Influence of Sugar Availability and Indoor Microclimate on Survival of ...

5 downloads 0 Views 476KB Size Report
of 14 d for mosquitoes fed blood alone, 29 d for sugar alone and 33 d for ... KEY WORDS Anopheles survival, indoor climate, sugar, blood, vectorial capacity.
BEHAVIOR, CHEMICAL ECOLOGY

Influence of Sugar Availability and Indoor Microclimate on Survival of Anopheles gambiae (Diptera: Culicidae) Under Semifield Conditions in Western Kenya BERNARD A. OKECH,1,2,3 LOUIS C. GOUAGNA,1 GERARD F. KILLEEN,1, 4 BART G. J. KNOLS,1 EPHANTUS W. KABIRU,3 JOHN C. BEIER,4 GUIYAN YAN,5 AND JOHN I. GITHURE1,2

J. Med. Entomol. 40(5): 657Ð663 (2003)

ABSTRACT The inßuence of indoor microclimate on survival of female Anopheles gambiae sensu stricto Giles (Diptera: culicidae) mosquitoes fed on different nutrition sources was evaluated in a semiÞeld experimental hut exposed to ambient climate in western Kenya. Cages of mosquitoes (n ⬇ 50 per cage) were placed in nine positions within the hut combining three different sides and three different heights. At each height and side, mosquitoes were offered either human blood (once every 2 d), glucose (6% wt:vol) or a combination of the two diets over three experiments so that each cage position received one diet source. The effect of diet on survival was signiÞcant with mean survival times of 14 d for mosquitoes fed blood alone, 29 d for sugar alone and 33 d for blood plus sugar. Sugar availability decreased the odds of mortality ⬇85% compared with the blood group. Micro heterogeneities of temperature but not relative humidity also inßuenced survival although to a much lesser extent. The side but not height within the hut at which mosquitoes were placed, inßuenced survival but could not be explained by either temperature or relative humidity differences. The potential inßuence of seemingly minor heterogeneities of indoor microclimate upon vector longevity and vectorial capacity may merit further investigation. Also, the availability of sugar was shown to be a potentially crucial determinant of vectorial capacity. Compared with blood alone, the availability of sugar served to increase survival potential of vectors beyond ages at which they are old enough to transmit malaria. KEY WORDS Anopheles survival, indoor climate, sugar, blood, vectorial capacity

THE LONGEVITY OF ADULT mosquitoes is dependent on both environmental and physiological factors. Temperature and humidity are important environmental physical factors that can directly or indirectly impact mosquito survival (Clements 1963). Physiological factors, such as energy reserve levels, are inherent characteristics that sustain the biological functioning of the insect. Anopheles gambiae s.s. is predominantly endophilic and anthropophagic, particularly in East Africa (White et al. 1972, Highton et al. 1979, Githeko et al. 1994, Bøgh et al. 1998). This important malaria vector spends most of its adult life resting within houses and members of this complex can even survive during adverse conditions by aestivating indoors for several months (Gillies and de Meillon 1968, Omer and Cloud1 International Centre of Insect Physiology and Ecology (ICIPE), P. O. Box 30772, Nairobi, Kenya. 2 Kenya Medical Research Institute, P. O. Box 54840, Nairobi, Kenya. 3 Department of Zoology, Kenyatta University, P. O. Box 43844, Nairobi, Kenya. 4 Department of Tropical Medicine, School of Public Health and Tropical Medicine, Tulane University, New Orleans, LA 70112. 5 Department of Biological Sciences, State University of New York, Buffalo, NY 14260.

sley-Thompson 1970). The choice of resting site by malaria vectors, over distances as little as a few centimeters, can be inßuenced by microclimate and even be modiÞed by parasite infection (Fialho and Schall 1995). Endophilic malaria vectors are not distributed randomly among houses within villages; rather they are aggregated and the bulk of mosquitoes may be found in a relatively small proportion of houses (Lindsay et al. 1995, Smith et al. 1995, Ribeiro et al. 1996, Woolhouse et al. 1997). Although this is clearly inßuenced by spatial relationships between houses, potential hosts and larval habitats, selection of the most favorable houses based upon microclimate, predator abundance or other factors also may play a role. African malaria vectors routinely aggregate at particular sites within houses such as dark crevices, the inside of the roof and under furniture (Gillies and de Meillon 1968). Although the dependence of vector survival upon temperature and humidity over broad temporal and spatial scales is well established (Ijumba et al. 1990, Martens et al. 1994, Craig et al. 1999), the inßuence of climatic heterogeneities over short distances and times remains vaguely understood. The extent of microclimatic heterogeneity within and between neighboring houses has never been studied in detail

0022-2585/03/0657Ð0663$04.00/0 䉷 2003 Entomological Society of America

658

JOURNAL OF MEDICAL ENTOMOLOGY

and its role in mosquito resting site choice and survival have never been quantiÞed. Other than microclimate, the availability of sugar and blood can considerably inßuence survival, feeding behavior, and malaria transmission potential (Nayar and Sauerman 1975a, 1975b, 1975c; Beier 1996; Gary and Foster 2001). All the important African malaria vectors are anautogenous and require at least one blood meal to complete each gonotrophic cycle and successfully lay eggs (Briegel and Ho´ rler 1993). Furthermore, blood can act as an energy source, and mosquitoes can readily survive on blood alone when sugar availability is limited (Nayar and Sauerman 1975a, 1975b, 1975c; Beier 1996; Gary and Foster 2001). Nevertheless, sugar plays an important role in mosquito physiology and is thought to be advantageous over a blood meal because it is immediately used to produce energy (Clements 1992, Foster 1995). The inßuence of diet on the survival of African malaria vectors remains to be elucidated in the Þeld and our understanding of the factors that determine the longevity of malaria vectors remains extremely limited. This study attempts to evaluate and quantify the magnitude indoor microclimate has on mosquito survival. The study was, therefore, designed to assess the longevity of mosquitoes as a function of microclimate and diet under semiÞeld conditions, in a grass thatch village hut in southwestern Kenya.

Materials and Methods Study Area. The study was conducted on the shore of Lake Victoria at ICIPE-Mbita Point Research and Training Centre in Suba District, southwestern Kenya. This is an area of stable, endemic malaria (Mutero et al. 1998) where An. gambiae Giles, An. arabiensis Patton, and An. funestus Giles all contribute to transmission (Mutero et al. 1998, Minakawa et al. 1999, Minakawa et al. 2001). The area has two rainy seasons beginning from March to July and October to December. The survival experiments were conducted during the months of September to October 2000 and January to April 2001. Another experiment was set up from October to November 2001 to examine the inßuence of prevailing weather on the indoor climatic conditions of the experimental hut compared with three other Þeld huts. Mosquitoes. An. gambiae mosquitoes, originally colonized from specimens collected at Njage, 70 km from Ifakara, southwest Tanzania in April 1996, were reared in an insectary according to standard rearing procedures, as previously described (Benedict 1997). Brießy, ⬇200 larvae per 20 ⫻ 15-cm tray were reared by feeding on 10% slurry of parts 2:1 of Tetramin Baby E Þsh food and yeast. Pupae were collected and kept in 30 ⫻ 30 ⫻ 30 meshed cages in which emerging adults had access to 6% glucose and water. Between 40 and 50 (the exact number was known for each cage), 2Ð3-d-old adult females were aspirated into each of nine 15 ⫻ 15 ⫻ 15-cm meshed cages for survival studies before the start of the experiments.

Vol. 40, no. 5

Survival Studies. After preliminary Þeld studies in huts near the research station, the study was moved to a semiÞeld environment to avoid confounding as a result of human activity. In this environment, a local village hut (wall dimensions 3.2 m ⫻ 2.8 m ⫻ 1.7 m) was constructed within a large mesh-walled screenhouse (dimensions: 11.4 m ⫻ 7.1 m ⫻ 4.2 m) exposed to ambient climatic conditions (Fig. 1). Within the house, each of the nine experimental cages was assigned to one combination of vertical heights (0.5, 1.5, and 2.5 m) and one of three sides of the house (walls facing approximately east, south, and west). Each cage was suspended from a pole or directly from the ceiling using steel wire, the Þrst centimeter of which was coated with Tanglefoot, a castor oil-based odorless formulation for protecting against crawling insects, ants in particular. The three cages at each level were randomly assigned to either blood, sugar, or a combination of the two diets so that for the whole study three experiments (three replicates) were performed, one for each diet at each cage position. The sugar-fed cages were allowed access to 6% (wt:vol) glucose solution continuously in a small cotton pad because in wild carbohydrate sources constantly available. Even though the wet cotton pads were expected to increase relative humidity in the sugar-fed cages than in the blood-fed cages, they did not (relative humidity in sugar-fed cages: 57.35%, in sugar plus blood cages: 65.58%, and in blood-fed cages: 69.0%). Because in nature mosquitoes would normally refeed every 2 d, if not more often (Briegel and Ho´ rler 1993, Beier 1996), the practicable option we had was to offer blood-fed cages blood every 2 d. We covered the mosquitoes with a black cloth, and offered a human forearm from the same individual for 15 min once every second day in the evening hours. Mosquitoes that did not engorge on Þrst blood meal were culled resulting in a reduction in the starting numbers for the blood group. For the mosquitoes fed blood, oviposition substrates in this experiment were not provided, although it was observed that eggs had been deposited on dry paper towels that were placed on the bottom of the cages. Nonetheless, the experiments were conducted during the dry season (see study area) when oviposition sites would normally be scarce and difÞcult to reach or completely absent. Therefore, oviposition data were not collected and the experiment proceeded with no provision for extra oviposition substrates. HOBO data loggers were placed inside each cage to record temperature and humidity at hourly intervals throughout the experiment. All experiments started at 1700 hours each day. Any dead mosquitoes were removed from the cages and counted at this time and, when necessary, blood meals were provided. The experiment proceeded until no mosquitoes remained alive. Modeling and Statistical Analysis. The mean survival times for mosquito groups on different dietary regimes, pooled over all positions, were calculated using the Kaplan-Meier survival analysis. The predictors of mosquito survival were estimated by a Cox proportional hazards regression model as a function of diet, mean daily temperature and humidity, height,

September 2003

OKECH ET AL.: INDOOR SURVIVAL OF Anopheles gambiae

659

Fig. 1. SemiÞeld experimental system and design. Cages shown as white boxes in broken lines are inside the hut at the three different heights (0.5, 1.5, and 2.5 m) and aligned in the middle of each of the three different sides of the house (A) on the east, south and west-facing walls and grass thatch roof (B).

and side of hut at which the mosquitoes were held. The proportion surviving was calculated as the proportion of mosquitoes surviving longer than any time, d, whereas the daily survival probability was calculated as the proportion of mosquitoes surviving at any given time, d. The effects of continuous variables, such as temperature and humidity, were tested in the Cox regression model directly and as transformations thereof because their relationship to mosquito survival is known to be nonlinear (Martens et al. 1994, Craig et al. 1999). The untransformed functions of temperature and humidity provided a better model Þt and were used in the Þnal model. To evaluate the effect that different dietary regimes could possibly have upon malaria transmission potential, we estimated the minimum sporogonic incubation

period of Plasmodium falciparum in the semi Þeld environment, that was calculated as 14 d using the overall mean temperature of all cages in all experiments, and standard models based on direct laboratory measurements (Craig et al. 1999). Results Initially, the effects of diet, height, and side of hut were tested by Cox regression to see whether any of these factors inßuenced survival within the hut. The diet (P ⬍ 0.0001) offered to mosquitoes as well as the side of hut (P ⬍ 0.0001) at which they were placed signiÞcantly affected their survival probability. Height (P ⫽ 0.155) of the cage within house did not inßuence mosquito survival. The effect of diet on survival was clear with mean survival times of 14 d for mosquitoes

660

JOURNAL OF MEDICAL ENTOMOLOGY

Table 1. Cox regression analysis of the combined influence of environmental and dietary variables upon survival of mosquitoes Variable



Dieta 0b Blood aloneb Sugar ⫺1.904 Sugar ⫹ Blood ⫺2.496 Side of huta East side of hut 0.245 0b South side of hutb West side of hut 0.504 Height of habitat NS Temperature 0.222 Relative humidity NS

Standard DF Probability Exp, (␤) error 0b 0.092 0.102 0.090 0b 0.091 NA 0.027 NA

2 NAb 1 1 2 1 NAb 1 2 1 1

0.0001 NAb 0.0001c 0.0001c 0.0001 0.006c NAb 0.0001c 0.827 0.0001 0.412

NA NAb 0.149 0.082 NA 1.277 NAb 1.655 NA 1.248 NA

Categorical variables are represented by b and reference group is represented by a. SigniÞcance in relation to the reference is represented by c. ␤ ⫽ estimatable regression coefÞcient, in which a negative ␤ value indicates increased survival compared with the reference group. Exponential of beta, (Exp ␤), is a value by which the odds of mortality changes per unit increase in covariate factors: Exp(␤) ⬎ 1, odds of mortality increase, ⬍ 1, odds of mortality decrease. NA ⫽ not applicable, NS ⫽ not signiÞcant at P ⫽ 0.05 level.

fed blood alone, 29 d for sugar alone, and 33 d for blood plus sugar. The availability of sugar to the mosquitoes decreased the odds of mortality (calculated from exponential of the Beta covariate coefÞcient) by more than ⬇85% compared with the mosquitoes fed blood alone (Table 1). Although the height of cages within the hut had no inßuence upon survival, the modest although clear effect of the side at which mosquitoes are placed demonstrates that there is considerable heterogeneity in the suitability of different sites within a single house for mosquito survival. In a related study to compare the climatic conditions of our semiÞeld hut versus three Þeld houses, performed over a 1-wk period, there were no significant differences between mean temperatures and mean relative humidityÕs between the semiÞeld hut and the Þeld houses [analysis of variance (ANOVA): F ⫽ 0.889; df ⫽ 3; P ⫽ 0.455 and F ⫽ 2.521; df ⫽ 3; P ⫽ 0.072, respectively]. However, the indoor climatic conditions for the experimental hut was cooler and less humid compared with the Þeld houses. In the experimental hut, the temperature ranged between 14.1 and 28.1⬚C over all experiments, and mean daily temperature varied with height but not side of house (ANOVA: F ⫽ 11.06; df ⫽ 2; P ⬍ 0.0001 and F ⫽ 0.079; df ⫽ 2; P ⫽ 0.924, respectively). Temperature increased slightly but consistently with height (23.50 ⫾ 0.07, 23.98 ⫾ 0.08, and 24.04 ⫾ 0.09⬚C for 0.5, 1.5, and 2.5 m, respectively). Mean daily relative humidity, which ranged between 23.3 and 90.6% (Fig. 1), varied with height and side of house (ANOVA: F ⫽ 28.75; df ⫽ 2; P ⬍ 0.001 and F ⫽ 40.14; df ⫽ 2; P ⬍ 0.001, respectively). Mean daily relative humidity decreased with height (69.2 ⫾ 0.6, 64.3 ⫾ 0.9, and 58.1 ⫾ 1.4% for 0.5, 1.5, and 2.5m, respectively) and was 57.15 ⫾ 1.1, 70.28 ⫾ 0.6, and 63.02 ⫾ 1.2% for east, south, and west facing sides, respectively. Upon examining these trends, we hypothesized that such micro heterogeneities of climate, humidity in particular,

Vol. 40, no. 5

might be responsible for the differences in survival of mosquitoes in different sides of the house. To test whether microheterogeneities of temperature was responsible for the different survival probabilities on different sides of the hut, we tested a model incorporating untransformed functions of temperature and humidity to see whether they are potential determinants of mosquito survival. All signiÞcant terms from the Þrst Cox regression were reselected in the new model with minor quantitative changes in their regression coefÞcients (Table 1). Despite the association of both relative humidity and daily survival with the side of the house at which mosquitoes were placed, relative humidity proved not to be a signiÞcant determinant of survival probability over the range observed here (Table 1). By including both temperature and side of house in the same regression model, these factors failed to negate the signiÞcance of either one, resulting in parameter estimates of both terms that are essentially the same as when separately Þtted. We therefore concluded that the differential survival of mosquitoes at each side of the hut was independent of temperature and humidity and was not caused by microheterogeneities in these microclimatic factors. Using the Þnal model, we examined the inßuence of diet on survival over the entire lifespan of female mosquitoes. It was clear that sugar, either alone or in combination with blood, confers much greater longevity than blood alone (Fig. 2A). Furthermore, by examining the daily survival probabilities of mosquitoes over their entire lifespan (Fig. 2B), it was clear that mosquitoes fed blood alone not only die faster when they are young but also that they senesce faster as they age (Fig. 2A). We earlier estimated the minimum sporogonic incubation period of P. falciparum to be just over 14 d under the climatic conditions recorded in the experimental hut with an overall mean temperature of 23.8⬚C using the model of Craig et al. (1999). Thus, the availability of sugar under these experimental conditions served to increase the survival of female vectors to ages ⬎14 d, during which time they can transmit malaria. Discussion In this study we have demonstrated that the availability of sugar and, to a lesser extent, the microhabitat within houses in which mosquitoes rest, can substantially inßuence their longevity and thus the potential to transmit malaria. An important role for sugar feeding in vector survival and possibly malaria transmission was suggested by these studies, indicating that sugar feeding behavior and the roles of natural carbohydrate sources warrant much more extensive investigation. This represents the Þrst demonstration that resting in different parts of a house can inßuence a mosquitoÕs chance of survival and that factors other than temperature and humidity may be involved. Although the availability of sugar clearly increases the longevity of female An. gambiae and their potential to transmit malaria under these conditions, the impact

September 2003

OKECH ET AL.: INDOOR SURVIVAL OF Anopheles gambiae

Fig. 2. Proportion surviving (A) and daily survival probability (B) of female Anopheles gambiae fed on blood alone (solid line), sugar alone (dotted line), or a combination of sugar and blood (long dash). Senescence effect is clearly demonstrated in the daily survival probability curve and the raggedness of the curves reßects daily compensations for temperature ßuctuations.

of sugar observed in this study may be somewhat exaggerated because we allowed access to blood only on every second day whereas vectors may feed more often upon blood if given the opportunity (Briegel and Ho´ rler 1993, Beier 1996, Straif and Beier 1996, Gary and Foster 2001). Thus, the impact of feeding on blood only may have been underestimated. Although this helped to demonstrate clearly the role of sugar as a supplement to blood and is consistent with other studies in which daily access to blood was allowed (Gary and Foster 2001), other researchers have shown that An. gambiae can survive almost as long on blood alone (Straif and Beier 1996), because they can increase the number of blood meals they take to make up for this energetic shortfall (Briegel and Ho´ rler 1993).

661

In terms of malaria vectorial capacity, it has been suggested that the shorter lifespan of An. gambiae feeding on blood alone is more than made up for by increased feeding frequency (Gary and Foster 2001) and actually results in increased transmission potential. Although we did not measure blood-feeding frequency, an increased activity in blood feeding such as doubling of blood feeding in the absence of sugar in the diet would still result in a substantially reduced transmission potential based on the survival potentials we have estimated. Also, wild mosquitoes taking extra blood meals are not provided with them within the safety of a cage so the cumulative hazards associated with host seeking and blood feeding may result in even shorter lifespans in natural populations of vectors feeding predominantly on blood alone compared with those reported herein. However, this apparent conßict may be resolved by examining the way in which survival was quantiÞed and modeled. Previous studies were based on traditional models of mosquito survival in which the daily survival probability of female vectors does not change as they age. Here we have demonstrated that not only is senescence substantial over normal age ranges of wild mosquito populations, but also that it is accelerated considerably by sugar starvation. Given that it is the oldest mosquitoes that transmit malaria and that failure to consider senescence can cause severe biases in the estimates of vectorial capacity (Clements and Paterson 1981), it is perhaps understandable that our estimates of survival in older age groups are more sensitive to sugar deprivation. Overall, our results suggest that malaria transmission may be substantially enhanced by the availability of sugar sources to wild vector populations and that, in addition to host and larval habitats, this may be a crucial and, as yet understudied resource for malaria vector populations. We have demonstrated that very small differences in temperature but not humidity within a hut can inßuence mosquito survival in an area such as Mbita Point, western Kenya. This does not mean that relative humidity cannot limit vector survival and malaria transmission, as reported elsewhere (Ijumba et al. 1990, Craig et al. 1999), but rather that the wide but high range of humidity observed on the shores of Lake Victoria are all equally hospitable to mosquitoes. The extreme temperature sensitivity of An. gambiae females under these semiÞeld conditions contrasts sharply with laboratory based studies. In our semiÞeld experimental hut, mosquitoes are clearly much more sensitive to small temperature increases than those maintained in laboratory incubators (Martens 1994, Craig et al. 1999). An interesting observation is that some factor or factors other than temperature and humidity inßuences the chances of mosquito survival, depending upon which side of the hut they were placed in. Although some obvious possibilities such as airßow or light intensity heterogeneities can be considered, it remains to be seen what precisely makes such a clear difference between sites within such a small, enclosed space.

662

JOURNAL OF MEDICAL ENTOMOLOGY

The results of this study suggest some important questions that may form the basis of further investigations into somewhat overlooked determinants of mosquito behavior, life histories and transmission potential. Although the importance of larval habitat and blood meal host availability is well established, if somewhat poorly understood (Killeen et al. 2000, Killeen et al. 2001), the role of sugar sources remains essentially unexplored. Some studies have established a clear relation between nectar feeding and the availability of nectar sources for Aedes aegypti (MartinezIbarra et al. 1997). We suggest that increased efforts should be focused on establishing which sugar sources are most commonly used by wild malaria vectors and how their availabilities inßuence mosquito population dynamics and malaria transmission potential. Additionally, the potential inßuence of seemingly minor heterogeneities of indoor microclimate may merit further investigation. Furthermore, uncaged vectors may select particular locations within houses that favor or inhibit parasite development and may even constitute good targets for mosquito control measures such as indoor spraying, or traps. Acknowledgments We thank Woodbridge Foster (Ohio State University, Columbus, OH) for insightful suggestions and comments on the original draft of the manuscript, Daniel Impoinvil for useful discussions, and Elizabeth Walczak for statistical review. We also thank Basilio Njiru, Hassan Akello, Jackton Arija, and Enock Aloo for their technical assistance. Many thanks to all other ICIPE-Mbita staff for providing logistical support. This study was supported by National Institutes of Health (NIH) grants 1U19 (AI) 45511, 1D43TWO1143, and 5D43 TW 00920. B.A.O. was supported by the International Centre of Insect Physiology and Ecology (ICIPE), and African Regional Postgraduate Programme in Insect Science (ARPPIS) training fellowship.

References Cited Beier, J. C. 1996. Frequent blood-feeding and restrictive sugar-feeding behavior enhance the malaria vector potential of Anopheles gambiae s. l. and An. funestus (Diptera: Culicidae) in western Kenya. J. Med. Entomol. 33: 613Ð 618. Benedict, M. Q. 1997. Care and maintenance of anopheline mosquito colonies, pp. 3Ð12. In J. M. Crampton, C. B. Beard and C. Louis [eds.], Molecular biology of insect disease vectors: a methods manual. Chapman & Hall, New York. Bøgh, C., E. M. Pedersen, D. A. Mukoko, and J. H. Ouma. 1998. Permethrin-impregnated bed net effects on resting and feeding behaviour of lymphatic Þlariasis vector mosquitoes in Kenya. Med. Vet. Entomol. 12: 52Ð59. Briegel, H., and E. Ho´ rler. 1993. Multiple blood meals as a reproductive strategy in Anopheles (Diptera: culicidae). J. Med. Entomol. 30: 975Ð985. Clements, A. N. 1963. The physiology of mosquitoes. Pergamon Press, New York. Clements, A. N., and G. D. Paterson. 1981. The analysis of mortality and survival rates in wild populations of mosquitoes. J. Appl. Ecol. 18: 373Ð399.

Vol. 40, no. 5

Clements, A. N. 1992. Development, nutrition and reproduction, the biology of mosquitoes. Chapman & Hall, New York. Craig, M. H., R. W. Snow, and D. le Sueur. 1999. A climatebased distribution model of malaria transmission in subSaharan Africa. Parasitol. Today 15: 105Ð111. Fialho, R. F., and J. J. Schall. 1995. Thermal ecology of a malaria parasite and its insect vector: consequences for the parasiteÕs transmission success. J. Anim. Ecol. 64: 553Ð 562. Foster, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev. Entomol. 40: 443Ð 474. Gary, R.E.J., and W. A. Foster. 2001. Effects of available sugar on the reproductive Þtness and vectorial capacity of malaria vector Anopheles gambiae (Diptera: Culicidae). J. Med. Entomol. 38: 22Ð28. Gillies, M. T., and B. de Meillon. 1968. The Anophelinae of Africa, south of the Sahara (Ethiopian zoogeographical Region). South African Institute of Medical Research, Johannesburg, South Africa. Githeko, A. K., M. W. Service, C. M. Mbogo, F. K. Atieli, and F. O. Juma. 1994. Origin of blood meals in indoor and outdoor resting malaria vectors in western Kenya. Acta Tropica 58: 307Ð316. Highton, R. B., J. H. Bryan, P.F.L. Boreham, and J. A. Chandler. 1979. Studies on the sibling species Anopheles gambiae Giles and Anopheles arabiensis Patton (Diptera: Culicidae) in the Kisumu area, Kenya. Bull. Entomol. Res. 69: 43Ð53. Ijumba, J. N., R. W. Mwangi, and J. C. Beier. 1990. Malaria transmission potential of Anopheles mosquitoes in the Mwea- Tebere irrigation scheme, Kenya. Med. Vet. Entomol. 4: 425Ð32. Killeen, G. F., F. E. McKenzie, B. D. Foy, C. Schieffelin, P. F. Billingsley, and J. C. Beier. 2000. A simpliÞed model for predicting malaria entomologic inoculation rates based on entomologic and parasitologic parameters relevant to control. Am. J. Trop. Med. Hyg. 62: 535Ð544. Killeen, G. F., F. E. McKenzie, B. D. Foy, C. Bøgh, and J. C. Beier. 2001. The availability of potential hosts as a determinant of feeding behaviours and malaria transmission by mosquito populations. Trans. R. Soc. Trop. Med. Hyg. 95: 469 Ð 476. Lindsay, S. W., J.R.M. Armstrong Schellenberg, H. A. Zeiler, R. J. Daly, F. M. Salum, and H. A. Wilkins. 1995. Exposure of Gambian children to Anopheles gambiae vectors in an irrigated rice production area. Med. Vet. Entomol. 9: 50 Ð58. Martens, W., J. Rotmans, and L. W. Niessen. 1994. Climate change and malaria risk: an integrated modelling approach, pp. 1Ð33. In Global Dynamics and Sustainable Development Programme: GLOBO report series no. 3. National Institute of Public Health and Environmental Protection, Bilthoven, The Netherlands. Martinez-Ibarra, J. A., M. H. Rodriguez, J. I. ArredondoJimenez, and B. Yuval. 1997. Inßuence of plant abundance on nectar feeding by Aedes aegypti (Diptera: Culicidae) in southern Mexico. J. Med. Entomol. 34: 589 Ð93. Minakawa, N., J. I. Githure, J. C. Beier, and G. Yan. 2001. Anopheline mosquito survival strategies during the dry period in western Kenya. J. Med. Entomol. 38: 388 Ð92. Minakawa, N., C. M. Mutero, J. I. Githure, J. C. Beier, and G. Yan. 1999. Spatial distribution and habitat characterization of anopheline mosquito larvae in Western Kenya. Am. J. Trop. Med. Hyg. 61: 1010 Ð1016. Mutero, C. M., J. H. Ouma, B. K. Agak, J. A. Wanderi, and R. S. Copeland. 1998. Malaria prevalence and use of self-pro-

September 2003

OKECH ET AL.: INDOOR SURVIVAL OF Anopheles gambiae

tection measures against mosquitoes in Suba District, Kenya. East Afr. Med. J. 75: 11Ð15. Nayar, J. K., and D.M.J. Sauerman. 1975a. The effects of nutrition on survival and fecundity in Florida mosquitoes. 1. Utilization of sugar for survival. J. Med. Entomol. 12: 92Ð98. Nayar, J. K., and D.M.J. Sauerman. 1975b. The effects of nutrition on survival and fecundity in Florida mosquitoes. 2. The utilization of a blood meal for survival. J. Med. Entomol. 12: 99 Ð103. Nayar, J. K., and D.M.J. Sauerman. 1975c. The effects of nutrition on survival and fecundity in Florida mosquitoes. 3. The utilization of blood and sugar for fecundity. J. Med. Entomol. 12: 220 Ð225. Omer, S. M., and J. L. Cloudsley-Thompson. 1970. Survival of female Anopheles gambiae Giles through nine month dry season in Sudan. Bull. W.H.O. 42: 319 Ð330. Ribeiro, J.M.C., F. Seulu, T. Abose, G. Kidane, and A. Teklehaimanot. 1996. Temporal and spatial distribution of anopheline mosquitoes in an Ethiopian village: implications for malaria control strategies. Bull. W.H.O. 74: 299 Ð 305.

663

Smith, T., J. D. Charlwood, W. Takken, M. Tanner, and D. J. Spiegelhalter. 1995. Mapping densities of malaria vectors within a single village. Acta Tropica 59: 1Ð18. Straif, S. C., and J. C. Beier. 1996. Effects of sugar availability on the blood-feeding behavior of Anopheles gambiae (Diptera:Culicidae). J. Med. Entomol. 33: 608 Ð12. White, G. B., S. A. Magayuka, and P.F.L. Boreham. 1972. Comparative studies on sibling species of the Anopheles gambiae Giles complex (Diptera: Culicidae): bionomics and vectorial activity of species A and species B at Segera, Tanzania. Bull. Entomol. Res. 62: 295Ð317. Woolhouse, M.E.J., C. Dye, J. F. Etard, T. Smith, J. D. Charlwood, G. P. Garnett, P. Hagan, J.L.K. Hii, P. D. Ndhlovu, R. J. Quinnell, C. H. Watts, S. K. Chaniawana, and R. M. Anderson. 1997. Heterogeneities in the transmission of infectious agents: implications for the design of control programs. Proc. Natl. Acad. Sci. U.S.A. 94: 338 Ð 342. Received for publication 24 April 2002; accepted 20 March 2003.