Opinion

TRENDS in Parasitology

Vol.22 No.1 January 2006

Can mosquitoes help to unravel the community structure of Plasmodium species? Christophe Boe¨te1,2,3 and Richard E.L. Paul4,5 1

Institut de Recherche pour le De´veloppement, Laboratoire Ge´ne´tique et Evolution des Maladies Infectieuses, 911 avenue Agropolis, BP 64501, 34394 Montpellier Cedex 05, France 2 Laboratory of Entomology, Wageningen University, PO Box 8031, Binnenhaven 7, 6700 EH Wageningen, The Netherlands 3 Joint Malaria Programme, Kilimanjaro Christian Medical Centre, PO Box 2229, Moshi, Kilimanjaro, Tanzania 4 Laboratoire d’Entomologie Me´dicale, Institut Pasteur de Dakar, 36 Avenue Pasteur, BP 220, Dakar, Se´ne´gal 5 Biochimie et Biologie Mole´culaire des Insectes, Institut Pasteur, CNRS-FRE 2849, 25 rue de Dr Roux, 75724 Paris cedex 15, France

There has been a recent revival in attempts to understand changes in patterns of abundance of Plasmodium spp. that infect humans. This has been driven by the purportedly beneficial effects of co-infection on clinical pathology and the recognition of Plasmodium vivax as a public health problem in its own right. In contrast to the attention given to mixed-species infections in humans, parasite infections and interactions within the mosquito vector remain poorly documented, even though the distribution of vector-borne parasites such as Plasmodium spp. depends on vector–vertebrate and, crucially, vector–parasite interactions. To understand malaria epidemiology and to design appropriate control measures, this gap must be readdressed.

Focus on mixed-species infections A series of recent articles in Trends in Parasitology was dedicated to the neglected problem of Plasmodium vivax and it reignited the debate about the importance of mixedspecies infections in the epidemiology of malaria [1–4]. Most of the reviews in the series concern Asia, where P. vivax is particularly abundant, and they pinpoint particular clinical and biological outcomes of mixedspecies infections. The theme underpinning such interest in mixed-species infections is that, because mixed infections are beneficial in terms of reducing Plasmodium falciparum disease severity [5], a control method that disproportionally affects other Plasmodium species could cause disturbing, unwanted outcomes for the burden of P. falciparum malaria. However, predicting how specific control methods differentially affect Plasmodium species is not always obvious. Indeed, during the 1950s and 1960s – when vector control, rather than drug treatment, strategies were all the rage – P. falciparum was replaced by Plasmodium malariae in some areas of Tanzania. Such a change in species composition was attributed to the efficacy of vector-control programs and, potentially, the longer duration of human infection by P. malariae [6]. This Corresponding author: Boe¨te, C. ([email protected]). Available online 22 November 2005

conclusion was unsubstantiated and by no means obvious. Estimating the consequence of such species-dependent control outcomes for Plasmodium species composition requires knowledge of how an intervention would differentially affect each species when considered alone and of the epidemiological consequences of co-infection in uncontrolled situations. Without such information, it would be impossible to generate predictions and to draw conclusions about Plasmodium spp. population dynamics when control methods are employed. Current views about the impact of co-infection on parasite epidemiology are based almost entirely on data regarding parasitic infections of humans. Parasite species spend the majority of their lifecycles within the human host and, thus, have greater opportunity to interact within humans than within mosquitoes. However, as clearly parameterized in the classic Ross–Macdonald model of malaria [7–26] (Box 1), the distribution of vector-borne parasites depends on vector–vertebrate and vector– parasite interactions, in addition to interactions within vertebrates. Given that vector-based (as opposed to therapy-based) interventions represent the mainstay of prevention strategies, it seems essential to consider initially species differences that relate to vectorial parameters and, thus, the consequences of vector-based interventions for species distribution. Subsequently, the consequences of putative interactions of parasite species that impinge on the vector-related parameters of the classic model should be considered. To highlight the necessity for considering vector-related parameters in more detail, we present a worked example of a recent case of change in parasite species composition following intervention. The Brazilian situation In the Amazon, P. vivax and P. falciparum are sympatric and their dynamics are not fully understood. In the 1930s, they were estimated to represent 40% (P. vivax) and 60% (P. falciparum) of the human malaria cases in Para´ state (Brazil) [27], and similar proportions were observed in the 1980s in Brazil [28]. In a recent article by Po´voa et al. [28],

www.sciencedirect.com 1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.pt.2005.11.007

Opinion

22

TRENDS in Parasitology

Vol.22 No.1 January 2006

Box 1. Co-infections and the basic reproductive ratio R0 The mathematical modelling of malaria that was developed by Ross [7] and Macdonald [8] provided a quantitative framework within which to consider parasite transmission and was based on the classical ecological measure of fitness: the basic reproductive number R0, which represents the number of secondary cases of malaria arising from a single case in a fully susceptible population. For malaria, in which there are both vector and vertebrate hosts, R0 can be defined as follows (Equation I), involving parameters concerning the mosquito, the vertebrate host and transmission between these two hosts: R0 Z

ma2 bc KmT e rm

The number of mosquitoes per human host (m) is important when two species of vectors co-exist that have different susceptibility to two parasite species. The mosquito mortality rate (m), biting rate (a) and the duration of the sporogonic rate of development in the mosquito (T) might alter when co-species infections occur within the mosquito. There is conflicting evidence concerning the effect of infection on m [11,12] and a [13–15]. Although there is some suggestion that Plasmodium–microfilaria co-infections increase m, reduce T [16] and affect parasite development [17], no studies to date have addressed the impact of mixed Plasmodium species infections on these mosquito parameters.

(I)

where m is the number of mosquitoes per human host, a is the biting rate of the mosquitoes, b is the proportion of infective bites that leads to an infection in humans, c is the proportion of mosquito bites on infected humans that results in an infected mosquito, m is the mosquito mortality rate, T is the duration (in days) of sporogonic development in the vector and r is the rate of human recovery from infection. In a population in which two malaria parasite species coexist, the R0 value of each species might be influenced by the presence of other species in a variety of ways.

The vertebrate host parameter r If there is cross-species immunity, as is thought to be the case for Plasmodium vivax and Plasmodium falciparum [9], then r must be modified according to the prevalence rates of both species, in much the same way as when immunity that is developing to a single species is incorporated into a modified version of the Ross–Macdonald equation [10].

Anopheles darlingi is described as re-emerging in Brazil in Belem and Para´, leading to an increase in the number of malaria cases, all species confounded. It is striking that this increase in malaria incidence is due to P. vivax (and, to a much lesser extent, P. malariae), whereas the incidence of P. falciparum is decreasing. In addition, Po´voa et al. observed a modification of vector species composition. At first sight, the focus on case treatment rather than vector control by a World Bank (http://www.worldbank. org/) project [29] in the 1990s might explain the reduction in the number of P. falciparum malaria cases in humans but it does not intuitively provide a reason for the increase in numbers of cases of P. vivax malaria. Treatment of P. falciparum malaria is often followed by a clinical case of P. vivax malaria, as seen in Thailand [30], potentially as a result of the release of P. vivax from suppression by P. falciparum. This could arise either through recrudescence of a previously undetectable P. vivax blood-stage infection or through a relapse with renewed blood infections from latent hypnozoite-stage parasites in the liver. Thus, it could account in part for the increase in P. vivax incidence, especially because mixed infections would have been treated as though they were infections with P. falciparum alone [29] (i.e. the P. vivax infections were ‘cryptic’). However, even if mixed infections in humans were underestimated, as shown recently [31,32], their proportions would remain extremely low [28]. Moreover, if P. vivax co-infections were to reduce P. falciparum infection pathology, as recently suggested [3,5], treatment would inadvertently focus on www.sciencedirect.com

The mosquito parameters m, m, a and T

Transmission parameters b and c The proportion of infective bites that leads to an infection in humans (b) can be influenced by the presence of another plasmodial species but this potential effect remains poorly documented. The proportion of bloodmeals from infected individuals that infects the mosquito (c) depends on: (i) the effect of co-species infection in the human on gametocyte quantity and viability; and (ii) any parasite interactions during sporogonic development within the vector. Although there is some indication that co-species infection can alter gametocyte quantities [18], data are often based on studies using extremely small sample sizes [19] or on induced-infection studies [20], and the equivalence to actual transmission success (i.e. c) remains unclear. Although transmission success generally increases with gametocyte density [21,22], there is considerable variation and a clear indication of age-dependent effects [23–25]. During sporogonic development, species can affect each other directly – for example, by interference during fertilization [26] – or indirectly through the consequence of infection for mosquito vectorial capacity (see earlier and main text).

P. falciparum unispecies infections. Thus, the contribution of cryptic P. vivax infections might not be sufficient to account for the increase in P. vivax incidence following the drug treatment strategy, especially because the increase was considerably greater than the decrease in P. falciparum incidence. Therefore, it is difficult to explain such changing patterns of parasite species prevalence on the basis of treatment alone. Indeed, treatment strategy was not evoked as a reason for the increase in incidence of P. malariae; rather, the increase was suggested to be the result of the greater proximity of human and monkey environments, which increased the possibility of human exposure to mosquitoes infected with P. malariae from the monkey reservoir. If treatment strategy cannot entirely explain the observed patterns, could vector-orientated arguments have contributed to the observed change in P. vivax and P. falciparum incidence rates? Mosquito survival and sporogonic development One of the most important parameters determining the basic reproductive number (R0) is the relationship between mosquito longevity and the duration of sporogonic development within the mosquito (Box 1). Mosquito longevity can vary considerably among species and is strongly affected by environmental factors. By contrast, sporogonic development seems to be determined largely by temperature but varies among Plasmodium species; for a given temperature, the duration of development is shortest for P. vivax, intermediate for P. falciparum and longest for P. malariae [2,33]. Interventions that reduce mosquito

Opinion

TRENDS in Parasitology

lifespan would, therefore, be expected to impact most on P. malariae and least on P. vivax. However, under natural conditions, Plasmodium spp. manipulate the thermoregulatory behavior of their vector in a manner that accelerates its development [34] and, thus, the classical laboratory values of species developmental rates must be accepted with a certain amount of caution. In addition to exogenous factors determining mosquito survival, there is increasing evidence that the parasite has a negative effect on the fitness of the mosquito [35], although there is conflicting evidence concerning any significant effect of the parasite on mosquito longevity [11,12]. Increasing parasite burden has, however, been shown to reduce mosquito fecundity significantly [36,37], and it has been suggested that parasites actively reduce their burden on the mosquito [38,39]. Concurrent Plasmodium species infection in the mosquito If each Plasmodium species has evolved strategies to infect but not ‘overinfect’ mosquitoes, concurrent infection of a mosquito by multiple species is likely to have direct and/or indirect effects on parasite development. At present, there is no direct evidence concerning the effect of mixed Plasmodium spp. infections on the efficiency or speed of sporogonic development. Indirect approaches based on comparisons of mixed-species infection rates in sympatric human and mosquito populations have revealed conflicting results but were based on limited datasets [40,41]. There is a need for laboratory studies to determine the sporogonic consequences of mixed Plasmodium species infections, especially because co-infection with microfilaria and Plasmodium has been shown to increase mosquito mortality rates and reduce parasite development [16,17]. In addition to the important effects that mixed infections might have on one another during the relatively long period of sporogonic development, the Ross–Macdonald model indicates another parameter of importance: the proportion of bloodmeals from infected humans that gives rise to an infected mosquito (c) (Box 1). The impact of mixed infections on this parameter might be negligible, requiring the ‘improbable’ event of simultaneous ingestion of gametocytes of two Plasmodium species in one bloodmeal, but requires consideration. In an experimental system, despite no apparent interspecific effects within the vertebrate host, mosquito infection rates by one of the two co-infecting parasite species were reduced [26]. In addition, co-infections in the human host and effects on transmission to mosquitoes have been documented during cross-sectional surveys [19], although temporal variation in Plasmodium spp. prevalence rates in humans requires a longitudinal approach to the study of transmission to mosquitoes. Mosquito–parasite compatibility Species-specific differences in vector–parasite compatibility (vectorial competence or capacity) have an impact on the relative distribution of different parasite species. For instance, Anopheles gambiae and Anopheles pharoensis have been shown to be exclusive vectors of P. falciparum and P. vivax, respectively, in Gambella (Ethiopia), where www.sciencedirect.com

Vol.22 No.1 January 2006

23

these mosquitoes coexist [42]. It has been suggested that an increase in the number of P. vivax malaria cases in Southeast Asia could be due to changes in vector composition, with an increased overall susceptibility to P. vivax [43,44]. Recent evidence suggests that there is a genetic basis to variation in An. gambiae susceptibility to P. falciparum [45]. Differences in mosquito susceptibility to circumsporozoite protein variant types of P. vivax have also been reported between two species of Anopheles: Anopheles albimanus is more susceptible to VK210, whereas Anopheles pseudopunctipennis is more susceptible to VK247 [46]. Moreover, it has been shown that An. albimanus is the main vector of the VK210 phenotype but that An. pseudopunctipennis transmits both phenotypes in southern Mexico [47]. If such variation in compatibility occurs within a parasite species, it is extremely likely that variability would be at least as strong if two parasite species were considered. Thus, changes in vector species or genotype composition within a population – following, for example, vector-control interventions – could have a considerable impact on transmission of sympatric parasite species. Brazil revisited Longitudinal studies across a wide range of endemic settings have repeatedly shown that, at the onset of malaria transmission, the prevalence of P. vivax increases, followed by an increase in P. falciparum prevalence [48]. This reflects fundamental differences in the biology of the two species, such as faster sporogonic development, the existence of latent hypnozoite liver stages and the rapid production of gametocytes in P. vivax. In the area of Brazil that was studied by Po´voa et al. [28], transmission is seasonal and, therefore, it is likely that the parasites persist in a human infectious reservoir from which they must infect the first mosquitoes of the season. Such seasonality would initially give P. vivax the advantage because it infects mosquitoes first, and P. vivax gametocyte densities might dominate those of P. falciparum in mixed infections that have remained from the previous transmission season (i.e. the reservoir of infection). As shown classically, once transmission is underway, this initial advantage of P. vivax is rapidly lost after the spread of P. falciparum and the apparent asexual-stage dominance of the latter in mixed infections within the human host. If drug treatment preferentially targeted the novel P. falciparum infections, the increase in prevalence of this parasite would be considerably retarded and dependent on its pre-existing infectious reservoir. The increase in anopheline density would exacerbate the initial advantage of P. vivax that would further interfere with P. falciparum transmission from the untreated human reservoir. Thus, in light of the biology of the two Plasmodium species, the observed changes in parasite species composition are explained most easily through a combination of treatment strategy and changing anopheline density. In addition, the increased anopheline density was due largely to An. darlingi, the major South American vector of malaria parasites. The relative compatibility of An. darlingi with P. vivax and P. falciparum is unknown but might differ and vary in time and space.

24

Opinion

TRENDS in Parasitology

The accumulation of such information from many sites for this and other Anopheles species would be invaluable. In conclusion, measurable vector-associated parameters could significantly amplify small changes in prevalence rates that result from intervention strategies such as targeted drug treatment. Interspecific interactions and malaria Analysing species interactions, especially those involving parasites, is a major challenge for basic science and public health. Implementing control measures that perturb current species dynamics might lead to epidemiological changes but the extent to which such changes are permanent or predictable depends on whether the system is at equilibrium and, if so, on the stability of that equilibrium. Indeed, it is not clear that, as recently stated, parasite species live in ‘harmony’ [3] or have attained an equilibrium state. Intervention measures that target the vector or vectorial transmission, in a variety of forms, have been the mainstay of malaria control for the past 60 years. The latest high-technology vector-based hope, transgenic mosquitoes, is being considered despite serious misgivings [49–52]. Negative interactions between parasite species during development in the mosquito host could reduce the need for 100% efficacy in the mosquito resistance mechanism [52]. However, these interactions might also favor the spread of P. vivax or any other minor malaria species, leading to epidemiological changes unpredicted by technology-driven approaches. Because vector-based strategies will probably continue to have an important role in malaria control and because the importance of the lesser Plasmodium species is appreciated more than previously, it is essential to consider the vectorial component of malaria transmission to gain an appreciation of malaria epidemiology and to design malariacontrol programs in the most efficient way. Acknowledgements We thank Chris Drakeley and Vincent Robert for fruitful discussion and helpful comments about the manuscript. We are grateful for the constructive criticism of the anonymous referees. We apologize to the authors whose work was not cited directly because of space limitations. C.B. was supported by a Marie Curie Intra-European fellowship at Wageningen University and by the Institut de Recherche pour le De´veloppement. C.B. is grateful to the Joint Malaria Programme: a collaboration among the National Institute for Medical Research, Kilimanjaro Christian Medical Centre, the London School of Hygiene and Tropical Medicine, and the University of Copenhagen Centre for Medical Parasitology.

References 1 Sina, B. (2002) Focus on Plasmodium vivax. Trends Parasitol. 18, 287–289 2 Sattabongkot, J. (2004) Plasmodium vivax transmission: chances for control? Trends Parasitol. 20, 192–198 3 Snounou, G. and White, N.J. (2004) The co-existence of Plasmodium: sidelights from falciparum and vivax malaria in Thailand. Trends Parasitol. 20, 333–339 4 Zimmerman, P.A. et al. (2004) Why do we need to know more about mixed Plasmodium species infections in humans? Trends Parasitol. 20, 440–447 5 Luxemburger, C. et al. (1997) The epidemiology of severe malaria in an area of low transmission in Thailand. Trans. R. Soc. Trop. Med. Hyg. 91, 256–262 www.sciencedirect.com

Vol.22 No.1 January 2006

6 Bruce-Chwatt, L.J. (1963) A longitudinal survey of natural malaria infection in a group of West African adults. West Afr. Med. J. 12, 141–173 7 Ross, R. (1911) The Prevention of Malaria, Murray 8 Macdonald, G. (1957) The Epidemiology and Control of Malaria, Oxford University Press 9 Maitland, K. et al. (1997) Plasmodium vivax and P. falciparum: biological interactions and the possibility of cross-species immunity. Parasitol. Today 13, 227–231 10 Koella, J.C. (1991) On the use of mathematical models of malaria transmission. Acta Trop. 49, 1–25 11 Anderson, R.A. et al. (2000) Plasmodium falciparum sporozoites increase feeding-associated mortality of their mosquito hosts Anopheles gambiae s.l. Parasitology 120, 329–333 12 Ferguson, H. and Read, A. (2002) Why is the effect of malaria parasites on mosquito survival still unresolved? Trends Parasitol. 18, 256–261 13 Koella, J.C. and Packer, M.J. (1996) Malaria parasites enhance bloodfeeding of their naturally infected vector Anopheles punctulatus. Parasitology 113, 105–109 14 Koella, J.C. et al. (1998) The malaria parasite, Plasmodium falciparum, increases the frequency of multiple feeding of its mosquito vector, Anopheles gambiae. Proc. Biol. Sci. 265, 763–768 15 Koella, J.C. et al. (2002) Stage-specific manipulation of a mosquito’s host-seeking behavior by the malaria parasite Plasmodium gallinaceum. Behav. Ecol. 13, 816–820 16 Burkot, T.R. et al. (1990) The prevalence of naturally acquired multiple infections of Wuchereria bancrofti and human malaria in Anopheles. Parasitology 100, 369–375 17 Albuquerque, C.M. and Ham, P.J. (1995) Concomitant malaria (Plasmodium gallinaceum) and filaria (Brugia pahangi) infections in Aedes aegypti: effect on parasite development. Parasitology 110, 1–6 18 Price, R. et al. (1999) Risk factors for gametocyte carriage in uncomplicated falciparum malaria. Am. J. Trop. Med. Hyg. 60, 1019–1023 19 Graves, P.M. et al. (1988) Measurement of malaria infectivity of human populations to mosquitoes in the Madang area, Papua New Guinea. Parasitology 96, 251–263 20 McKenzie, F.E. et al. (2002) Plasmodium malariae infection boosts Plasmodium falciparum gametocyte production. Am. J. Trop. Med. Hyg. 67, 411–414 21 Schall, J.J. (2000) Transmission success of the malaria parasite Plasmodium mexicanum into its vector: role of gametocyte density and sex ratio. Parasitology 121, 575–580 22 Paul, R.E. et al. (2002) Plasmodium sex determination and transmission to mosquitoes. Trends Parasitol. 18, 32–38 23 Muirhead-Thomson, R.C. (1957) The malaria infectivity of an African village population to mosquitoes (Anopheles gambiae). Am. J. Trop. Med. Hyg. 6, 971–979 24 Boudin, C. et al. (1993) High human malarial infectivity to laboratorybred Anopheles gambiae in a village in Burkina Faso. Am. J. Trop. Med. Hyg. 48, 700–706 25 Paul, R.E. et al. (2004) Mosquitoes and transmission of malaria parasites – not just vectors. Malar. J. 3, 39 26 Paul, R.E. et al. (2002) Interspecific competition during transmission of two sympatric malaria parasite species to the mosquito vector. Proc. Biol. Sci. 269, 2551–2557 27 Davis, N.C. (1931) A note on the malaria carrying Anopheles in Bele´m, Para´n and in Natal, Rio Grande do Norte, Brazil. Riv. Malariol. 10, 43–51 28 Po´voa, M.M. et al. (2003) Malaria vectors, epidemiology, and the reemergence of Anopheles darlingi in Bele´m, Para´, Brazil. J. Med. Entomol. 40, 379–386 29 Akhavan, D. et al. (1999) Cost-effective malaria control in Brazil. Cost effectiveness of a malaria control program in the Amazon Basin of Brazil, 1988–1996. Soc. Sci. Med. 49, 1385–1399 30 Looareesuwan, S. et al. (1987) High rate of Plasmodium vivax relapse following treatment of falciparum malaria in Thailand. Lancet 2, 1052–1055 31 Mayxay, M. et al. (2004) Mixed-species malaria infections in humans. Trends Parasitol. 20, 233–240 32 Cavasini, M.T. et al. (2000) How prevalent is Plasmodium malariae in Rondoˆnia, Western Brazilian Amazon? Rev. Soc. Bras. Med. Trop. 33, 489–492

Opinion

TRENDS in Parasitology

33 Bruce-Chwatt, L.J. (1985) Essential Malariology, Heinemann 34 Fialho, R.F. and Schall, J.J. (1995) Thermal ecology of a malaria parasite and its insect vector: consequences for the parasite’s transmission success. J. Anim. Ecol. 64, 553–562 35 Koella, J.C. (1999) An evolutionary view of the interactions between anopheline mosquitoes and malaria parasites. Microbes Infect. 1, 303–308 36 Hurd, H. et al. (1995) Interactions between blood-feeding, fecundity and infection in mosquitoes. Parasitol. Today 11, 411–416 37 Ahmed, A.M. et al. (2001) Effects of malaria infection on vitellogenesis in Anopheles gambiae during two gonotrophic cycles. Insect Mol. Biol. 10, 347–356 38 Al-Olayan, E.M. et al. (2002) Apoptosis in the malaria parasite, Plasmodium berghei: a possible mechanism for limiting intensity of infection in the mosquito. Int. J. Parasitol. 32, 1133–1143 39 Hurd, H. (2003) Manipulation of medically important insect vectors by their parasites. Annu. Rev. Entomol. 48, 141–161 40 Arez, A.P. et al. (1997) Transmission of mixed malaria species and strains by mosquitoes, as detected by PCR, in a study area in GuineaBissau. Parassitologia 39, 65–70 41 Arez, A.P. et al. (2003) Transmission of mixed Plasmodium species and Plasmodium falciparum genotypes. Am. J. Trop. Med. Hyg. 68, 161–168 42 Nigatu, W. et al. (1992) Plasmodium vivax and P. falciparum epidemiology in Gambella, South-West Ethiopia. Trop. Med. Parasitol. 43, 181–185 43 Limrat et al. (2001) Anopheles barbirostris/campestris as a probable vector of malaria in Arayaprathet, Sa Keo Province. Southeast Asian J. Trop. Med. Public Health 32, 739–744

Vol.22 No.1 January 2006

44 Somboon, P. et al. (1994) Susceptibility of Thai zoophilic Anophelines and suspected malaria vectors to local strains of human malaria parasites. Southeast Asian J. Trop. Med. Public Health 25, 766–770 45 Lambrechts, L. et al. (2005) Host genotype by parasite genotype interactions underlying the resistance of anopheline mosquitoes to Plasmodium falciparum. Malar. J. 4, 3 46 Gonzalez-Ceron, L. et al. (1999) Differential susceptibilities of Anopheles albimanus and Anopheles pseudopunctipennis to infections with coindigenous Plasmodium vivax variants VK210 and VK247 in southern Mexico. Infect. Immun. 67, 410–412 47 Rodriguez, M.H. et al. (2000) Different prevalences of Plasmodium vivax phenotypes VK210 and VK247 associated with the distribution of Anopheles albimanus and Anopheles pseudopunctipennis in Mexico. Am. J. Trop. Med. Hyg. 62, 122–127 48 Christophers, R. (1949) Endemic and epidemic prevalence. In Malariology (Vol. 2) (London, B.M., ed.), pp. 698–721, W.B. Saunders 49 Boe¨te, C. (2005) Malaria parasites in mosquitoes: laboratory models, evolutionary temptation and the real world. Trends Parasitol. 21, 445–447 50 Boe¨te, C. (2005) Malaria-refractoriness in mosquito: just a matter of harbouring genes? In Genetically Modified Mosquitoes for Malaria Control (Boe¨te, C., ed.), Eurekah/Landes Bioscience (www.eurekah. com) 51 Chevillon, C. et al. (2005) Thinking transgenic vectors in a population context: some expectations and many open-questions. In Genetically Modified Mosquitoes for Malaria Control (Boe¨te, C., ed.), Eurekah/Landes Bioscience (www.eurekah.com) 52 Boe¨te, C. and Koella, J.C. (2002) A theoretical approach to predicting the success of genetic manipulation of malaria mosquitoes in malaria control. Malar. J. 1, 3

Endeavour the quarterly magazine for the history and philosophy of science You can access Endeavour online via ScienceDirect, where you’ll find a collection of beautifully illustrated articles on the history of science, book reviews and editorial comment.

featuring

Selling the silver: country house libraries and the history of science by Roger Gaskell and Patricia Fara Carl Schmidt – a chemical tourist in Victorian Britain by R. Stefan Ross The rise, fall and resurrection of group selection by M.E. Borello Mary Anning: the fossilist as exegete by T.W. Goodhue Caroline Herschel: ‘the unquiet heart’ by M. Hoskin Science in the 19th-century zoo by Oliver Hochadel The melancholy of anatomy by P. Fara and coming soon Etienne Geoffroy St-Hillaire, Napoleon’s Egyptian campaign and a theory of everything by P. Humphries Losing it in New Guinea: The voyage of HMS Rattlesnake by J. Goodman The accidental conservationist by M.A. Andrei Powering the porter brewery by J. Sumner Female scientists in films by B.A. Jones and much, much more . . . Locate Endeavour on ScienceDirect (http://www.sciencedirect.com) www.sciencedirect.com

25