Predation efficiency of indigenous larvivorous fish species on Culex pipiens L. larvae (Diptera: Culicidae) in drainage ditches in Argentina

102

June 2006

Journal of Vector Ecology

Predation efficiency of indigenous larvivorous fish species on Culex pipiens L. larvae (Diptera: Culicidae) in drainage ditches in Argentina Gerardo A. Marti , M. de las Mercedes Azpelicueta1, María C. Tranchida, Sebastián A. Pelizza, and Juan J. García Centro de Estudios Parasitológicos y de Vectores-Cepave (UNLP-Conicet), 2 No. 584, (1900) La Plata, Argentina Division of Vertebrate Zoology, Museum of Natural Sciences, Paseo del Bosque, Universidad Nacional de La Plata, 1900 La Plata, Argentina

1

Received 19 September 2005; Accepted 30 November 2005 ABSTRACT: Two neotropical freshwater fish species, Cnesterodom decemmaculatus (Poeciliidae) and Jenynsia multidentata (Anablepidae), were collected from human-made ditches, a common habitat of the house mosquito Culex pipiens in La Plata, Argentina. Cnesterodom decemmaculatus was recorded in 62 of the 100 examined ditches, whereas J. multidentata was collected from only 21 ditches sympatrically with C. decemmaculatus. Culex pipiens was the only mosquito species collected, and its larvae and pupae were found in 38 of the 100 ditches. Fish and mosquito larvae and pupae were collected together in only two ditches and were significantly negatively correlated. Siphons of larval Culex and remnants of chironomid larvae, copepods, aquatic mites, and fish were present in the gut contents of two C. decemmaculatus from mosquito-positive ditches, while diatoms and filamentous algae were recorded in every fish dissected. Adult C. decemmaculatus and J. multidentata needed approximately 6.2 h to completely digest one Cx. pipiens 4th instar larva under laboratory conditions. When fish were confined with a density of 60 or fewer Cx. pipiens 4th instar larvae, C. decemmaculatus and J. multidentata adults consumed 100% of them in one day but only 35% and 42%, respectively, when confined with 150 larvae. Eradication of Cx. pipiens from a ditch, where densities had averaged 250 immatures per dip, was achieved 17 days after the introduction of 1,700 C. decemmaculatus. Journal of Vector Ecology 31 (1): 102-106. 2006. Keyword Index: Culex pipiens, larvivorous fish, Cnesterodom decemmaculatus, Jenynsia multidentata, mosquito control.

INTRODUCTION Culex pipiens L. is a serious nuisance and pest mosquito in many areas of Argentina. However, the broadscale application of synthetic insecticides is often associated with high levels of insecticide resistance and environmental and health concerns. Because of these concerns, biological vector control plays an increasingly important role in integrated management strategies. Some fish species have been studied as potential biocontrol agents of larval and pupal mosquitoes. Even though it is possible to find many studies on larvivorous mosquito species (Gerberich and Laird 1968, Bay 1985, Meisch 1985, Nelson and Keenan 1992, Torrente et al. 1993, Lee 2000, Martinez-Ibarra et al. 2002, Hurst et al. 2004), to our knowledge there are no reports from Argentina. The mosquito fish Gambusia affinis (Baird and Girard) has been the most studied fish species for use as a biological control agent worldwide, but the World Health Organization (1982) discourages the introduction of exotic species because of the potential for negative ecological consequences (CDC 1973). Human-made drainage ditches in suburban areas of many cities in Argentina are among the most common larval habitats of Cx. pipiens and huge numbers of mosquitoes emerge throughout the year (Campos et al. 1993). The objectives of this study were to identify some larvivorous fish species indigenous to Argentina and to conduct laboratory and field

trials to determine the predation efficacy of these fish species on Cx. pipiens larvae. MATERIALS AND METHODS Study site A survey for larvivorous fish was conducted in 100 ditches in an area of 3.6 ha (600m x 600m) located 6 km from the city of La Plata (34º 55´S, 57º 57´W), Buenos Aires province, Argentina, from November 2003 to April 2004. These ditches by the sides of streets are part of a continuous system connected to the El Gato stream located 300 m from the study site, which flows into the Río de La Plata. These ditches averaged 50 cm wide and 15 cm deep and drained excess rain water and, occasionally, domestic waste water that contained detergent, soap, bleach, and grease from nearby houses. By the ditches there were areas with aquatic vegetation and/or trash. Collection techniques Fish were collected with a mesh aquatic net (25 cm x 25 cm) and transported to the laboratory in a 10-liter plastic bucket containing water from the collection site aerated by a battery-operated portable air pump. Mosquito larvae and pupae were collected with a 300 ml dipper. Densities of immature mosquito stages were calculated as averages from five dips per ditch. Fish identifications were based upon the

Vol. 31, no. 1

Journal of Vector Ecology

103

taxonomic keys of Ringuelet et al. (1967) and Ghedotti (1998).

Time for complete digestion of a mosquito larva Field-collected adult males (n=100) and females (n=100) of both fish species were individually placed in glass containers (10 cm x 10 cm x 10 cm) with 500 ml of dechlorinated water. Each fish was fed one field-collected Cx. pipiens 4th instar larva. Three fish from each species and sex were fixed in 10% formaldehyde at 15 min intervals after feeding. Fixed specimens were dissected and their digestive tracts removed. Gut contents were washed out and examined for the presence of mosquito larval remains under a light microscope. Laboratory feeding tests Feeding tests were conducted at 25±1°C and 12:12 (L:D) photoperiod. After a period of seven days of acclimation to laboratory conditions, during which the fish from ditches were kept in dechlorinated water and fed with Cx. pipiens larvae, adult fish (n=5) were individually placed in glass containers (10 cm x 10 cm x 10 cm) with 500 ml of dechlorinated water. To determine the number of larvae consumed by a single fish in 1-h, densities from 10 to 60 (increasing by 10) fieldcollected Cx. pipiens 4th instar larvae were added to each of the five containers allocated to each density. Surviving larvae were recorded from each container after the 1-h period. This experiment was repeated five times on different dates under similar conditions. To determine the daily consumption of mosquito larvae, single adult fish (n=5) were kept in a similar manner as the previous experiment. Densities from 10 to 150 (increasing by 10) Cx. pipiens 4th instar larvae were added to each of the five containers allocated to each density. Surviving larvae were counted daily and removed from each container. Individual fish received the same number of larvae every 24 h during a 7-day test period. Predation efficacy Larvae and pupae of Cx. pipiens were collected with a 300 ml dipper from ditches in the study area. They were counted and transported to the ditches under study in a plastic bucket with 5 liters of site water. An average of 8,000 larvae per ditch were released in three ditches (average 4 m long, 0.5 m wide, 0.15 m deep) into which fish were previously recorded, but their density had not been estimated. Numbers of larvae per dip (n=5) were recorded at time 0 (immediately after larval releases), 1, 3, and 6 h post-release from both sides of the ditches. After a 2-h post-release period, 10 fish from each ditch were collected with a net, dissected, and their gut contents examined for presence of mosquito larval remains as described previously.

Field release of fish in a ditch with mosquito larvae A ditch (5 m long, 0.5 m wide, 0.15 m deep) with Cx. pipiens larvae and pupae and without fish was selected for a preliminary study to determine the predatory efficacy of fish under field conditions. Fish were collected with an aquatic net in ditches from the study area, identified, counted, and transported in a 10-liter bucket with site water for release into the test ditch. Larval and pupal densities of Cx. pipiens in the ditch were determined from daily mean numbers per dip (n=5) from a week prior to introduction until four weeks after the introduction of fish. A control ditch without fish introduction located 50 m from the studied ditch was sampled for the same time period. Statistical analysis Analysis of variance (ANOVA) and comparison of means by Tukey’s test (P = 0.05) were utilized to compare larvae consumption by fish species with different larvae densities at 1 and 24 h. RESULTS Endemic fish and mosquito species Two species of fish were identified from the examined ditches, Cnesterodom decemmaculatus (Jenyns) (Poeciliidae) (2.2 ± 0.6 cm long) and Jenynsia multidentata (Jenyns) (Anablepidae) (2.6 ± 0.3 cm long). Fish were recorded in 62 of the examined ditches (n=100), C. decemmaculatus in all 62 fish-positive ditches, and J. multidentata in 21 of these, always with C. decemmaculatus. Culex pipiens was the only mosquito species collected in these ditches, its larvae and pupae were recorded in 38% of the ditches (n=38). Larval density ranged from less than one to a maximum of 478 per dip. Fish and mosquitoes were collected in the same ditches in only two ditches, and mosquito larval density was very low in these two, less than one larva per 10 dips. A G-test for independence with Williams’ correction (Sokal and Rohlf 1995) indicated a significant negative association between the presence of fish and Cx. pipiens immatures (Gcorr.= 98.4, 1 df, P