Control of Larval Aedes aegypti (Diptera: Culicidae) by Cyclopoid Copepods in Peridomestic Breeding Containers GERALD G. MARTEN,l GERARDO BORJAS,2 MARY CUSH,l EDUARDO FERNANDEZ, AND JANET W. REID3 Division de Enfermedades de Transmision Vectorial, Ministerio de Salud Publica de Honduras, Tegucigalpa, Honduras J. Med. Entomol. 31(1): 36-44 (1994) ABSTRACT Mesocyclops longisetus (Thiebaud), Mesocyclops thermocyclopoides Harada, Mesocyclops venezolanus Dussart, and Macrocyclops albidus (J urine) were tested for their effectiveness in controlling Aedes aegypti (L.) larvae in a variety of containers around homes in EI Progreso, Honduras. All four cyclopoid species killed >20 larvae per cyclopoid per d under container conditions. M. longisetus was most effective, not only because it was the most voracious predator, but also because it survived best in the containers. M. longisetus maintained long-term populations in 200-liter drums, tires, vases, and cement tanks (without drains), providing the cyclopoids were not dried or poured out. M. longisetus reduced third- and fourth-instar Ae. aegypti larvae by >980/0 compared with control containers without cyclopoids. M. longisetus should be of practical value for community-based Ae. aegypti control if appropriate attention is directed to maintaining it in containers after introduction. KEY WORDS Copepoda, Aedes aegypti, biological control THE INTEGRATED DENGUE CONTROL PROJECT second-instar mosquitoes, can maintain virtually in El Progreso, Honduras, a cityof:::::::80,000 inhabi­ 1000/0 control of container-breeding Aedes for as tants, is concerned with community-based Aedes long as the cyclopoids survive in the container aegypti (L.) control (Fernandez et al. 1992). The (Riviere & Thirel 1981, Marten 1984, Suarez et project uses mechanical methods of source re­ al. 1984). Mesocyclops longisetus (Thiebaud) duction, such as cleaning water storage contain­ and Macrocyclops albidus (Jurine) currently ers to interrupt larval development, storing tires are used to control Aedes larvae in tires (Mar­ and domestic containers so they do not collect ten 1990a,b,c), and Mesocyclops aspericornis rainwater, and eliminating tires or containers (Daday) has proved effective against Aedes lar­ that are not needed. vae in field trials with tires, water-storage drums, Some breeding sites are not amenable to me­ and cisterns (Riviere & Thirel 1981, Suarez et al. chanical methods. A sporadic water supply may 1984, Riviere et al. 1987a). prevent the cleaning of water storage containers Cyclopoids are more effective for Aedes con­ frequently enough to control Ae. aegypti larvae. trol than other aquatic invertebrates that prey on Because of work outside the home, some house­ mosquito larvae because their high reproductive wives m<lY not have the time to clean their con­ capacity and broad diet (including phytoplank­ tainers frequently. Water storage tanks without a ton, protozoa, and small animals) enable them to drain rarely may be cleaned, and some people maintain abundant populations independent of have tires they are not willing or able to store out mosquito larval abundance. Their small size and of the rain. All of these require an alternate high reproductive capacity also make cyclopoids method to eliminate Ae. aegypti production. inexpensive and convenient for large-scale pro­ Cyclopoid copepods are a promising method duction and distribution (Riviere et al. 1987a, of biological control. The larger species of Marten 1990c, Suarez et al. 1992). these microcrustaceans, which prey on first- and We conducted laboratory experiments and field trials to evaluate the effectiveness of cyclo­ poids for the long-term control of Ae. aegypti 1 New Orleans Mosquito Control Board, 6601 Lakeshore Drive, New Orleans, LA 70126, and Department of Tropical larvae in domestic containers in El Progreso. Medicine, Tulane University, New Orleans, LA 70112. The purpose of the field trials was to determine 2 Departamento de Biologia, Universidad Aut6noma de which containers can be controlled by using cy­ Honduras, Tegucigalpa, Honduras. clopoids and which species of cyclopoids are 3 Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC most effective. Four species common in Hondu­ 20560. ras were tested: Macrocyclops albidus, Mesocy­ 0022-2585/94/36-44$02.00/0 © 1994 Entomological Society of An1erica January 1994 MARTEN ET AL.: CYCLOPOID COPEPODS FOR Ae. Aegypti CONTROL 37 Table I. Percentage of Ae. aegypti production from in their yard for eventual use or sale. Most of domestic containers in EI Progreso, Honduras these tires contain leaves because they are shaded by trees, and they seldom dry out, partic­ Container 0/0 Production ularly during the rainy season. Laundry tanks 44 Animal drinking containers vary in size and Tires 29 200-liter drums 12 shape. They include plastic bowls and tires cut Animal drinking containers 7 in half lengthwise, but mosquito production is Buckets and potsa 3 greatest from shallow cement pools that are em­ Cisterns and poolsb 2 Vases with live plants 1 bedded in the ground without a drain. Animal OtherC 2 drinking containers seldom produce mosquitoes while in active use because the water is changed Based on percentages of pupae in containers at 500 houses frequently and many animals (particularly ducks during September-October 1991. Breteau index, 38; total num­ and geese) eat the larvae in their drinking water. ber of pupae, 1,764. a More than 4-liter capacity. However, Ae. aegypti production can be high b Cement construction, on or in the ground, no drain. from a drinking container in a yard without ani­ C Tin cans, bottles, car batteries, toilets, pipes, scrap iron, mals. wheel barrels, and tree holes. Buckets, 40-liter plastic drums, pots, and large plastic bowls are numerous in yards, because much of the cooking and cleaning is done out­ clops longisetus, Mesocyclops venezolanus side. Most of these containers do not breed Ae. Dussart, and Mesocyclops thermocyclopoides aegypti because they are in active use and larvae Harada. Because the first three species occur are eliminated before they can complete their throughout much ofLatin America, the results of development. However, they may breed mosqui­ these tests should apply broadly to that region. toes if they are not in active use. A few families use cisterns (cement tanks, sev­ Materials and Methods eral meters long, embedded in the ground) to Containers. Laundry tanks, tires, 200-liter (55 store water. The cisterns have makeshift covers, gal) drums, animal drinking containers, buckets, and mosquitoes have easy access. Ornamental large pots, cisterns, ornamental pools, and vases pools are similar to cisterns except they are shal­ are responsible for nearly all the Ae. aegypti pro­ lower and not covered. The cisterns and pools duction in EI Progreso (Table 1). are seldom cleaned and often contain enough Most homes in EI Progreso have a cement detritus to provide abundant food for mosquito laundry tank outside the house to store water for larvae. They can be a primary source of Ae. ae­ washing clothes and dishes. The tanks are gypti in neighborhoods where they are located. needed because city tap water may be available Various containers-vases, tin cans, and plastic for only a few hours a day. Clothes and dishes are soft drink bottles tipped on their side with one not washed in the tanks themselves; water is side cut out-are used to hold live plants (e.g., dipped out for use and replaced from the tap philodendron) in water. Often the plant is a cut­ when available. Most housewives clean their ting with few roots, the water in the vase is clean, tanks about once a week by draining, scrubbing, and the carrying capacity for mosquito larvae is and refilling them. As with all other water stor­ low. Mosquito production from vases with a age containers in this study, the water from these large quantity of detritus can be high. tanks is not used for drinking. Field Collection and Preliminary Screening. Water also is stored in 200-liter drums for Cyclopoid copepods were collected from several household cleaning activities such as mopping hundred randomly selected aquatic sites in the floors arid washing down the toilet in the latrine. vicinity of EI Progreso. All species with an adult Houses without a laundry tank may use water female body length>1 mm (from the front ofthe from the drum to wash clothes and dishes. The cephalothorax to the end of the caudal ramus) drums are always uncovered. Water is dipped were considered large enough to be possible out as needed and replaced with tap water. Be­ predators of mosquito larvae. They were placed cause the water for floors or toilets does not have in dishes (10 cm diameter) for 24 h with newly to be very clean, many drums are cleaned only hatched Ae. aegypti larvae at a ratio of 10 larvae once a month. Drums are cleaned by tipping to per cyclopoid. Larval head capsules and man­ " pour out the water, spraying the inside with a gled dead bodies were considered to be a con­ hose, and returning the drum to an upright posi­ sequence of cyclopoid predation. tion. Because the drums are outside and trees are Following culture procedures described by numerous, even drums that are cleaned fre­ Suarez et al. (1992), laboratory colonies of Mac­ quently often have sediment or leaves at the bot­ rocyclops albidus sensu stricto, Mesocyclops tom. longisetus var. curvatus Dussart, M. venezola­ Some people have discarded tires associated nus, and M. thermocyclopoides sensu lato were with a home business, and others keep old tires established to produce the numbers required for 38 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 31, no. 1 the predation experiments and field trials de­ abundant during this period. The field trials with scribed below. vases started 2 mo later than other containers and Laboratory Predation Experiments. To assess lasted for 20 wk. the relative capacity ofeach ofthe four cyclopoid The containers used for the field trials in­ species to kill Ae.