera represented by the collected mosquitoes, 96% were Culex. Fifty-three percent of the collected Culex species were attracted to traps containing Bermuda grass infusion. Rabbit chow infusion collected the least number of Culex species (4%). The genus Aedes represented 4% of the total mosquitoes collected. Most of the Aedes (39%) were collected by acacia leaf baited infusions. Traps baited with rabbit chow infusions attracted the least number of Aedes (3%).

The total number of mosquitoes collected for each infusion by experiments one and two is presented in Figure 3. Experiment two had significantly lower numbers compared to experiment one. Bermuda grass infusion performance was still superior over all other infusions during this period. Averaging across species and sites, there was a significant main effect of infusion in both experiment one (P = 0.0017) and experiment two (P < 0.0001). In both experiments, the Bermuda grass infusions attracted the highest number of mosquitoes. In experiment one, the algae and oak infusions attracted nearly as many mosquitoes as the Bermuda grass infusion, and the differences among these infusions were not statistically significant. The Bermuda grass infusion attracted significantly more mosquitoes than the acacia infusion in experiment one (P = 0.0025). In experiment two, significantly more mosquitoes were collected using the Bermuda grass infusion than any of the other infusions (P <0.001 for all comparisons). There was a significant site to infusion interaction in both experiment one (P = 0.0070) and experiment two (P = 0.0074), indicating that the differences among infusions varied by site. Overall, in all sites, except site two (experiment one), the Bermuda grass infusion attracted the highest number of mosquitoes. After adjusting for multiple comparisons, the differences between Bermuda grass and the other infusions were not statistically significant in sites one, two and three, but were significant (P < 0.05) in sites four, five, and six (experiment two). Significant species to infusion interaction was observed in experiment two (P = 0.0024) but not in experiment one (P = 0.0712). This finding suggests that the relative effectiveness of the 4 (Bermuda grass, oak leaves, acacia leaves, and algae) infusions was similar for all species in experiment one. In experiment two, after adjusting for multiple comparisons, there was no significant difference among infusion with respect to the number of Ae albopictus mosquitoes attracted. The Bermuda grass infusion attracted significantly more of both Cx erraticus and Cx nigripalpus than any other infusion, and attracted significantly more Cx quinquefasciatus than any infusion except acacia.



Our data supports reports from several other studies that Bermuda grass infusion is highly attractive to Cx quinquefasciatus 2,7,14,15 The Bermuda grass infusion baited trap collected 2 to 3 times more Cx quinquefasciatus than all of the other infusions. Over half of the Culex species collected in this study were in the traps that contained Bermuda grass infusion. The Bermuda grass infusion also attracted more Cx nigripalpus and Cx erraticus mosquitoes than the other infusions. In fact, the numbers of Cx erraticus collected with the Bermuda grass infusion were higher than the numbers of Cx quinquefasciatus. Burkett-Cadena and Mullen (16) reported, in contrast to our results, that small numbers of Cx erraticus were collected with gravid trap, and inferred that the gravid trap was not a useful tool in collecting females of Cx erraticus. In their study, they used oak leaves, pine straw, red (dyed) hardwood mulch, and composted manure to make 4 different infusions. Similar to their results, the traps baited with oak leaf infusion used in our study collected modest numbers of Cx erraticus (Figure 3). Therefore, it is likely that the low numbers of Cx erraticus collected in their study were impacted by the selection of infusions. The infusions selected for this study collected low numbers of Cx coronator Bayer, Cx interrogator, Cx salinarius Coq. and Cx tarsalis Coquillett. With the exception of Cx interrogator, they all were commonly collected (in the vicinity of the trap sites) in moderate numbers using the CDC standard light trap. It is possible that the age of the infusions impacted the collection of these common species. Isoe et al (17) showed that Cx tarsalis and Cx quinquefasciatus attraction to Bermuda grass infusion was affected by age. They suggested that Cx tarsalis preferred relatively clean water for oviposition sites, whereas Cx quinquefasciatus preferred more polluted and turbid waters. The oak, algae, and acacia infusions used in this study were the least turbid. Culex coronator, Cx salinarius and Cx tarsalis were only attracted to these 3 infusions. After the Bermuda grass infusion, the oak leaf infusion attracted the second highest number of mosquitoes. In addition, the oak leaf infusion attracted 11 different mosquito species, while the Bermuda grass infusion attracted 7 species.

Aedes albopictus was the only non-Culex mosquito species collected by all 5 infusions used in this study. Overall, the numbers of Ae albopictus were low. However, their presence in the traps was common when compared with other Aedes species collected during this study. Excluding the algae infusion, the acacia infusion attracted 2 to 3 times more Ae albopictus than the other 3 infusions. However, the mean number was low. This lack of preference for a certain type of infusion is most likely due to their oviposition behavior. Mogi and Mokry (18) referred to this behavior as skip oviposition. This occurs when female mosquitoes lay their eggs in several containers as opposed to laying their entire clutch in one container. (19-21)

The rabbit chow infusion yielded the lowest results from all the infusions. Similar reports of low responses to this infusion were reported by Isoe and Millar, (15) Allan et al, (14) and Jackson et al. (22) In contrast to the other infusions, the rabbit chow infusion attracted significant numbers of flies (Calliphoridae). It is possible that the attraction of these flies was antagonistic to the trapping efficiency. The high levels of calliphorid flies around the traps created an intense situation in which the flies were fighting over the oviposition media. The trapping location that had the most significant issues with the flies was the one at the horse stables. Ironically, the rabbit chow attracted the least amount of mosquitoes at this location.

In conclusion, Bermuda grass, acacia leaves, oak leaves, and algae infusions were effective as gravid trap baits for the attraction of Cx quinquefasciatus, Cx nigripalpus, and Cx erraticus. Additional studies focusing on the concentration of organic material and age of these infusions might show how to increase attraction rates of low density gravid trap species such as Ae albopictus and Cx tarsalis.


We thank Prabir Chakraborty and Annette Mitchell for assistance in data analysis; Samuel Melendez, Irene Brown, Jacob Caraway, Angela Moore, and Diego Cano for field assistance; and Dan Strickman, Dan Kline, and Sandy Allan for their constructive comments and review of this manuscript.


(1.) Braks MA, Carde' RT. Improving efficacy of box gravid traps for collecting Culex quinquefasciatus. J Vector Ecol. 2006;32(1):83-89.

(2.) Burkett-Cadena ND, Mullen GR. Field comparison of Bermuda-hay infusions of emergent aquatic vegetation for collecting female mosquitoes. J Am Mosq Control Assoc. 2007;23(2):117-123.

(3.) Ritchie SA. Hay infusion and isopropyl alcohol-baited CDC light trap: a simple effective trap for gravid Culex mosquitoes. Mosq News. 1984;44:404-407.

(4.) Reiter P. A revised version of CDC gravid mosquito trap. J Am Mosq Control Assoc. 1987;3:325-327.

(5.) Weber RG, Horner TA. The ability of Culex pipiens and Culex restuans to locate small ovisites in the field. Proc Annu Meet NJ Mosq Control Assoc 1992. 1993;79:96-103.

(6.) Lampman RL, Novak RJ. Oviposition preferences of Culex pipiens and Culex restuans for infusion-baited traps. J Am Mosq Control Assoc. 1996;12:437-455.

(7.) Bentley MD, Day JF. Chemical ecology and behavioral aspects of mosquito oviposition. Annu Rev Entomol. 1989;34:401-421.

(8.) O'Gower AK. Environmental stimuli and the oviposition behavior of Aedes aegypti var. queenslandis. Anim Behav. 1963;11:189-197.

(9.) Bentley MD, McDaniel IN, Lee YP, Stiehl B, Yatagai M. Studies of Aedes triseriatus oviposition attractants produced by larvae of Aedes triseriatus and Aedes atropalpus (Diptera: Culicidae). J Med Entomol. 1976;13:112-115.

(10.) Reiter P. A portable battery-operated trap for collecting gravid Culex mosquitoes. Mosq News. 1983;43:496-498.

(11.) Gjullin CM, Johnson JO, Plapp FW Jr. The effect of odors released by various waters on the oviposition sites selected by two species of Culex. Mosq News. 1965;25:268-271.

(12.) Du Y, Millar JG. Eletroantenogram and oviposition bioassay responses of Culex quinquefasciatus and Culex tarsalis (Diptera: Culicidae) to chemicals in odors from Bermuda grass infusions. J Med Entomol. 1999;36:158-166.

(13.) Darsie RF, Ward RA. Identification and Geographical Distribution of the Mosquitoes of North America, North of Mexico. Fresno, CA: American Mosquito Control Association; 1981.

(14.) Allan SA, Bernier UR, Kline DL. Evaluation of oviposition substrates and organic infusions on collecting of Culex in Florida. J Am Mosq Control Assoc. 2005;21(3):268-273.

(15.) Isoe J, Millar JG. Characterization of factors mediating oviposition site choice by Culex tarsalis. J Am Mosq Control Assoc. 1995;11:21-28.

(16.) Burkett-Cadena ND, Mullen GR. Comparison of infusions of commercially available garden products for collection of container-breeding mosquitoes. J Am Mosq Control Assoc. 2008;24(2):236-243.

(17.) Isoe J, Beehler JW, Millar JG, Mulla MS. Oviposition responses of Culex tarsalis and Culex quinquefasciatus to aged Bermuda grass infusions. J Am Mosq Control Assoc. 1995;11:39-44.

(18.) Mogi M, Mokry J. Distribution of Wyeomyia smithii (Diptera: Culicidae) eggs in pitcher plants in Newfoundland, Canada. Trop Med. 1980;22: 1-12.

(19.) Fay RW, Perry AS. Laboratory studies of the ovipositional preferences of Aedes aegypti. Mosq News. 1965;26:531-535.

(20.) Rozeboom LE, Rosen L, Ikeda J. Observations on oviposition by

Aedes (S.) albopictus Skuse and A. (S.) polynesienses marks in nature. J Med Entomol. 1973;4:397-399.

(21.) Chadee DD, Corbet PS, Greenwood JJD. Egg laying yellow fever mosquitoes avoid sites containing eggs laid by themselves or by conspecifics. Entomol Exp Appl. 1990;57:295-298.

(22.) Jackson BT, Paulson SL, Youngman RR, Scheffel SL, Hawkins B. Oviposition preferences of Culex restuans and Culex pipens (Diptera: Culicidae) for selected infusions in oviposition traps and gravid traps. J Am Mosq Control Assoc. 2005;21:360-365.

CPT Lee P. McPhatter, MS, USA

Cara H. Olsen, DrPH

COL Mustapha Debboun, MS, USA

CPT McPhatter is Chief, Entomology Section, Department of Preventive Medicine, Brooke Army Medical Center, Fort Sam Houston, Texas.

Dr Olsen is an Assistant Professor and Biostatistics Consultant in the Department of Preventive Medicine and Biometrics, Uniformed Services University of the Health Sciences, Bethesda, Maryland.

COL Debboun is Director, Center for Health Education and Training, Academy of Health Sciences, US Army Medical Department Center & School, Fort Sam Houston, Texas.
Distribution by species of female mosquitoes collected by
5 different infusion types during 2 trapping occasions in
2008 (Sep 3 to Oct 11, Oct 15 to Nov 22) at Fort Sam
Houston Military Reservation, Texas.

Mosquito Species                      Infusion Type

                         Bermuda    Rabbit Chow    Acacia   Oak

Aedes aegypti               0            0            0       1
Aedes albopictus            6            2           26       8
Aedes triseriatus           0            0            0       2
Aedes vexans                3            0            2       2
Aedeszoosophus              1            0            0       0
Aedes atropalpus            0            0            0       1
Culex coronator             0            0            0       0
Culex erraticus            450          26           60      79
Culex interrogator          38           3           19      42
Culex nigripalpus          198          24           43      63
Culex quinquefasciatus     336          30           84     154
Culex salinarius             0           0            4       0
Culex tarsalis               0           0            0       2
Orthopodmyia alba            0           0            0       0
Total                     1,032         85          238     354

Mosquito Species         Infusion Type


Aedes aegypti                  1
Aedes albopictus              17
Aedes triseriatus              0
Aedes vexans                   0
Aedeszoosophus                 0
Aedes atropalpus               0
Culex coronator                1
Culex erraticus               64
Culex interrogator            24
Culex nigripalpus             76
Culex quinquefasciatus       107
Culex salinarius               3
Culex tarsalis                 0
Orthopodmyia alba              1
Total                        294
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.