PREVALENCE OF WEST NILE VIRUS IN TREE CANOPY-INHABITING CULEX PIPIENS AND ASSOCIATED MOSQUITOES JOHN F. ANDERSON, THEODORE G. ANDREADIS, ANDY J. MAIN, AND DANIEL L. KLINE Department of Entomology and Department of Soil and Water, The Connecticut Agricultural Experiment Station, New Haven, Connecticut; American University in Cairo, Cairo, Egypt; U.S. Department of Agriculture, Agriculture Research Service, Center for Medical, Agricultural and Veterinary Entomology, Gainesville, Florida Abstract. Culex pipiens was the dominant mosquito captured in a West Nile virus (WNV) focus in Stratford, Connecticut. More Cx. pipiens were captured in Centers for Disease Control miniature light traps baited with CO 2, quail/hamster traps, and mosquito magnet experimental (MMX) traps placed in the tree canopy than in similar traps placed near the ground. Significantly more Cx. pipiens were captured in MMX traps placed in the canopy than in the other traps tested. Ninety-two percent and 85% of the 206 and 68 WNV isolations were from Cx. pipiens in 2002 and 2003, respectively; 5% and 12% were from Cx. salinarius. Eighty-five percent and 87% of the isolates were from mosquitoes captured in the canopy in each of the two years. The significantly larger numbers of WNV isolates from Cx. pipiens captured in the canopy are attributed to the significantly larger numbers of Cx. pipiens captured in the canopy in comparison to those captured in traps near the ground. INTRODUCTION West Nile virus (WNV) was initially isolated in the New World from mosquitoes and birds in the greater New York City area in 1999. 1,2 Subsequently, the virus spread and was detected in 44 states and the District of Columbia in 2002. 3 A total of 3,389 human cases were reported. Species of Culex are considered to be the most important vectors, though WNV has been isolated or detected from >20 species of mos- quitoes in the United States. 1,2,4–9 Surveillance of arboviruses in mosquitoes is most fre- quently conducted with dry ice-baited Centers for Disease Control (CDC) miniature light traps placed relatively close to the ground, 10,11 although animal baited traps also have been used. 12 However, numerous species of mosquitoes are recog- nized to preferentially inhabit tree canopies and/or to fly at tree canopy height, 13–15 including Culex pipiens. 16–19 Vertical stratification may be influenced by humidity, temperature, light, 13 and possibly by availability of hosts. 20 Culex pipiens,a species that preferentially feeds on birds, 21–24 is an important and competent vector of WNV in both the Old and New Worlds. 25–29 The likely importance of this species in the natu- ral history of WNV in the northeastern United States prompted us to evaluate the prevalence of WNV-infected Cx. pipiens and associated species at ground and tree canopy lev- els using three different types of mosquito traps in a known focus for WNV in Connecticut. MATERIALS AND METHODS Experiments were conducted on Water Pollution Control Authority land of the Town of Stratford, Connecticut. This site (41°10'41N, 73°07'34W) is located adjacent to the Hou- satonic River where it flows into Long Island Sound and was a focal area for WNV in 2001. 6 Trapping commenced on July 8, 2002 and continued until October 17, 2002; in 2003, collec- tions were made from May 20 through November 8. Three different types of traps placed at two different heights were evaluated in 2002; two traps were tested in 2003. Traps were replicated three times each night and were placed in a ran- domized design. A trap was placed near the base of a tree ∼1.5 meters above the ground (ground level), and another trap of the same design was placed in the tree canopy ∼7.6 meters above the ground. Tree height was ∼10.7 meters. The three types of traps tested were a CDC trap (Model 512 with an aluminum dome; John W. Hock Co., Gainesville, FL), 10 a mosquito magnet experimental (MMX) trap, (American Biophysics Corp., East Greenwich, RI), 30 and live quail or hamster traps. 12 The CDC trap uses a motor-driven rotary fan to move mosquitoes attracted by a small light and CO 2 from dry ice stored in a container above the trap to a holding net suspended beneath the trap. The MMX trap is constructed of an ∼11.4-liter clear polyvinyl chloride pretzel container with a fan blowing CO 2 out the bottom and another fan providing airflow into the bottom of the trap. Carbon dioxide was supplied from a 20-lb compressed gas cylinder with a flow rate of 500 mL/min. 30 The live quail or hamster trap was a lard can or a modification (hamsters were used in place of quail on two nights). 12 The modification was a can measuring 63.5 cm long with a diameter of 34.3 cm with a screen cone leading inward into the can from each end. The quail or hamster was placed in a side door in which the bait animal was protected with a screen from feeding mosquitoes. The trap was hung in a horizontal position from the tree. The entire trap was washed using soap to remove odors whenever a different type of animal was being used. The use of quail and hamsters conformed to the guidelines approved by The Connecticut Agricultural Experiment Station’s Animal Care and Use Committee. Only the MMX and CDC traps were evaluated in 2003. Traps were operated overnight and retrieved the following morning. Captured mosquitoes were knocked down with dry ice in the field, quickly aspirated and transferred into a flat- bottomed shell vial measuring 17 × 55 mm. The vial was sealed with a rubber stopper and the juncture of the vial and stopper was wrapped with three layers of 1.9-cm wide water- proof tape. The vial was labeled and stored on dry ice until taken to the laboratory where the vial was transferred to a -80°C freezer. Mosquitoes were identified using the key of Darsie and Ward. 31 Specimens were placed on a cold table and identified with the aid of a dissecting microscope. Female mosquitoes were grouped according to species, date, type of trap, height, and location. Numbers of mosquitoes per pool ranged from 1 to 50. Mosquitoes were kept on regular ice until processed for viruses. For attempted isolation of viruses, mosquitoes were tritu- Am. J. Trop. Med. Hyg., 71(1), 2004, pp. 112–119 Copyright © 2004 by The American Society of Tropical Medicine and Hygiene 112