SEARCH

SEARCH BY CITATION

Keywords:

  • Anopheles arabiensis;
  • An. gambiae;
  • An. quadriannulatus;
  • cannibalism;
  • competition;
  • larval populations;
  • malarial vectors;
  • mosquito larvae;
  • population dynamics;
  • predation;
  • Kenya

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Among the aquatic developmental stages of the Anopheles gambiae complex (Diptera: Culicidae), both inter- and intra-specific interactions influence the resulting densities of adult mosquito populations. For three members of the complex, An. arabiensis Patton, An. quadriannulatus (Theobald) and An. gambiae Giles sensu stricto, we investigated some aspects of this competition under laboratory conditions. First-instar larvae were consumed by fourth-instar larvae of the same species (cannibalism) and by fourth-instar larvae of other sibling species (predation). Even when larvae were not consumed, the presence of one fourth-instar larva caused a significant reduction in development rate of first-instar larvae. Possible implications of these effects for population dynamics of these malaria vector mosquitoes are discussed.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The most important vector of malaria in sub-Saharan Africa, Anopheles gambiae Giles sensu lato, exploits dynamic freshwater habitats, such as shallow rain-fed pools, puddles and car tracks (Service, 1993). Biotic factors in and around these aquatic habitats, such as the presence of predators, parasites and pathogens (Service, 1973, 1977; Aniedu et al., 1993) and availability and quality of food (Gimnig et al., 2002), affect the survival and growth of An. gambiae s.l. larvae. Abiotic factors, such as temperature (Lyimo et al., 1992) and alkalinity (le Sueur & Sharp, 1988), also affect larval population dynamics.

Annual wet seasons occur throughout the natural range of An. gambiae s.l. but the rainfall is often erratic. A few days without rain may cause the larval habitats to dry out rapidly (Minakawa et al., 2001), whereas large amounts of rain might cause breeding sites to be flushed out (Charlwood et al., 1995; Gimnig et al., 2001). These dynamic conditions may cause stress amongst, and extreme competition between, larvae occupying a breeding site.

Schneider et al. (2000) showed that An. arabiensis and An. gambiae, the two most prevalent species of the An. gambiae complex in Africa, differ in their sensitivity to crowded conditions. In mixed species larval populations, An. gambiae s.s. survived better – to the detriment of An. arabiensis. By contrast, An. arabiensis developed significantly faster to the pupal stage than An. gambiae s.s.Schneider et al. (2000) also investigated the possibilities of cannibalism and/or predation by direct behavioural studies. As they could not detect larvae eating others, they ruled out this mechanism as a possible explanation for the observed high mortality in their experimental trays.

Cannibalism has been described in malaria vectors from several parts of the world. Balfour (1921) first reported that An. gambiae s.l. (then known as An. costalis Loew) larvae become cannibalistic when deprived of food. Hinman (1934) concluded, from dissecting 600 larvae of An. punctipennis Say, that this Nearctic species was cannibalistic. Among malaria vectors of the Indian subcontinent, Reisen & Emory (1976) observed that nearly all first-instar larvae of An. stephensi Liston were consumed by fourth instars of the same species; similarly Reisen et al. (1982) found that An. culicifacies Giles shows cannibalistic behaviour. Shoukry (1980) investigated the widespread African malaria vector An. pharoensis Theobald (Gillies & DeMeillon, 1968) and found body parts of young conspecific larvae in the midgut of older larvae. Other anophelines reportedly cannibalistic, although evidence is scant, are An. maculatus Theobald (Senior-White, 1926) and An. messeae Falleroni (Gordeev & Troshkov, 1990). From these studies it can be concluded that cannibalism is not unusual among anopheline mosquitoes. To improve understanding of this behaviour and its potential ecological implications, we investigated the occurrence of predation and cannibalism within and between members of the An. gambiae complex. In this context, we assessed effects of larval food availability and the presence of older instars on the development of younger instars.

Materials and methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Mosquitoes

For the predation experiment, first and fourth-instar larvae of laboratory-adapted strains of An. gambiae s.s. (SUAKOKO) and An. quadriannulatus Theobald (SANGQUA) were used. These strains originated from Liberia and Zimbabwe, respectively, and were reared in climate-controlled rooms at the Laboratory of Entomology, Wageningen University, the Netherlands, at 28°C and 80% relative humidity, with a 12 h photoperiod. Adults were fed twice per week on human blood (An. gambiae s.s.) or daily on cow blood (An. quadriannulatus) and kept in 30-cm cubic cages provided with 6% sucrose solution as maintenance diet. Larvae were reared in 2.5-L trays filled with tapwater and fed Tetramin® fish-food. Experiments were conducted in the climate-controlled rooms.

For the cannibalism experiment, first and fourth-instar larvae of An. arabiensis were used. These larvae were third generation offspring of female An. arabiensis that were collected in Ahero (00°10′ S, 34°55′ E), Kenya. This strain was maintained under seminatural conditions at the Centre for Vector Biology and Control Research (CVBCR) of the Kenya Medical Research Institute (KEMRI) in Kisumu, Kenya. Adult An. arabiensis females were offered bloodmeals daily on a human arm and kept in cages as above. Larvae were reared in plastic trays filled with water (from a river 20 km to the east of Kisumu) and fed Tetramin® fish-food.

Predation

Using reciprocal combinations of An. gambiae s.s. and An. quadriannulatus, fourth-instar larvae of one species (the potential predator) were placed individually in plastic cups (6 cm Ø; 3.5 cm high) containing 50 ml distilled water. Ten first-instar larvae of the other species (potential prey) were added to each cup and fed once with Tetramin® at a rate of 0.1, 0.2 or 0.3 mg/larva or none. The experiment was replicated eight times for each predator/prey combination and quantity of food.

For scoring possible predation, the first-instar larvae were counted every 30 min for 4 h. If fewer than 10 larvae remained, the fourth-instar larva was removed, killed and stored (in an Eppendorf tube with silica-gel for desiccation) for species diagnostic DNA-assay. If no larvae had disappeared during the first 4 h of the experiment, the number of first-instar larvae was checked again 20 h later. As before, if fewer than 10 first-instar larvae were present after the total period of 24 h exposure to the fourth-instar larva, it was removed and processed as described above.

Detection of ‘prey’ DNA

From cups from which first-instar larvae had disappeared, the fourth-instar larva was subjected to a species diagnostic DNA-assay (using the rDNA-PCR method of Scott et al., 1993). DNA was extracted from the larva as described by Schneider et al. (2000) and 1 µl of the undiluted extract was used for PCR. It was expected that if a fourth-instar larva had consumed any first-instar larva(e) of the other species, then the diagnostic DNA-bands of both species (prey as well as predator species) would be detected, instead of only the diagnostic band of the fourth-instar (predator) larva of known identity. For the positive controls we used extracts of fourth-instar larvae directly taken from the mosquito cultures.

Cannibalism

To test whether cannibalism occurs, and to assess the effect of the presence of fourth-instar larvae on the development of first-instar larvae, we used An. arabiensis to compare four treatments: (A) as a control, 24 first-instar larvae were reared individually, without any other larvae in their plastic well (see below); treatments (B), (C) and (D) each comprised 24 first-instar larvae reared individually in the presence of a fourth-instar larva. Treatments (A) and (B) were fed 0.4 mg Tetramin® per well per day; treatment (C) was fed 0.8 mg/well/day; treatment (D) was fed 1.6 mg/well/day.

We used plastic plates containing 12 wells (normally used for tissue culture) as rearing units. These wells (2.2 cm Ø, 1.7 cm deep, 6.5 ml volume) were filled with river water (as above). First-instar larvae were placed individually in these wells, with or without a conspecific fourth-instar larva. The plates were kept on a laboratory table (smeared with Vaseline® to prevent ants from disturbing the experiment). Mean temperatures during the experiment were 20.8 ± 0.4°C minimum and 29.5 ± 1.0°C maximum.

We recorded daily whether the experimental larva (i.e. the larva that had been put in the well as a first instar) was still present and alive. If so, we recorded its instar stage, and the experiment was terminated as soon as this larva moulted to the third instar. To preclude pupation of the fourth-instar larvae during the experiment, they were replaced daily by new fourth-instar An. arabiensis larvae from the CVBCR insectary.

Larval faeces examination

Because the specific rDNA-PCR assay cannot distinguish first from fourth-instar DNA of the same species, we collected larval faeces (using a fine pipette) from the bottom of the wells in which the first-instar An. arabiensis larva had disappeared. Faeces (mostly of the fourth-instar larva) from each well was spread on a glass slide and the particles were then examined (under a microscope at 100 × magnification) for remnants of the missing first-instar larva.

Statistical analysis

In the predation experiment, χ2-tests were used to test for differences in predation frequency (i) between the treatments with different amounts of food and (ii) between the two predator/prey combinations.

In the cannibalism experiment, time until death was analysed by means of Kaplan–Meier survival analysis (Kleinbaum, 1996). If the experimental larvae survived, their development time to third instar was recorded and analysed with Student's t-tests.

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Predation

Predation by the fourth-instar larvae of each sibling species on first instars of the other occurred readily under the experimental conditions (Table 1) and the quantity of food available (Tetramin®) had no significant effect on the predation rate (χ2-test, P > 0.05). During the first 4 h of exposure, more An. gambiae s.s. (13/32 = 41%) than An. quadriannulatus (8/32 = 25%) consumed at least one of the 10 first-instar larvae available, but this difference was not significant (P > 0.05). After 24 h, predation of first instars had occurred in approximately half of the experimental cups, with no significant difference observed between the predator species: An. gambiae s.s. 18/32 (56%) vs. An. quadriannulatus 14/32 (44%).

Table 1. . Numbers of cases of interspecific predation on first-instar larvae (L1, 10/test) by single fourth-instar larvae (L4) of Anopheles gambiae s.s. (GAM) and Anopheles quadriannulatus (QUAD), with initial availability of different quantities of food (Tetramin®).
Food quantity‘Predator’‘Prey’ No. casesTotal no. casesNo. cases PCR positive
(mg/L1)L4L1Replicatesafter 4hafter 24hwith prey DNA
  • *

    See text for significance of differences.

0GAMQUAD8350
0.1GAMQUAD8450
0.2GAMQUAD8340
0.3GAMQUAD8340
Total*  3213180
0QUADGAM8241
0.1QUADGAM8230
0.2QUADGAM8243
0.3QUADGAM8232
Total*  328146

Detection of ‘prey’ DNA

DNA of An. gambiae s.s. was detected in six of 14 An. quadriannulatus fourth-instar larvae that had consumed at least one first-instar larva of the other species (see five examples in Fig. 1). By contrast, we did not detect the diagnostic DNA of An. quadriannulatus in any of the 18 fourth-instar larvae of An. gambiae s.s. that had apparently consumed first instars of the other species (Table 1).

image

Figure 1. DNA banding patterns of fourth-instar (L4) An. quadriannulatus larvae that had preyed on first instar (L1) An. gambiae s.s. larvae. Lane 1, low ladder; lane 2, negative control; lane 3, positive control An. quadriannulatus; lane 4, positive control An. gambiae s.s.; lanes 5–9, L4 An. quadriannulatus that consumed L1 An. gambiae s.s.

Download figure to PowerPoint

Cannibalism

Table 2 shows the percentage survival and average development time of An. arabiensis larvae from first to third instar when individually kept in a 6.5 ml well. Comparing those given the minimum quantity of food (0.4 mg/well), development was significantly reduced (t-test, pairwise comparison of treatments A and B, P < 0.05) and mortality was significantly greater (Kaplan–Meier pairwise comparison of treatments A and B: 16.7% vs. 87%, P < 0.001) in the presence of a fourth-instar larva than when kept alone. Even with more food, only about 20% of the younger larvae survived to the third instar in the presence of a conspecific fourth-instar larva, and there was no significant effect of the amount of food on their survival (Kaplan–Meier pairwise comparisons of treatments B, C and D, P > 0.05). In the presence of a fourth-instar larva, the quantity of food did not have a significant effect on development time of a younger larva (t-test, pairwise comparisons of treatments B, C and D, P > 0.05), although there was a trend towards faster development when more food was present.

Table 2. .Anopheles arabiensis larval survival percentage and mean ± SD time from first to third instar for individuals kept in a 6.5 ml well, with or without a conspecific fourth instar (L4), starting with different quantities of Tetramin® food/well.
    Development time
Treatment* FoodSurvivalmean±SD (days)
codeL4mg/wellL1 to L3from L1 to L3
  • *

    See text for significance of pairwise comparisons between results of each treatment.

A0.483.3%4.5±0.5 (n=20)
B+0.413.0%8.3±1.5 (n=3)
C+0.820.8%6.6±1.3 (n=5)
D+1.618.2%5.8±1.0 (n=4)

Larval faeces examination

Cadavers of An. arabiensis larvae that died in isolation were always found in the well, whereas cadavers could not be found for any of the younger larvae that disappeared in the presence of a fourth-instar larva. Therefore, we checked the faeces of 43 fourth-instar larvae in wells from which the younger larva had disappeared (57 cases). Among the samples of larval faeces examined, remnants of first-instar larvae were recovered in 11 out of 43 cases (26%). Fragments found in or attached to the faeces included abdominal hairs, mouthparts and the complete head capsule (Fig. 2).

image

Figure 2. Remnants of the heads of first instar stages recovered in faeces of fourth-instar stages of An. arabiensis; a–c, different individuals.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Assuming that first-instar larvae were eaten alive or after being killed by fourth instars, this study shows that older larvae of the An. gambiae complex are able to prey on younger larvae of conspecifics (cannibalism) as well as on larvae of closely related species (predation). We saw no indications that larval consumption involved scavenging of cadavers that had died from other causes. PCR assays confirmed the consumption of An. gambiae s.s. first instars by An. quadriannulatus fourth instars, although not every specimen showed the diagnostic rDNA band for the prey species. This insensitivity could be caused by rapid degeneration of the prey DNA in the predator's midgut due to its alkalinity with digestive enzyme activity (Dadd, 1975). It remains unclear why our PCR assay failed to detect An. quadriannulatus rDNA in the apparently predatory fourth-instar larvae of An. gambiae s.s. The amount of food available did not influence the occurrence of predation in An. gambiae s.s. or An. quadriannulatus, indicating that predation is a facultative process and thus not food dependent.

Compared to the low mortality rate (∼17%) of An. arabiensis larvae between first and third instar when kept in isolation, the much higher rates of disappearance (79–87%) of smaller instars reared in the presence of a conspecific fourth-instar larva indicated the frequent occurrence of cannibalism in An. arabiensis. This happened even under conditions with abundant food, indicating that cannibalism is not necessarily induced by food deprivation, as was the case with predation. The recovery of body parts of first-instar larvae in the faeces of An. arabiensis fourth-instar larvae confirmed our inference of cannibalism from the feeding experiments.

Also for An. arabiensis, the presence of a fourth-instar larva significantly prolonged the development of a smaller larva to the third instar. As this occurred with adequate food available, the retarded growth rate may be attributed to frequent interactions between larvae causing disproportionate stress for the younger instar. Such larval stress could both disturb feeding and increase energy consumption with deleterious effects on the rate of development. For the European woodland mosquito Aedes cantans Meigen, Renshaw et al. (1993) suggested that frequent contact between larvae may interrupt their feeding and could produce effects similar to those of food shortage per se.

Although larval densities of the An. gambiae complex used in these studies of cannibalism and predation are comparable with those used in similar studies on other anophelines (Reisen & Emory, 1976; Shoukry, 1980; Reisen et al., 1982), it remains questionable to what extent cannibalistic and predacious behaviour occur under more natural conditions. The experimental densities used in this study (roughly 0.53 larva/cm2 in the cannibalism study, 0.35 larva/cm2 in the predation study) are relatively high compared with densities usually found in natural habitats (e.g. 0.016 ± 0.018 larva/cm2, C.J.M. Koenraadt, unpublished data; 0.13 larva/cm2 in Gimnig et al., 2002). As the smaller instars had little chance of escape from attacks by the older instars within the small experimental containers, this would have induced more stress and risk of predation than under field conditions. However, larval densities in natural breeding sites are not homogeneously distributed and tend to be clustered (Service, 1971), so frequent contact between larvae within a site is not unlikely, resulting in opportunities for cannibalism and predation. Examining the midgut contents from field collected larvae might reveal whether or not cannibalism and predation occur among such anopheline populations under field conditions.

Studies on the population dynamics of the aquatic stages of An. gambiae s.l. have shown high mortality rates in natural breeding sites. Predation by nonanophelines and the presence of parasites and pathogens have been identified as major causes of larval mortalities up to 98% (Service, 1973, 1977), but cannibalism and predation may be additional regulatory mechanisms for larval populations. Although we would expect that cannibalism and predation are more likely to occur under conditions of nutritional stress, we hypothesize that they also occur under abundant food conditions with densities in a breeding site becoming higher as the habitat shrinks through drying out. Eliminating larvae of sibling species or conspecifics may facilitate faster development with less density-dependent competition for the available resources. Rapid development is a prerequisite for successful reproduction of An. gambiae s.l. because larval habitats may dry out within a few days.

In the context of optimal larval survival, it is important to note that An. gambiae s.l. females tend to avoid oviposition sites containing older instar larvae (McCrae, 1984). In this way, predation of offspring may be avoided. It remains unclear, however, which physical or chemical cues female An. gambiae s.l. use for selection of sites optimal for the development of their offspring (Takken & Knols, 1999; Blackwell & Johnson, 2000). Studies combining the interpretation of female oviposition preference and larval survival should investigate the trade-offs between such behavioural and ecological factors for each sibling species, considering that divergences of their oviposition sites seem to be important drivers of speciation within the An. gambiae complex (Coluzzi et al., 2002). Even so, sharing of breeding sites is a common feature among freshwater members of the An. gambiae complex in tropical Africa, notably for the two main malaria vectors An. arabiensis and An. gambiae s.s. (White & Rosen, 1973; Gimnig et al., 2001; Edillo et al., 2002). The laboratory study by Schneider et al. (2000) indicated that An. gambiae s.s. outcompetes An. arabiensis among mixed larval populations, which may help to explain some of the differential fluctuations observed in sibling species composition and temporal dynamics of sympatric adult populations. Results from this study showed that cannibalism and predation occur within and between members of the An. gambiae complex and therefore these phenomena may also influence the outcome of competitive interactions between these sibling species, affecting their resultant adult population densities.

Acknowledgements

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Dr Andrew Githeko for hospitality to CJMK during the field-work of this study in Kenya; we thank E. Fischer, S. Majambere and A. de Bakker for assisting with the experiments and N. Mulaya is acknowledged for the photomicrographic pictures. This paper is published with permission of the Director of the Kenya Medical Research Institute. Financial support was obtained from WOTRO WAA-912.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aniedu, I., Mutinga, M.J. & Mutero, C.M. (1993) Vertical estimates of survivorship of larvae and pupae of Anopheles gambiae Giles complex in Baringo District Kenya. Insect Science and its Application, 14, 3948.
  • Balfour, A. (1921) Mosquito breeding in saline waters. Bulletin of Entomological Research, 12, 2934.
  • Blackwell, A. & Johnson, S.N. (2000) Electrophysiological investigation of larval water and potential oviposition chemo-attractants for Anopheles gambiae s.s. Annals of Tropical Medicine and Parasitology, 94, 389398.
  • Charlwood, J.D., Kihonda, J., Sama, S. et al. (1995) The rise and fall of Anopheles arabiensis (Diptera: Culicidae) in a Tanzanian village. Bulletin of Entomological Research, 85, 3744.
  • Coluzzi, M., Sabatini, A., Della Torre, A., Di Deco, M.A. & Petrarca, V. (2002) A polytene chromosome analysis of the Anopheles gambiae species complex. Science, 298, 14151418.
  • Dadd, R.H. (1975) Alkalinity within the midgut of mosquito larvae with alkaline-active digestive enzymes. Journal of Insect Physiology, 21, 18471853.
  • Edillo, F.E., Toure, Y.T., Lanzaro, G.C., Dolo, G. & Taylor, C.E. (2002) Spatial and habitat distribution of Anopheles gambiae and Anopheles arabiensis (Diptera: Culicidae) in Banambani village, Mali. Journal of Medical Entomology, 39, 7077.
  • Gillies, M.T. & DeMeillon, B. (1968) The Anophelinae of Africa South of the Sahara. Publication no. 54. South African Institute for Medical Research, Johannesburg.
  • Gimnig, J.E., Ombok, M., Kamau, L. & Hawley, W.A. (2001) Characteristics of larval anopheline (Diptera: Culicidae) habitats in Western Kenya. Journal of Medical Entomology, 38, 282288.
  • Gimnig, J.E., Ombok, M., Otieno, S., Kaufman, M.G., Vulule, J.M. & Walker, E.D. (2002) Density-dependent development of Anopheles gambiae (Diptera: Culicidae) larvae in artificial habitats. Journal of Medical Entomology, 39, 162172.
  • Gordeev, M.I. & Troshkov, N.I. (1990) Inversion polymorphism of the malaria mosquito Anopheles messeae. IX. Cannibalism in larvae as a selection factor. Genetika, 26, 15971603.
  • Hinman, E.H. (1934) Predators of the Culicidae (mosquitoes). I – The predators of larvae and pupae, exclusive of fish. Journal of Tropical Medicine and Hygiene, 37, 129134.
  • Kleinbaum, D.G. (1996) Survival Analysis – a Self-Learning Text. Springer-Verlag, New York.
  • Lyimo, E.O., Takken, W. & Koella, J.C. (1992) Effect of rearing temperature and larval density on larval survival, age at pupation and adult size of Anopheles gambiae. Entomologia Experimentalis et Applicata, 63, 265271.
  • McCrae, A.W. (1984) Oviposition by African malaria vector mosquitoes. II. Effects of site tone, water type and conspecific immatures on target selection by freshwater Anopheles gambiae Giles, sensu lato. Annals of Tropical Medicine and Parasitology, 78, 307318.
  • Minakawa, N., Githure, J.I., Beier, J.C. & Yan, G. (2001) Anopheline mosquito survival strategies during the dry period in western Kenya. Journal of Medical Entomology, 38, 388392.
  • Reisen, W.K. & Emory, R.W. (1976) Cannibalism in Anopheles stephensi Liston. Mosquito News, 36, 198200.
  • Reisen, W.K., Azra, K. & Mahmood, F. (1982) Anopheles culicifacies (Diptera: Culicidae): horizontal and vertical estimates of immature development and survivorship in rural Punjab Province, Pakistan. Journal of Medical Entomology, 19, 413422.
  • Renshaw, M., Service, M.W. & Birley, M.H. (1993) Density-dependent regulation of Aedes cantans (Diptera: Culicidae) in natural and artificial populations. Ecological Entomology, 18, 223233.
  • Schneider, P., Takken, W. & McCall, P.J. (2000) Interspecific competition between sibling species larvae of Anopheles arabiensis and An. gambiae. Medical and Veterinary Entomology, 14, 165170.
  • Scott, J.A., Brogdon, W.G. & Collins, F.H. (1993) Identification of single specimens of the Anopheles gambiae complex by the polymerase chain reaction. American Journal of Tropical Medicine and Hygiene, 49, 520529.
  • Senior-White, R. (1926) Physical factors in mosquito ecology. Bulletin of Entomological Research, 16, 187248.
  • Service, M.W. (1971) Studies on sampling larval populations of the Anopheles gambiae complex. Bulletin of the World Health Organisation, 45, 169180.
  • Service, M.W. (1973) Mortalities of the larvae of the Anopheles gambiae Giles complex and detection of predators by the precipitin test. Bulletin of Entomological Research, 62, 359369.
  • Service, M.W. (1977) Mortalities of the immature stages of species B of the Anopheles gambiae complex in Kenya: comparison between rice fields and temporary pools, identification of predators, and effects of insecticidal spraying. Journal of Medical Entomology, 13, 535545.
  • Service, M.W. (1993) Mosquito Ecology. Field Sampling Methods, 2nd edn. Elsevier Applied Sciences, London.
  • Shoukry, A. (1980) Cannibalism in Anopheles pharoensis Theo. Experientia, 36, 308309.
  • Le Sueur, D. & Sharp, B.L. (1988) The breeding requirements of three members of the Anopheles gambiae Giles complex (Diptera: Culicidae) in the endemic malaria area of Natal, South Africa. Bulletin of Entomological Research, 78, 549560.
  • Takken, W. & Knols, B.G.J. (1999) Odor-mediated behavior of Afrotropical malaria mosquitoes. Annual Review of Entomology, 44, 131157.
  • White, G.B. & Rosen, P. (1973) Comparative studies on sibling species of the Anopheles gambiae Giles complex (Dipt., Culicidae). II. Ecology of species A and B in savanna around Kaduna, Nigeria, during transition from wet to dry season. Bulletin of Entomological Research, 62, 613625.