The use of annual killifish in the biocontrol of the aquatic stages of mosquitoes in temporary bodies of fresh water; a potential new tool in vector control
© Matias and Adrias; licensee BioMed Central Ltd. 2010
Received: 13 January 2010
Accepted: 21 May 2010
Published: 21 May 2010
Mosquitoes that breed in temporary pools in remote areas that dry up seasonally are especially difficult to control through chemical or biological means. The annual killifish has been suggested as a means of eradicating the aquatic stages of mosquitoes in transient pools because they can maintain permanent populations in such habitats by undergoing suspended animation or diapause during the embryonic stages to survive periodic drought. However, very little is known about the predatory activity of annual killifish and their usefulness in mosquito control.
The annual killifish, Nothobranchius guentheri, native to Tanzania, was used in this investigation. Food preference was tested under laboratory conditions by feeding juvenile killifish with 2nd instar mosquito larvae of Culex quinquefasciatus in the presence of alternative food sources, such as rotifers and chironomid larvae. Semi-field tests were conducted by introduction of hibernating killifish embryos and juvenile fish to artificial ponds in an outdoor open environment that allowed natural oviposition of Cx. quinquefasciatus. Food preference studies show that N. guentheri preferred to prey on mosquito larvae than either chironomid or rotifers. When hibernating killifish embryos were added to ponds simultaneously with the addition of freshwater, the embryos hatched and fed on mosquito larval population resulting in complete elimination of the immature stages. The introduction of juvenile fish to ponds with high density of mosquito larvae resulted in total eradication of the mosquito population due to predation by fish. Complete biocontrol of the mosquito larval population was achieved in the presence of 3 fish per m2 of pond surface area.
The annual killifish provides yet another tool that may be employed in the eradication diseases carried by mosquitoes through vector control, particularly in temporary bodies of freshwater. The fish can be conveniently transported in the absence of water in the form of hibernating embryos. Once introduced either as embryos or juveniles in ponds, the annual killifish can effectively reduce the larval population because of its aggressive predatory activity.
Vector control using pesticides remains an important component of all mosquito control program worldwide. However, the persistent use of pesticides caused the development of chemically resistant substrains of mosquitoes. This chemical resistance in mosquitoes is increasing with new reports of resistance emerging in various areas [1–3]. For this reason and because of the ecosystem damage caused by such chemicals, there has been renewed interest in biological control techniques to complement mosquito control programs. The use of larvivorous fish is the oldest biocontrol method that is still employed today. The mosquito fish (Gambusia affinis), native to northeastern United States, was eventually selected for worldwide mosquito control because of its high larvivorous capacity, high fecundity and adaptability to new environments. However, the success of Gambusia in mosquito control was overshadowed by its intrinsic aggressive nature that drove many other native species of aquatic organisms to the brink of extinction . More recent efforts had been focused on developing indigenous fish and other less aggressive aquarium fish as alternative to Gambusia in providing the same mosquito larval control in permanents streams and pools [5–7].
Mosquito larval control in breeding sites comprising pools that dry up seasonally limits the use of conventional fish. Transporting fish to remote areas is both logistically challenging and expensive. For these reasons, attention had been focused on a unique group of freshwater fish, collectively known as annual killifish, native to Africa and South America that can maintain populations in seasonal pools. These killifish can survive prolonged drought, sometimes lasting as much as five years, by undergoing suspended animation or diapause at three stages of their embryonic development [8, 9]. At the end of the wet season, the adults die and the population survives in the form of diapausing embryos encased in the dry substrate . These diapause embryos are more resistant to environmental extremes, such as high temperature, anoxia and physical/chemical damage, compared to non-diapause stages [10–14]. At the start of the first rains, the embryos hatch to re-populate the seasonal pools . This life cycle mimics their main food source, which are mosquitoes and other insects that also begin their life cycle in the same seasonal pools. This convergent evolution provides a compelling reason for the interest in the potential of annual killifish for vector control in seasonal habitats.
The first record of annual killifish being used for this purpose was during World War II when Vanderplank initiated the introduction of Nothobranchius taeniopygus to combat malaria in seasonal swamps in Tanganyika and Kenya . In 1952, the noted ichthyologist, George Myers, observed that the annual killifish may have potential in biological control since mosquito bites were diminished in areas normally populated by these fish . Years later, others have made similar recommendations [18, 19]. Some preliminary attempts made by the World Health Organization to use the annual killifish for mosquito control provided some information about the conditions necessary for proper introduction of annual killifish [18, 20]. However, the limited financial support for a long term biocontrol research and inadequate knowledge about the environmental factors that control diapause had presented operational difficulties.
Much more is now known about the life cycle of Nothobranchius guentheri, an annual killifish indigenous to Tanzania. A hypothesis regarding its survival strategy in nature has already been proposed . What is lacking before any serious attempts at field introductions is basic information regarding the larvivorous nature of these killifish. As part of an ongoing program to develop the use of annual killifish for mosquito control, the laboratory and semi-field studies in this report are designed to explore the larvivorous potential of N. guentheri and ways these fish might be employed in mosquito control.
The breeding population of the annual killifish, Nothobranchius guentheri, was established at the Poseidon Sciences field laboratory in the Philippines where all the experiments described in this paper was conducted. This population was derived from the original population maintained in the United States since 1975 and interbred recently with N. guentheri breeding pairs supplied by Mr. Fred Behrman of Athens Aquatics (New York, USA) and Mr. Chris Butcher (Florida, USA). The killifish were maintained in 10 gallon aquaria according to previously described methods . The spawning occurred in the tanks with substrate of fine sand. The fertilized eggs were collected and incubated in petri dishes at 25°C until they reach the pre-hatching state, where the embryo can remain quiescent for at most 3 months. Thereafter, they were transferred into moist sphagnum horticultural moss (Mosser Lee Co., Millston, Wisconsin, USA) at densities of 50 embryos per 25 g of moist peat moss in a transparent polyethylene plastic bag. After 30 days under incubation temperature of 25 ± 2°C, the embryos were competent to hatch and remained in the pre-hatching stage until placed in water for the hatching process to commence. Upon hatching, the fry were kept in glass aquaria and fed newly hatched Artemia for the first 2 weeks. Thereafter, the fry were fed live fruit fly larvae (Drosophila melagonaster) until they reach the juvenile size of 1.5 cm in length.
Laboratory and semi-field studies were conducted in the island of Panay in the Philippines ( 10°40'25.17N, 121°57'39.47E) during the rainy season from May 2009 and ending in October, 2009. The water temperature ranged from 27° to 30°C, with pH in the range from 7.0 to 8.0. The dissolved oxygen level was 4 ppm. The artificial ponds were constructed near banana trees, with the overhanging leaves providing partial shade. The nearby trees also provided shade and leaves were permitted to fall naturally onto the experimental ponds. The only species of mosquitoes occurring naturally at this location was Culex quinquefasciatus. These mosquitoes were allowed to naturally oviposit in the experimental ponds. Other aquatic species present in the ponds in significant numbers were chironomids or midge larvae (Chironomus plumosus) and rotifers (Brachionus sp.).
Feeding with rotifers and mosquito larvae
To study the food preference of N. guentheri, the prey organisms and the killifish were kept in glass tanks (11 cm depth × 29 cm length × 23 cm width) containing 7 liters of pond water. Freshwater rotifers were collected from an existing pond by passing the water through a fine mesh net. The concentrated freshwater rotifers were placed in the glass tanks filled with pond water. The rotifer count was determined by taking a 1 ml sample of the water and counting the number of rotifers under the microscope. 2nd instar larvae of Cx. quinquefasciatus were collected from ponds and also introduced in the glass tank either separately or in combination with rotifers in the presence of 1 male fish, which was starved for 24 hours prior to the experiment. The rotifer and mosquito larval consumption was obtained at 30 min and 120 min after killifish introduction to the tank.
Feeding with chironomid and mosquito larvae
Chironomid larvae and 2nd instar larvae of Cx. quinquefasciatus were manually collected from the artificial pond and introduced into the test aquaria with killifish as described above. The number of larvae remaining in the tank after a given time period was determined.
All laboratory tests were conducted during the light phase of the artificial photoperiod (14 h light phase: 10 hour dark phase) at water temperature of 27 ± 1.5°C. All freshwater used in this study was obtained from an artificial well.
All of the mosquito larvae and pupae were harvested from each test pond by using fine mesh dip nets. Given the small size of the test ponds, it was possible to harvest all mosquito larvae to obtain more reliable data on larval abundance. Cx. quinquefasciatus, the only mosquito species found in the pond, was characterized according to the stages (instars and pupae). The larvae and pupae were placed in shallow pans for counting and returned back to the pond within 3 hours. For simplicity, the total mosquito count represents all larvae and pupae combined. The chironomids were counted by determining the total number of intact cocoons found underneath leaves in the bottom substrate. The rate of growth of the annual killifish in the ponds was determined by measuring the standard length as the distance from the tip of the snout to the end of the caudal peduncle.
Effect of introduction of annual killifish fish embryos
To determine if newly introduced embryos of N. guentheri can hatch, survive and feed on mosquito larvae, 1 m × 1 m × 0.3 m plastic lined ponds were constructed using 0.75 mm polyethylene plastic sheeting. These artificial ponds were filled with freshwater initially to a height of 5 cm. The killifish embryos in peat moss were introduced to the pond on the same day. Each group is composed of 5 ponds. No other food was provided to the fries and the pond water was allowed to be augmented by the daily rains to a final height of 0.3 m. The bottom substrate was composed of sandy-muddy soil at a thickness of 1 cm. Cx. quinquefasciatus was allowed to naturally oviposit eggs simultaneously with the introduction of the fish embryos.
Effect of the introduction of juvenile killifish
Artificial ponds (1 m × 2 m) were constructed and filled with water to height of 0.3 m. Cx. quinquefasciatus was allowed to naturally oviposit in the ponds. After one week, the ponds showed substantial presence of mosquito larvae at which time a total of 5 juvenile killifish at 25 days of age (3 females and 2 males) measuring 1.5 cm in length were placed in each pond. The total number of mosquito larvae and pupae was determined at weekly intervals. The controls represented ponds that did not receive any killifish. Each test group consisted of 3 ponds. After 28 days, the killifish in the experimental ponds were removed to determine the time for mosquito population to recover in the absence of fish. Also at day 28, five killifish (3 females and 2 males) were added to the control ponds to determine the effect of the killifish in ponds with an already high mosquito density. No mortalities were observed during the transfer of the killifish and during length measurements.
To determine the fish density required to eradicate mosquito larval population in the ponds, juvenile male annual killifish were added to 1 m2 ponds at densities from 0 to 5 fish per pond. Each test group was composed of 5 ponds. Cx. quinquefasciatus was allowed to naturally oviposit eggs for 1 week to provide a natural population of mosquito larvae. The total number of larvae present in the ponds was determined 4 days after fish introduction.
The statistical significance between control and experimental groups was evaluated using Student's t test.
Food preference of male N. guentheri in the presence of rotifers and mosquito larvae as prey.
Mosquito larvae consumed
(% of initial count)
(% of initial count)
Mosquito larvae + rotifers
33.3 ± 3.3
61.3 ± 4.4
7.3 ± 3.3
24.1 ± 5.4
Mosquito larvae alone
32.0 ± 3.2
69.6 ± 4.5
30.3 ± 3.8
53.3 ± 3.9
The data described in this report validate the potential of using the annual killifish, N. guentheri, for mosquito control because of its preference for mosquito larvae as prey, the successful demonstration of introducing the fish in the ponds in the form of diapausing eggs and the eradication of mosquito larval population in the ponds. The successful use of fish in mosquito control has been described before. Gosh and Dash  has demonstrated success in controlling An. culicifacies in wells and in ponds in India using only fish as the method of control.
Food preference for mosquito larvae was evident when either chironomids or rotifers were simultaneously presented as prey. The size of the prey is not likely a major factor since the average size of the 2nd instar mosquito larvae at 0.6 mm was not markedly different from the mean length for chironomid at larvae at 0.7 mm. Rotifers were much smaller prey measuring approximately 0.2 mm. The annual killifish would eat rotifers and chironomids as alternate food source if no other food was available, but overwhelmingly preferred mosquito larvae. Like most annual killifish, N, guentheri is a surface feeder and specially adapted to feed on aquatic invertebrates at or near the surface of the pond. The annual killifish will actively chase after the mosquito larvae upon sensing movements. Hence, the preference for mosquito larva that spends most of its time at the surface. In contrast, chironomids are bottom dwelling and typically found underneath decaying leaves. Rotifers are free swimming throughout the water column and likely preferred by the young killifish fry during the first week after hatching because of their small size.
The high predatory activity of N. guentheri on culicine larvae strongly suggests that it is a suitable candidate for mosquito control. The results in this study show that a minimum of 3 killifish is adequate to eradicate the larval population from 1 m2 of pond surface area. Whether anopheline mosquitoes will be equally preyed upon like the culicine mosquitoes in this study or whether other annual killifish will have similar aggressive predatory activity remains to be established. Recently, Louca, et al  demonstrated that culicine mosquitoes avoid ovipositing in water containing fish. This may explain the total absence of new culicine mosquitoes in the present study after the introduced fish had decimated the larval and pupal stages. However, Anopheles gambiae do not share the same oviposition behavior and will continue to lay eggs in the presence of fish in the pond . The present study is now being expanded to look into the predatory behavior of annual killifish in Tanzania where An. gambiae is endemic.
The onset of diapause sets the annual killifish apart from conventional fish and provides a novel means of providing biological control in transitory bodies of freshwater. The convenience of transporting and disseminating embryos for mosquito control purposes remain a major advantage particularly in remote areas where insecticide resistance is rampant and places where chemical use is not wanted by the local population. Even if re-introduction of new embryos becomes necessary during the next rainy season, this task can be easily accomplished by local mosquito control officers or even by the local population. The likelihood that introduced eggs will maintain sustainable populations in any new habitats is still open for exploration. The substrate in a typical Nothobranchius habitat is composed of black or gray mud, commonly referred to as "black cotton soil" or vertisols which consist of swelling clays, such as montmorillonite, with high water retaining capacity [18, 28]. Nothobranchius is never observed to maintain populations in habitats with red clays or oxisols. This habitat requirement may limit the type of environments that may sustain Nothobranchius species.
There are, however, ecological issues of concern before annual killifish can be employed for this purpose. Introduction of N. guentheri to other areas, even within Tanzania, may pose some risk to contamination of other transient pools inhabited by other species of Nothobranchius. However, field studies have shown that in many instances Nothobranchius species can be found in natural settings already intermixed with other species of annual killifish [18, 29]. While N. guentheri is used for this study merely as a demonstration of the potential of using annual killifish for mosquito control and should be limited for use in its natural geographic range, there are other species, such as N. melanospilus, that have a wider distribution in Tanzania that may be better adapted to the different conditions found in the country (B. Watters, personal communication).
Another issue of concern is whether introduction of this species in new habitats may pose risks to other indigenous aquatic organisms. Annual killifish survive only in temporary pools and require a period of drying to complete its life cycle. For this reason, annual killifish are never found in permanent bodies of water. Since they are typically the dominant inhabitant in a transient pool, annual killifish may not have evolved behavioral adaptations to compete against other fishes. In multi-species aquaria for example, annual killifish do not survive well. The aggressive nature of N. guentheri is only evident when males compete for the attention of females for reproduction and to establish dominance hierarchies . While it is likely for introduced killifish to migrate to permanent waters due to flooding or transport of eggs by birds or other animals, it is quite unlikely that annual killifish can maintain permanent presence in such habitats. This question needs to be further elucidated with controlled studies to accurately determine the potential ecological impact of introducing annual killifish to new environments.
There are many practical advantages of the annual killifish in mosquito control. First, there is a wide range of indigenous species of annual fish found in South America and sub-Saharan Africa. It is therefore feasible to use indigenous annual killifish species for any given malarial region rather than introducing an exotic species to new habitats. Second, the use of diapausing embryos may allow in suitable areas to have recurring populations in transient bodies of freshwater so that re-introductions will not be necessary, thereby reducing cost of maintaining mosquito control. Vectors of malaria and dengue propagate in both permanent and in transient pools, such as tire tracks, water collecting stations, animal footprints and depressions created by mining activities. Transporting conventional live fish for biocontrol measures in such situations will be difficult and costly. On the other hand, transporting thousands of annual killifish embryos in peat moss or in a more convenient delivery system would make it easier and more economical. Third, the annual killifish are small and less likely to be used as food source. The small size also allows easy access to shallow areas where mosquito larvae tend to congregate. Fourth, the annual killifish is not aggressive against other fish, cannot survive in permanent bodies of freshwater and likely pose minimal ecological issues. And finally fifth, as shown in this report, the annual killifish is sufficiently larvivorous to exert substantial control of the immature stages of the mosquito population. All of these characteristics satisfy World Health Organization's recommendations on the selection of larvivorous fish for mosquito control .
In summary, this study demonstrates that the annual killifish, Nothobranchius guentheri, aggressively preys on culicine larvae preferentially compared to other aquatic food sources under both laboratory and semi-field conditions. Total control of the immature stages of mosquitoes in freshwater can be achieved with the introduction of killifish. More important, introduction of hibernating or diapause embryos serves as a convenient means of disseminating killifish in transient pools. This study suggests that annual killifish may be useful in the future as part of the integrated program in vector control.
The authors are grateful for the technical assistance of the following individuals for maintaining the breeding populations: Kristine J. Matias, Nandy Brijlal and Katherine L. Matias. We also thank Jaleel A. Lorella, Lourdes Q. Adrias and Jacob A. Lorella for their assistance in the laboratory and semi-field studies. The expert assistance by Sheila V. Serrano and Coleen P. Sucgang during the preparation of this manuscript is gratefully acknowledged.
- Munhenga G, Mansendu HT, Brooke BD, Hunt RH, Koekemoer LK: Pyrethroid resistance in the major malaria vector Anopheles arabiensis from Gwave, a malaria-endemic area in Zimbabwe. Malar J. 2008, 7: 247-10.1186/1475-2875-7-247.PubMed CentralView ArticlePubMedGoogle Scholar
- Dabire KR, Diabate A, Djogbenou L, Ouari A, Guessan RN, Ouedraogo JB, Hougard JM, Chandre F, Baldet T: Dynamics of multiple insecticide resistance in the malaria vector Anopheles gambiae in a rice growing area in Southwestern Burkina Faso. Malar J. 2008, 7: 188-10.1186/1475-2875-7-188.PubMed CentralView ArticlePubMedGoogle Scholar
- Ranson H, Abdallah H, Badolo A, Moussa WH, Kerah-Hinzoumbe C, Yangalbe-Kalnone E, Sagnon N, Simard F, Coetzee M: Insecticide resistance in Anopheles gambiae: data from the first year of a multi-country study highlight the extent of the problem. Malar J. 2009, 8: 299-10.1186/1475-2875-8-299.PubMed CentralView ArticlePubMedGoogle Scholar
- Hamouda S, Samraoui B: Ecological impact of Gambusia affinis (Cyprinodontiform: Poeciliidae) on aquatic environments. J Anim Vet Advances. 2007, 6: 828-832.Google Scholar
- Howard AFV, Zhou G, Omlin FX: Malaria mosquito control using edible fish in western Kenya: Preliminary findings of a controlled study. BMC Public Health. 2007, 7: 199-10.1186/1471-2458-7-199.PubMed CentralView ArticlePubMedGoogle Scholar
- Ghosh A, Bhattacharjee I, Ganguly M, Mondal S, Chandra G: Efficacy of some common aquarium fishes as biocontrol agent of preadult mosquitoes. Bul Penel Kesehatan. 2004, 32: 144-149.Google Scholar
- Ghosh SK, Tiwari SN, Sathyanarayan TS, Sampath TRR, Sharma VP, Nanda N, Joshi H, Adak T, Subbarao SK: Larvivorous fish in wells target the malaria vector sibling species of Anopheles culicifacies complex in villages in Karnataka, India. Trans Royal Trop Med Hygiene. 2005, 99: 101-105. 10.1016/j.trstmh.2004.03.009.View ArticleGoogle Scholar
- Wourms JT: Developmental biology of annual fishes. I. Stages in the normal development of Austrofundulus myersi Dahl. J Exp Zool. 1972, 182: 143-168. 10.1002/jez.1401820202.View ArticlePubMedGoogle Scholar
- Wourms JT: Developmental biology of annual fishes. II. Naturally occurring dispersion and re aggregation of blastomeres during the development of annual fish eggs. J Exp Zool. 1972, 182: 169-200. 10.1002/jez.1401820203.View ArticlePubMedGoogle Scholar
- Matias JR, Markofsky J: The survival of the embryos of the annual fish, Nothobranchius guentheri, exposed to temperature extremes and the subsequent effects on embryonic diapause. J Exp Zool. 1978, 204: 219-227. 10.1002/jez.1402040209.View ArticleGoogle Scholar
- Matias JR: Embryonic diapause in annual fishes: Evaporative water loss and survival. Experientia. 1982, 38: 1315-1317. 10.1007/BF01954923.View ArticleGoogle Scholar
- Matias JR: The effect of exposure to gaseous ammonia on the duration of diapause II in the embryos of the annual fish Nothobranchius guentheri. Experientia. 1983, 39: 1148-1150. 10.1007/BF01943154.View ArticleGoogle Scholar
- Matias JR: The stage-dependent resistance of the chorion to external chemical damage and its relationship to embryonic diapause in the annual fish, Nothobranchius guentheri. Experientia. 1984, 40: 753-754. 10.1007/BF01949761.View ArticlePubMedGoogle Scholar
- Podrabsky JE, Lopez JP, Fan TWM, Higashi R, Somero GN: Extreme anoxia tolerance in embryos of the annual fish Austrofundulus limnaeus: insights from a metabolomics analysis. J Exp Zool. 2007, 210: 2253-2266.Google Scholar
- Wourms JT: Developmental biology of annual fishes. III. Pre-embryonic and embryonic diapause of variable duration in the eggs of annual fishes. J Exp Zool. 1972, 182: 389-414. 10.1002/jez.1401820310.View ArticlePubMedGoogle Scholar
- Vanderplank FL: Nothobranchius and Barbus species: indigenous anti-malaria fishes in East Africa. East Africa Med J. 1941, 17: 431-436.Google Scholar
- Myers GS: Annual Fishes. Aquar J. 1952, 22: 145-151.Google Scholar
- Wildekamp RH: Assignment report. Preliminary study of the Somalian Nothobranchius species of larvivorous fishes. EM/MAL/198. EM/VBC/41. SOM/MPD/001/RB. World Health Organization
- Markofsky J, Matias JR: Waterborne vectors of disease in tropical and subtropical areas and novel approach to mosquito control using annual fish. The Columbia University Seminars on Pollution and Water Resource. 1979, 12: H1-H17.Google Scholar
- Haas R: Assignment report. Malaria control programme. EM/VBC/18, EM/MAL/164. 1979, World Health OrganizationGoogle Scholar
- Matias JR: Annual fish biology and survival strategy in nature. [http://www.poseidonsciences.com/Annual_Fish_Biology_and_Survival_Strategy.pdf]
- Markofsky J, Matias JR: The effects of temperature and season of collection on the onset of diapause in embryos of the annual fish Nothobranchius guentheri. J Exp Zool. 1977, 202: 49-56. 10.1002/jez.1402020107.View ArticlePubMedGoogle Scholar
- Ghosh SK, Dash AP: Larvivorous fish against malaria vectors: a new outlook. Trans Royal Soc Trop Med Hygiene. 2007, 101: 1063-1064. 10.1016/j.trstmh.2007.07.008.View ArticleGoogle Scholar
- Hay S, Carlos AG, Gething P, Patilz AP, Tatem AJ, Noor AM, Kabaria CW, Manh BH, Elyazar IRF, Brooker S, Smith DL, Moyeed RA, Snow RW: A world malaria map: Plasmodium falciparum endemicity in 2007. Plos Med. 2009, 6: 286-View ArticleGoogle Scholar
- Watters BR: The genus Nothobranchius. [http://www.aka.org/aka/modules/content/index.php?id=9]
- Bhattacharjee I, Aditya G, Chandra G: Laboratory and field assessment of the potential of larvivorous, air-breathing fishes as predators of culicine mosquitoe. Biol Control. 2009,Google Scholar
- Louca V, Lucas ML, Green C, Majambere S, Fillinger U, Lindsay SW: Role of fish as predators of mosquito larvae on the floodplain of the Gambia River. J Med Entomol. 2009, 43: 546-556.View ArticleGoogle Scholar
- Watters B: The ecology and distribution of Nothobranchius fishes. J Am Killifish Assoc. 2009, 42: 58-61.Google Scholar
- Reichard M, Polacik M, Sedlacek : Distribution, colour polymorphism and habitat use of the African killifish, Nothobranchius furzeri, the vertebrate fish with the shortest lifespan. J Fish Biol. 2009,Google Scholar
- Haas R: Sexual selection in Nothobranchius guentheri (Pisces: Cyprinodontidae). Evolution. 1976, 30: 614-622. 10.2307/2407584.View ArticleGoogle Scholar
- World Health Organization: Use of fish for mosquito control. EM/MAL/289/E/G. 2003, WHOGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.