Unexpected behavioural adaptation of yellow fever mosquitoes in response to high temperatures.

Temperature is a major ecological driver of mosquito-borne diseases as it influences the life-history of both the mosquito and the pathogen harboured within it. Understanding the mosquitoes' thermal biology is essential to inform risk prediction models of such diseases. Mosquitoes can respond to temperatures by microhabitat selection through thermal preference. However, it has not yet been considered that mosquitoes are likely to adapt to changing temperatures, for example during climate change, and alter their preference over evolutionary time. We investigated this by rearing six cohorts of the yellow fever mosquito Aedes aegypti at two temperatures (24 °C, 30 °C) for 20 generations and used these cohorts to explicitly separate the effects of long-term evolution and within-generation acclimation on their thermal preferences in a thermal gradient of 20-35 °C. We found that warm-evolved mosquitoes spent 31.5% less time at high temperatures, which affects their efficiency as a vector. This study reveals the complex interplay of experimental evolution, rearing temperatures, and thermal preference in Ae. aegypti mosquitoes. It highlights the significance of incorporating mosquito microhabitat selection in disease transmission models, especially in the context of climate change.

Use of novel lab assays to examine the effect of pyrethroid-treated bed nets on blood-feeding success and longevity of highly insecticide-resistant Anopheles gambiae s.l. mosquitoes.

BACKGROUND: There is a pressing need to improve understanding of how insecticide resistance affects the functional performance of insecticide-treated nets (ITNs). Standard WHO insecticide resistance monitoring assays are designed for resistance surveillance and do not necessarily provide insight into how different frequencies, mechanisms or intensities of resistance affect the ability of ITNs to reduce malaria transmission. METHODS: The current study presents some novel laboratory-based assays that attempt to better simulate realistic exposure of mosquitoes to ITNs and to quantify impact of exposure not only on instantaneous mortality, but also on blood-feeding and longevity, two traits that are central to transmission. The assays evaluated the performance of a standard ITN (Permanet® 2.0; Vestergaard Frandsen), a 'next generation' combination ITN with a resistance-breaking synergist (Permanet® 3.0) and an untreated net (UTN), against field-derived Anopheles gambiae sensu lato mosquitoes from Côte d'Ivoire exhibiting a 1500-fold increase in pyrethroid resistance relative to a standard susceptible strain. RESULTS: The study revealed that the standard ITN induced negligible instantaneous mortality against the resistant mosquitoes, whereas the resistance-breaking net caused high mortality and a reduction in blood-feeding. However, both ITNs still impacted long-term survival relative to the UTN. The impact on longevity depended on feeding status, with blood-fed mosquitoes living longer than unfed mosquitoes following ITN exposure. Exposure to both ITNs also reduced the blood-feeding success, the time spent on the net and blood-feeding duration, relative to the untreated net. CONCLUSION: Although a standard ITN did not have as substantial instantaneous impact as the resistance-breaking net, it still had significant impacts on traits important for transmission. These results highlight the benefit of improved bioefficacy assays that allow for realistic exposure and consider sub- or pre-lethal effects to help assess the functional significance of insecticide resistance.

Exploring artificial habitat fragmentation to control invasion by infectious wildlife diseases.

One way to reduce the impacts of invading wildlife diseases is setting up fences that would reduce the spread of pathogens by limiting connectivity, similarly to exclusion fences that are commonly used to conserve threatened species against invasive predators. One of the problems with fences is that, while they may have the short-term benefit of impeding the spread of disease, this benefit may be offset by potential long-term ecological costs of fragmentation by fencing. However, managers facing situations where a pathogen has been detected near the habitat of a (highly) vulnerable species may be willing to explore such a trade-off. To aid such exploration quantitatively, we present a series of models trading off the benefits of fragmentation (potential reduction of disease impacts on susceptible individuals) against its costs (both financial and ecological, i.e. reduced viability in the patches created by fragmentation), and exploring the effects of fragmentation on non-target species richness. For all model variants we derive the optimal number of artificial patches. We show that pre-emptive disease fences may have benefits when the risk of disease exceeds the impacts of fragmentation, when fence failure rates are lower than a specific threshold, and when sufficient resources are available to implement optimal solutions. A useful step to initiate planning is to obtain information about the expected number of initial infection events and on the host's extinction threshold with respect to the focal habitat and management duration. Our approach can assist managers to identify whether the trade-offs support the decision to fence and how intensive fragmentation should be.

Parasite evolution in an age-structured population.

Although mortality increases with age in most organisms, senescence is missing from models of parasite evolution. Since virulence evolves according to the host's mortality, and since virulence influences the intensity of transmission, which determines the average age at infection and thus the mortality rate of a senescing host, we expected that epi-evolutionary feedbacks would underlie the evolution of virulence in a population of senescing hosts. We tested this idea by extending an age-structured model of epidemiological dynamics with the parasite's evolution. A straightforward prediction of our model is that stronger senescence forces the evolution of higher virulence. However, the model also reveals that the evolved virulence depends on the average age at infection, giving an evolutionary feedback with the epidemiological situation, a prediction not found when assuming a constant mortality rate with age. Additionally, and in contrast to most models of parasite evolution, we found that the virulence at the evolutionary equilibrium is influenced by whether the force of infection depends on the density or on the frequency of infected hosts, due to changes in the average age at infection. Our findings suggest that ignoring age-specific effects, and in particular senescence, can give misleading predictions about parasite evolution.

Sequential co-infections drive parasite competition and the outcome of infection.

Co-infections by multiple parasites are common in natural populations. Some of these are likely to be the result of sequential rather than simultaneous infections. The timing of the co-infections may affect their competitive interactions, thereby influencing the success of the parasites and their impact on the host. This may have important consequence for epidemiological and eco-evolutionary dynamics. We examined in two ecological conditions the effect of sequential co-infection on the outcome of infection by two microsporidians, Vavraia culicis and Edhazardia aedis, that infect the mosquito Aedes aegypti. The two parasites have different transmission strategies: V. culicis is transmitted horizontally either among larvae or from adults to larvae, while E. aedis can be transmitted horizontally among larvae or vertically from females to their eggs. We investigated how the timing and order of the co-infection and how the host's food availability affected the parasite's transmission potential (the percentage of individuals that harboured transmissible spores) and the host's juvenile survival, its age at emergence and its longevity. The outcome of co-infection was strongly affected by the order at which the parasites arrived. In co-infections, V. culicis had greater horizontal transmission if it arrived early, whereas the transmission potential of E. aedis, either vertical or horizontal, was not affected by the competitor V. culicis. The availability of food determined the duration of infection leading to variation in mortality and in the transmission potential. For both parasites low food decreased juvenile survival, delayed emergence to adulthood and increased horizontal transmission potential. High food increased juvenile survival and the probability of emergence with higher vertical transmission for E. aedis. Overall, our results suggest that early infection favours transmission and that (a) V. culicis plastically responded to co-infection, (b) E. aedis was not affected by co-infection but it was more susceptible to factors extending or decreasing the time it spent in the host (time of infection and food). Our results emphasize the complexity of the impact of co-infection on host-parasite interactions. In particular, the timing and order of sequential co-infections can result in different within-host dynamics and modify infection outcomes.

Consequences of larval competition and exposure to permethrin for the development of the rodent malaria Plasmodium berghei in the mosquito Anopheles gambiae.

BACKGROUND: Mosquitoes and other vectors are often exposed to sublethal doses of insecticides. Larvae can be exposed to the run-off of agricultural use, and adults can be irritated by insecticides used against them and move away before they have picked up a lethal dose. This sublethal exposure may affect the success of control of insect-borne diseases, for it may affect the competence of insects to transmit parasites, in particular if the insects are undernourished. METHODS: We assessed how exposure of larvae and adults to a sublethal dose of permethrin (a pyrethroid) and how larval competition for food affect several aspects of the vector competence of the mosquito Anopheles gambiae for the malaria parasite Plasmodium berghei. We infected mosquitoes with P. berghei and measured the longevity and the prevalence and intensity of infection to test for an effect of our treatments. RESULTS: Our general result was that the exposure to the insecticide helped mosquitoes deal with infection by malaria. Exposure of either larvae or adults decreased the likelihood that mosquitoes were infected by about 20%, but did not effect the parasite load. Exposure also increased the lifespan of infected mosquitoes, but only if they had been reared in competition. Larval competition had no effect on the prevalence of infection, but increased parasite load. These effects may be a consequence of the machinery governing oxidative stress, which underlies the responses of mosquitoes to insecticides, to food stress and to parasites. CONCLUSIONS: We conclude that insecticide residues are likely to affect the ability of mosquitoes to carry and transmit pathogens such as malaria, irrespective of the stage at which they are exposed to the insecticide. Our results stress the need for further studies to consider sublethal doses in the context of vector ecology and vector-borne disease epidemiology.

Larval exposure to a pyrethroid insecticide and competition for food modulate the melanisation and antibacterial responses of adult Anopheles gambiae.

The insecticides we use for agriculture and for vector control often arrive in water bodies, where mosquito larvae may be exposed to them. Not only will they then likely affect the development of the larvae, but their effects may carry over to the adults, potentially affecting their capacity at transmitting infectious diseases. Such an impact may be expected to be more severe when mosquitoes are undernourished. In this study, we investigated whether exposing larvae of the mosquito Anopheles gambiae to a sub-lethal dose of permethrin (a pyrethroid) and forcing them to compete for food would affect the immune response of the adults. We found that a low dose of permethrin increased the degree to which individually reared larvae melanised a negatively charged Sephadex bead and slowed the replication of injected Escherichia coli. However, if mosquitoes had been reared in groups of three (and thus had been forced to compete for food) permethrin had less impact on the efficacy of the immune responses. Our results show how larval stressors can affect the immune response of adults, and that the outcome of exposure to insecticides strongly depends on environmental conditions.

The ability of Anopheles gambiae mosquitoes to bite through a permethrin-treated net and the consequences for their fitness.

Insecticide-treated bed-nets (ITNs) control malaria by keeping mosquitoes from reaching people sleeping under a net and by killing mosquitoes. Most tests of ITNs consider their overall epidemiological outcome without considering the different behaviors underlying their effects. Here we consider one of these behaviors: that mosquitoes can bite through the net if its user is touching it. We assayed the ability of an insecticide-sensitive strain of the mosquito Anopheles gambiae to bite through a permethrin-treated or an untreated net, and their subsequent survival and fecundity. Despite the irritancy of permethrin, 71% of the mosquitoes took blood through the ITN (vs. 99% through the untreated net). The ITN reduced the time spent biting, the blood-meal size and the fecundity, and it killed about 15% of the mosquitoes within 24 hours of feeding (vs. 5% on the untreated net). However, the mosquito's survival was much higher than what we found in WHO cone assays, suggesting that the bloodmeal increased the mosquito's resistance to the insecticide. Thus, our results suggest that the irritancy and the toxicity of ITNs are reduced when mosquitoes contact and feed on their host, which will affect our understanding of the personal and community protection offered by the ITNs.

Malaria load affects the activity of mosquito salivary apyrase.

Mosquitoes infected by sporozoites, the infectious stage of malaria, bite more frequently than uninfected mosquitoes. One of the mechanisms underlying this behavioural change appears to be that the sporozoites decrease the activity of apyrase, an ADP-degrading enzyme that helps the mosquitoes to locate blood. Using the parasite Plasmodium berghei and the mosquito Anopheles gambiae, we confirmed that sporozoite infection alters the host-seeking behaviour of mosquitoes by making them more likely to refeed after a first blood meal, and that apyrase activity is one of the mechanisms of the increased biting persistence and motivation of infectious mosquitoes. We further showed that apyrase activity decreases as the sporozoite load increases, and that mosquitoes with lower apyrase activity take up less blood, making it more likely that they would return to top up their blood meal. Finally, by comparing full-sib families of mosquitoes, we showed that there was genetic variation for apyrase activity, but not for the resistance of parasites to be manipulated. Our results give new insights in understanding how malaria parasites change their hosts to affect their own transmission.

The interaction between permethrin exposure and malaria infection affects the host-seeking behaviour of mosquitoes.

BACKGROUND: Insecticide-treated bed nets (ITNs) help to control malaria by mechanically impeding the biting of mosquitoes, by repelling and irritating them and by killing them. In contrast to spatial repellency, irritancy implies that mosquitoes contact the ITN and are exposed to at least a sub-lethal dose of insecticide, which impedes their further blood-seeking. This would weaken the transmission of malaria, if mosquitoes are infectious. METHODS: It was therefore tested whether sub-lethal exposure to permethrin impedes blood-feeding differently in uninfected mosquitoes and in mosquitoes carrying the non-transmissible stage (oocysts) or the infectious stage (sporozoites) of the malaria parasite Plasmodium berghei. In addition, as the degree of irritancy determines the dose of insecticide the mosquitoes may receive, the irritancy to permethrin of infected and uninfected mosquitoes was compared. RESULTS: In this laboratory setting, sub-lethal exposure to permethrin inhibited the blood-seeking behaviour of Anopheles gambiae mosquitoes for almost 48 h. Although infection by malaria did not affect the irritancy of the mosquitoes to permethrin at either the developmental stage or the infectious stage, both stages of infection shortened the duration of inhibition of blood-seeking. CONCLUSIONS: The results suggest that the impact of ITNs may be weaker for malaria-infected than for uninfected mosquitoes. This will help to understand the global impact of ITNs on the transmission of malaria and gives a more complete picture of the effectiveness of that vector control measure.

The effect of parasite infection on the recombination rate of the mosquito Aedes aegypti.

Sexual reproduction and meiotic recombination generate new genetic combinations and may thereby help an individual infected by a parasite to protect its offspring from being infected. While this idea is often used to understand the evolutionary forces underlying the maintenance of sex and recombination, it also suggests that infected individuals should increase plastically their rate of recombination. We tested the latter idea with the mosquito Aedes aegypti and asked whether females infected by the microsporidian Vavraia culicis were more likely to have recombinant offspring than uninfected females. To measure the rate of recombination over a chromosome we analysed combinations of microsatellites on chromosome 3 in infected and uninfected females, in the (uninfected) males they copulated with and in their offspring. As predicted, the infected females were more likely to have recombinant offspring than the uninfected ones. These results show the ability of a female to diversify her offspring in response to parasitic infection by plastically increasing her recombination rate.

The relationship between size and longevity of the malaria vector Anopheles gambiae (s.s.) depends on the larval environment.

BACKGROUND: Understanding the variation in vector-borne disease transmission intensity across time and space relies on a thorough understanding of the impact of environmental factors on vectorial capacity traits of mosquito populations. This is driven primarily by variation in larval development and growth, with carryover effects influencing adult traits such as longevity and adult body size. The relationship between body size and longevity strongly affects the evolution of life histories and the epidemiology of vector-borne diseases. This relationship ranges from positive to negative but the reasons for this variability are not clear. Both traits depend on a number of environmental factors, but primarily on temperature as well as availability of nutritional resources. We therefore asked how the larval environment of the mosquito Anopheles gambiae Giles (sensu stricto) (Diptera: Culicidae) affects the relationship between body size and longevity. METHODS: We reared the larvae of An. gambiae individually at three temperatures (21, 25 and 29 °C) and two food levels (the standard and 50% of our laboratory diet) and measured adult size and longevity. We estimated the direct and indirect (via adult size) effects of food and temperature on longevity with a piecewise structural equation model (SEM). RESULTS: We confirmed the direct effects of food and temperature during larval development on body size, as wing length decreased with increasing temperature and decreasing food levels. While the overall relationship between size and longevity was weak, we measured striking differences among environments. At 25 °C there was no clear relationship between size and longevity; at 29 °C the association was negative with standard food but positive with low food; whereas at 21 °C it was positive with standard food but negative with low food. CONCLUSIONS: The larval environment influences the adult's fitness in complex ways with larger mosquitoes living longer in some environments but not in others. This confirmed our hypothesis that the relationship between size and longevity is not limited to a positive correlation. A better understanding of this relationship and its mechanisms may improve the modelling of the transmission of vector borne diseases, the evolution of life history traits, and the influence of vector control.

Malaria infection in mosquitoes decreases the personal protection offered by permethrin-treated bednets.

BACKGROUND: Insecticides targeting adult mosquitoes are the main way of controlling malaria. They work not only by killing mosquitoes, but also by repelling and irritating them. Indeed their repellent action gives valuable personal protection against biting mosquitoes. In the context of malaria control this personal protection is especially relevant when mosquitoes are infectious, whereas to protect the community we would prefer that the mosquitoes that are not yet infectious are killed (so, not repelled) by the insecticide. As the infectious stage of malaria parasites increases the motivation of mosquitoes to bite, we predicted that it would also change their behavioural response to insecticides. RESULTS: With two systems, a laboratory isolate of the rodent malaria Plasmodium berghei infecting Anopheles gambiae and several isolates of P. falciparum obtained from schoolchildren in Tanzania that infected Anopheles arabiensis, we found that mosquitoes harbouring the infectious stage (the sporozoites) of the parasite were less repelled by permethrin-treated nets than uninfected ones. CONCLUSIONS: Our results suggest that, at least in the laboratory, malaria infection decreases the personal protection offered by insecticide-treated nets at the stage where the personal protection is most valuable. Further studies must investigate whether these results hold true in the field and whether the less effective personal protection can be balanced by increased community protection.

Host genotype and environment affect the trade-off between horizontal and vertical transmission of the parasite Edhazardia aedis.

BACKGROUND: If a parasite is able to transmit horizontally or vertically, which transmission mode will it choose? We investigated how the growth conditions and the genotype of the mosquito Aedes aegypti affect the transmission mode of the parasite Edhazardia aedis. RESULTS: In poor conditions the parasites were more likely to be transmitted horizontally, whereas in favourable conditions they were more likely to be transmitted vertically. Unfavourable conditions delayed emergence, giving the parasite more time to produce its horizontally transmitted stage; in more favourable conditions mosquitoes have greater reproductive success, increasing the effectiveness of vertical transmission. In addition, the parasite's ability to transmit vertically was influenced by the genetic background of the host (i.e., its full-sib family), giving a genetic correlation between the host's life-history and which of the parasite's transmission mode it enables. In particular, genotypes with large bodies (and therefore high fecundity) were more likely to enable vertical transmission than genotypes with small bodies. This led to a trade-off among the host's families (which can be interpreted as a genetic correlation) for the parasite's transmission mode. CONCLUSIONS: Since horizontal transmission is linked to higher virulence than vertical transmission, the host's contribution to transmission mode has important consequences for the evolution of parasites with mixed-mode transmission.

The Role of the Environment in the Evolution of Tolerance and Resistance to a Pathogen.

Defense against parasites can be divided into resistance, which limits parasite burden, and tolerance, which reduces pathogenesis at a given parasite burden. Distinguishing between the two and understanding which defense is favored by evolution in different ecological settings are important, as they lead to fundamentally different evolutionary trajectories of host-parasite interactions. We let the mosquito Aedes aegypti evolve under different food levels and with either no parasite, a constant parasite, or a coevolving parasite (the microsporidian Vavraia culicis). We then tested tolerance and resistance of the evolved lines on a population level at the two food levels. Exposure to parasites during evolution increased resistance and tolerance, but there were no differences between the lines evolved with coevolving or constant parasites. Mosquitoes that had evolved with food restriction had higher resistance than those evolved with high food but similar tolerance. The mosquitoes that had restricted food when being tested had lower tolerance than those with normal food, but there was no difference in resistance. Our results emphasize the complexity and dependence on environmental conditions of the evolution and expression of resistance and tolerance and help to evaluate some of the predictions about the evolution of host defense against parasites.

Inoculating Anopheles gambiae Mosquitoes with Beads to Induce and Measure the Melanization Immune Response.

The stimulation of immune responses is a common tool in invertebrate studies to examine the efficacy and the mechanisms of immunity. This stimulation is based on the injection of non-pathogenic particles into insects, as the particles will be detected by the immune system and will induce the production of immune effectors. We focus here on the stimulation of the melanization response in the mosquito Anopheles gambiae. The melanization response results in the encapsulation of foreign particles and parasites with a dark layer of melanin. To stimulate this response, mosquitoes are inoculated with beads in the thoracic cavity using microcapillary glass tubes. Then, after 24 hr, the mosquitoes are dissected to retrieve the beads. The degree of melanization of the bead is measured using image analysis software. Beads do not have the pathogenic effects of parasites, or their capacity to evade or suppress the immune response. These injections are a way to measure immune efficacy and the impact of immune stimulations on other life history traits, such as fecundity or longevity. It is not exactly the same as directly studying host-parasite interactions, but it is an interesting tool to study immunity and its evolutionary ecology.

Effects of food variability on growth and reproduction of Aedes aegypti.

Despite a large body of knowledge about the evolution of life histories, we know little about how variable food availability during an individual's development affects its life history. We measured the effects of manipulating food levels during early and late larval development of the mosquito Aedes aegypti on its growth rate, life history and reproductive success. Switching from low to high food led to compensatory growth: individuals grew more rapidly during late larval development and emerged at a size close to that of mosquitoes consistently reared at high food. However, switching to high food had very little effect on longevity, and fecundity and reproductive success were considerably lower than in consistently well-fed mosquitoes. Changing from high to low food led to adults with similar size as in consistently badly nourished mosquitoes, but even lower fecundity and reproductive success. A rapid response of growth to changing resources can thus have unexpected effects in later life and in lifetime reproductive success. More generally, our study emphasizes the importance of varying developmental conditions for the evolutionary pressures underlying life-history evolution.

Larval environment influences vector competence of the malaria mosquito Anopheles gambiae.

Background: While environmental factors such as temperature can influence the vector competence of mosquitoes directly, for example by affecting the longevity of the mosquito and the development of the malaria parasite they may also have an indirect impact on the parasite's transmission. By influencing larval development, they may affect the adult traits that are important for the parasite's development and transmission. We studied the influence of two larval environmental factors, food availability and temperature, on the probability that mosquitoes infected with the malaria parasite survived to harbour sporozoites in their salivary glands. Materials and methods: Anopheles gambiae larvae were reared at 21ºC, 25ºC or 29ºC, and fed either a standard larval diet or half of it. Adults could blood feed on mice harbouring the infectious gametocytic stage of Plasmodium berghei ANKA transformed with green fluorescent protein (GFP). Survival was assessed every 24 hrs up to 21 days post infection, when surviving mosquitoes were dissected to check the salivary glands for sporozoites with a fluorescent microscope sensitive to GFP. Using a binomial GLM we analysed 'vector competence', i.e. if mosquitoes survived until dissection and harboured sporozoites in their salivary glands. Results: Vector competence dropped by about a third if we fed larvae half the standard food regime. The effect of temperature during the larval period depended strongly on the food regime. At low food, increasing temperature from 21ºC to 29ºC increased vector competence from about 0.18 to 0.48, whereas at standard food, vector competence dropped from about 0.67 at 21ºC to 0.56 at 29ºC. Conclusions: Thus, perceptions and models about the role of environmental change on the transmission of malaria should include how the environment changes adult life-history by influencing larval development.

An Epidemiological Model of the Effects of Insecticide-Treated Bed Nets on Malaria Transmission.

Insecticide-treated bed nets (ITNs) have become a central tool for malaria control because they provide personal and community-wide protection through their repellent and insecticidal properties. Here we propose a model that allows to assess the relative importance of those two effects in different epidemiological contexts and we show that these two levels of protection may oppose each other. On the one hand, repellency offers personal protection to the users of ITNs. The repellent action, however, is a two-edged sword, for it diverts infectious mosquitoes to non-users, thereby increasing their risk. Furthermore, with increasing ITN coverage, the personal protection effect of repellency decreases as mosquitoes are forced to perform multiple feeding attempts even on ITN users. On the other hand, the insecticidal property, which offers community-wide protection by killing mosquitoes, requires that mosquitoes contact the insecticide on the ITN and is thus counteracted by the repellency. Our model confirms that ITNs are an effective intervention method by reducing total malaria prevalence in the population, but that there is a conflict between personal protection, offered by repellency, and community-wide protection, which relies on the ITN's insecticidal properties. Crucially, the model suggests that weak repellency allows disease elimination at lower ITN coverage levels.

Larval and adult environmental temperatures influence the adult reproductive traits of Anopheles gambiae s.s.

BACKGROUND: Anopheles mosquito life-history parameters and population dynamics strongly influence malaria transmission, and environmental factors, particularly temperature, strongly affect these parameters. There are currently some studies on how temperature affects Anopheles gambiae s.s. survival but very few exist examining other life-history traits. We investigate here the effect of temperature on population dynamics parameters. METHODS: Anopheles gambiae s.s. immatures were reared individually at 23 ± 1 °C, 27 ± 1 °C, 31 ± 1 °C, and 35 ± 1 °C, and adults were held at their larval temperature or at one of the other temperatures. Larvae were checked every 24 h for development to the next stage and measured for size; wing length was measured as a proxy for adult size. Females were blood fed three times, and the number of females feeding and laying eggs was counted. The numbers of eggs and percentage of eggs hatched were recorded. RESULTS: Increasing temperatures during the larval stages resulted in significantly smaller larvae (p = 0.005) and smaller adults (p < 0.001). Adult temperature had no effect on the time to egg laying, and the larval temperature of adults only affected the incubation period of the first egg batch. Temperature influenced the time to hatching of eggs, as well as the time to development at every stage. The number of eggs laid was highest when adults were kept at 27 °C, and lowest at 31 °C, and higher adult temperatures decreased the proportion of eggs hatching after the second and third blood meal. Higher adult temperatures significantly decreased the probability of blood feeding, but the larval temperature of adults had no influence on the probability of taking a blood meal. Differences were observed between the first, second, and third blood meal in the times to egg laying and hatching, number of eggs laid, and probabilities of feeding and laying eggs. CONCLUSIONS: Our study shows that environmental temperature during the larval stages as well as during the adult stages affects Anopheles life-history parameters. Data on how temperature and other climatic factors affect vector life-history parameters are necessary to parameterise more reliably models predicting how global warming may influence malaria transmission.