Armigeres subalbatus colonization of damaged pit latrines: a nuisance and potential health risk to residents of resettlement villages in Laos.

During the resettlement of 6500 persons living around the Nam Theun 2 hydroelectric project in Laos, more than 1200 pour-flush latrines were constructed. To assess the role of these latrines as productive larval habitats for mosquitoes, entomological investigations using Centers for Disease Control (CDC) light traps, visual inspection and emergence trapping were carried out in over 300 latrines during the rainy seasons of 2008-2010. Armigeres subalbatus (Diptera: Culicidae) were nine times more likely to be found in latrines (mean catch: 3.09) than in adjacent bedrooms (mean catch: 0.37) [odds ratio (OR) 9.08, 95% confidence interval (CI) 6.74-15.11] and mosquitoes were active in and around 59% of latrines at dusk. Armigeres subalbatus was strongly associated with latrines with damaged or improperly sealed septic tank covers (OR 5.44, 95% CI 2.02-14.67; P < 0.001). Armigeres subalbatus is a nuisance biter and a putative vector of Japanese encephalitis and dengue viruses. Dengue virus serotype 3 was identified from a single pool of non-blood-fed female A. subalbatus using reverse transcription polymerase chain reaction. Maintaining a good seal around septic tanks by covering them with a layer of soil is a simple intervention to block mosquito exit/entry and contribute to vector control in resettlement villages. The scale-up of this simple, cheap intervention would have global impact in preventing the colonization of septic tanks by nuisance biting and disease-transmitting mosquitoes.

Preliminary studies developing methods for the control of Chrysomya putoria, the African latrine fly, in pit latrines in The Gambia.

OBJECTIVE: To explore ways of controlling Chrysomya putoria, the African latrine fly, in pit latrines. As pit latrines are a major source of these flies, eliminating these important breeding sites is likely to reduce village fly populations, and may reduce the spread of diarrhoeal pathogens. METHODS: We treated 24 latrines in a Gambian village: six each with (i) pyriproxyfen, an insect juvenile hormone mimic formulated as Sumilarv(®) 0.5 G, a 0.5% pyriproxyfen granule, (ii) expanded polystyrene beads (EPB), (iii) local soap or (iv) no treatment as controls. Flies were collected using exit traps placed over the drop holes, weekly for five weeks. In a separate study, we tested whether latrines also function as efficient flytraps using the faecal odours as attractants. We constructed six pit latrines each with a built-in flytrap and tested their catching efficiency compared to six fish-baited box traps positioned 10 m from the latrine. Focus group discussions conducted afterwards assessed the acceptability of the flytrap latrines. RESULTS: Numbers of emerging C. putoria were reduced by 96.0% (95% CIs: 94.5-97.2%) 4-5 weeks after treatment with pyriproxyfen; by 64.2% (95% CIs: 51.8-73.5%) after treatment with local soap; by 41.3% (95% CIs = 24.0-54.7%) after treatment with EPB 3-5 weeks after treatment. Flytraps placed on latrines collected C. putoria and were deemed acceptable to local communities. CONCLUSIONS: Sumilarv 0.5 G shows promise as a chemical control agent, whilst odour-baited latrine traps may prove a useful method of non-chemical fly control. Both methods warrant further development to reduce fly production from pit latrines. A combination of interventions may prove effective for the control of latrine flies and the diseases they transmit.

Habitat associations, reproduction and diet of the Guinean tilapia Tilapia guineensis of the Gambia River floodplains.

The ecology of the Guinean tilapia Tilapia guineensis a dominant species of the lower Gambia River floodplains and an important food source in parts of West Africa was studied to better understand the threat posed from construction of a barrage across the river. The catch per unit effort (CPUE) of T. guineensis was positively associated with conductivity and dissolved oxygen and negatively associated with water depth and the % vegetation cover. Diet studies indicated that the T. guineensis is primarily an iliophage. The peak of reproduction was at the beginning of the rainy season. CPUE peaked in May, just before the first rains, and subsequently declined, probably reflecting movement into newly flooded habitat. Median size at maturity was 11.6 cm total length, L(T), for females and 12.5 cm L(T) for males. L(T)-frequency analysis indicated several juvenile cohorts as well as very young fish on the floodplains each month, suggesting continuation of reproduction throughout the rainy season. The impending construction of a hydroelectric dam on the Gambia River is likely to affect T. guineensis negatively through anticipated changes in the hydrology of the river.

Suitability of two carbon dioxide-baited traps for mosquito surveillance in the United Kingdom.

Rapidly changing environments and an increase in human movement around the globe have contributed to a rise in new and emerging diseases, many of which are arthropod borne. The threat posed to the United Kingdom by such diseases is uncertain, and there is a real need to understand the distribution, seasonality and behaviour of potential vectors in the country. At present, there is no standard method for routine mosquito surveillance in the UK. Here we compared the catching efficiency of two carbon dioxide-baited traps, the CDC light trap and the MosquitoMagnet Pro trap, for collecting British mosquitoes. Two of each type of trap were operated at four sites in central and southern England from June to September, 2003. To determine whether trap height affected collections, three light traps were operated at 1, 2.5 and 5 m above the ground in one site in 2004. Both types of trap were efficient at catching mosquitoes, collecting 5414 mosquitoes of 16 species. MosquitoMagnet traps caught 2.7 times more mosquitoes than CDC light traps (P<0.001) and a wider range of species (16 species vs 11) than CDC light traps. Four to six times more female Culex pipiens s.l. were collected in light traps at 5 m (P<0.001) compared with traps at lower heights. MosquitoMagnet traps ran continuously for up to 8 weeks, whilst the battery of a CDC light trap had to be replaced every 24 hrs. Although MosquitoMagnets collected more specimens and a greater range of mosquito species, they were considerably more expensive, prone to breakdown and incurred higher running costs than the CDC light traps. MosquitoMagnets are useful tools for collecting mosquitoes during longitudinal surveys during the summer months, whilst CDC light traps are to be preferred for rapid assessments of the presence or absence of mosquitoes, particularly the important species Culex pipiens.

Responses of adult mosquitoes of two sibling species, Anopheles arabiensis and A. gambiae s.s. (Diptera: Culicidae), to high temperatures.

It is well known that amongst the sibling species of the Anopheles gambiae complex, A. arabiensis Patton predominates over A. gambiae sensu stricto Giles in hotter, drier parts of Africa. Here it was investigated whether A. arabiensis is better adapted to higher temperatures than A. gambiae s.s. at the microclimatic level. Bioassays were used to assess behavioural avoidance activity of adult mosquitoes in the presence of increasing temperature. Female mosquitoes were introduced into a holding tube from which they could escape into a cage through a one-way funnel. From a starting temperature of 28 degrees C they were exposed to a 2 degrees C rise in temperature every 30 min until all mosquitoes had escaped or been knocked down. As temperature increased, A. arabiensis left the holding tube at higher temperatures than A. gambiae s.s. (A. arabiensis mean activation temperature = 35.7 degrees C, 95% CIs = 35.4-36.1 degrees C; A. gambiae s.s. = 33.0 degrees C, 32.5-33.5 degrees C). To determine the relative ability of both species to survive at extremely high temperatures, batches of insects were exposed to 40 degrees C for different periods. It took considerably longer to kill 50% of A. arabiensis at 40 degrees C than it did A. gambiae s.s. (112 min vs. 67 min). These data show that adult A. arabiensis are better adapted to hotter conditions than A. gambiae s.s., a characteristic that is reflected in their spatial and temporal distribution in Africa.

Temperature-related duration of aquatic stages of the Afrotropical malaria vector mosquito Anopheles gambiae in the laboratory.

Vector abundance is an important factor governing disease risk and is often employed when modelling disease transmission. The longevity of the aquatic stages of mosquitoes (Diptera: Culicidae) dictates the rate of production of adults and hence the intensity of disease transmission. We examined how temperature influences the survival of larval stages (larvae and pupae) of Anopheles gambiae Giles sensu stricto and subsequent adult production of this most efficient malaria vector. Groups of 30 mosquitoes were reared at constant temperatures (from 10 to 40 degrees C) from the first instar and observed until death or metamorphosis of the last individual. Larvae developed into adults at temperatures ranging from 16 to 34 degrees C. Larval survival was shortest (< 7 days) at 10-12 degrees C and 38-40 degrees C, and longest (> 30 days) at 14-20 degrees C. Within the temperature range at which adults were produced, larval mortality was highest at the upper range 30-32 degrees C, with death (rather than adult emergence) representing over 70% of the terminal events. The optimal survival temperatures were lower than the temperatures at which development was quickest, suggesting a critical relationship between temperature and the life cycle of the insect. These data provide fundamental information about An. gambiae s.s. adult productivity at different temperatures, which may facilitate the construction of process-based models of malaria risk in Africa and the development of early warning systems for epidemics.

Effect of topography on the risk of malaria infection in the Usambara Mountains, Tanzania.

We investigated whether the risk of infection with malaria parasites was related to topography in the Usambara Mountains, Tanzania. Clinical surveys were carried out in seven villages, situated at altitudes from 300 m to 1650 m. Each village was mapped and incorporated into a Digital Terrain Model. Univariate analysis showed that the risk of splenomegaly declined with increasing altitude and with decreasing potential for water to accumulate. Logistic regression showed that altitude alone could correctly predict 73% of households where an occupant had an enlarged spleen or not. The inclusion of land where water is likely to accumulate within 400 m of each household increased the accuracy of the overall model slightly to 76%, but significantly improved predictions between 1000 m and 1200 m, where malaria is unstable, and likely to be epidemic. This novel approach illustrates how topography could help identify local areas prone to epidemics in the African highlands.

Effect of temperature on the development of the aquatic stages of Anopheles gambiae sensu stricto (Diptera: Culicidae).

Global warming may affect the future pattern of many arthropod-borne diseases, yet the relationship between temperature and development has been poorly described for many key vectors. Here the development of the aquatic stages of Africa's principal malaria vector, Anopheles gambiae s.s. Giles, is described at different temperatures. Development time from egg to adult was measured under laboratory conditions at constant temperatures between 10 and 40 degrees C. Rate of development from one immature stage to the next increased at higher temperatures to a peak around 28 degrees C and then declined. Adult development rate was greatest between 28 and 32 degrees C, although adult emergence was highest between 22 and 26 degrees C. No adults emerged below 18 degrees C or above 34 degrees C. Non-linear models were used to describe the relationship between developmental rate and temperature, which could be used for developing process-based models of malaria transmission. The utility of these findings is demonstrated by showing that a map where the climate is suitable for the development of aquatic stages of A. gambiae s.s. corresponded closely with the best map of malaria risk currently available for Africa.

Relationship between altitude and intensity of malaria transmission in the Usambara Mountains, Tanzania.

There is a consensus that malaria is a growing problem in African highlands. This is surprising because many parts of the highlands were considered too cold to support transmission. In this report, we examined how transmission of Plasmodium falciparum in six villages changed along an altitude transect in the Usambara Mountains, Tanzania, from 300 m to 1700 m. Routine entomological collections were made using spray catches and light traps for 15 mo. Direct estimates of entomological inoculation rates and indirect estimates of vectorial capacity suggested a >1000-fold reduction in transmission intensity between the holoendemic lowland and the hypoendemic highland plateau. Lowland transmission was perennial with a significant peak in the cool season after the long rains in May, when vectors densities were high. In the highlands, low temperatures prevented parasite development in mosquitoes during the cool season rains, and highland transmission was therefore limited to the warm dry season when vector densities were low. The primary effect of increasing altitude was a log-linear reduction in vector abundance and, to a lesser extent, a reduction in the proportion of infective mosquitoes. Highland malaria transmission was maintained at extraordinarily low vector densities. We discuss herein the implications of these findings for modeling malaria and suggest that process-based models of malaria transmission risk should be improved by considering the direct effect of temperature on vector densities. Our findings suggest that variation in the short rains in November and changes in agricultural practices are likely to be important generators of epidemics in the Usambaras.

Localized breeding of the Anopheles gambiae complex (Diptera: Culicidae) along the River Gambia, West Africa.

A study was undertaken to identify the major larval habitats of the Anopheles gambiae (Giles) complex in rural Gambia. Mosquito larvae and pupae were sampled along transects and in specific habitats in the central region of the country during the rainy seasons of 1996 and 1997. The sampling showed that the major breeding sites were located on the flooded alluvial soils bordering the river. The largest numbers of larvae were found during September, one month after the peak rains. Polymerase chain reaction analysis of specimens showed that Anopheles melas (Theobald) was the dominant species in the flooded areas (81.5%), followed by A. gambiae sensu stricto (Giles) (18.0%) and A. arabiensis (Patton) (0.5%). By sampling in specific habitats it was evident that A. arabiensis was mainly breeding in rain-fed rice fields along the edge of the alluvial soils. Anopheles melas and A. gambiae s.s. often coexisted but whereas A. melas were found in water with a salinity of up to 72% sea water (25.2 g NaCl l(-1)), A. gambiae s.s. only occurred in water with up to 30% sea water (10.5 g NaCl l(-1)). Anopheles melas larvae were found in association with plant communities dominated by sedges and grasses (Eleocharis sp., Paspalum sp., Sporobolus sp.) and sea-purslane Sesuvium portulacastrum (L.) and the presence of cattle hoof prints, whereas A. gambiae s.s. larvae mainly occurred in association with Paspalum sp. and Eleocharis sp. The study showed that even during the peak rainy season, breeding of the A. gambiae complex is almost entirely restricted to the extensive alluvial areas along the river.

Changes in house design reduce exposure to malaria mosquitoes.

House design may affect an individual's exposure to malaria parasites, and hence to disease. We conducted a randomized-controlled study using experimental huts in rural Gambia, to determine whether installing a ceiling or closing the eaves could protect people from malaria mosquitoes. Five treatments were tested against a control hut: plywood ceiling; synthetic-netting ceiling; insecticide-treated synthetic-netting ceiling (deltamethrin 12.5 mg/m2); plastic insect-screen ceiling; or the eaves closed with mud. The acceptability of such interventions was investigated by discussions with local communities. House entry by Anopheles gambiae, the principal African malaria vector, was reduced by the presence of a ceiling: plywood (59% reduction), synthetic-netting (79%), insecticide-treated synthetic-netting (78%), plastic insect-screen (80%, P < 0.001 in all cases) and closed eaves (37%, ns). Similar reductions were also seen with Mansonia spp., vectors of lymphatic filariasis and numerous arboviruses. Netting and insect-screen ceilings probably work as decoy traps attracting mosquitoes into the roof space, but not the room. Ceilings are likely to be well accepted and may be of greatest benefit in areas of low to moderate transmission and when used in combination with other malaria control strategies.

Irrigated crop production is associated with less malaria than traditional agricultural practices in Tanzania.

There is concern that crop irrigation that results in increased numbers of vector mosquitoes will lead to a rise in malaria in local communities. We evaluated the level of malaria experienced in 3 communities in northern Tanzania with different agricultural practices: rice irrigation, sugar-cane irrigation and traditional maize cultivation. Five cross-sectional surveys were used to measure the prevalence of infection with falciparum malaria in 1-4 years old children in each community over a period of 12 months. Active case detection was also carried out to record clinical episodes of malaria during the study period. Information on antimalarial measures was also recorded. Results from the cross-sectional surveys showed that the overall prevalence of malaria parasites was less near the rice irrigation (12.5%) and sugar-cane (16.9%) schemes than the savannah village (29.4%). There were also significantly fewer clinical episodes of malaria in the rice village (15 cases/1000 child-weeks at risk [cwar]) than either the sugar-cane (36 cases/1000 cwar) or savannah (40 cases/1000 cwar) villages. Overall, rice irrigation was associated with less malaria than alternative agricultural practices, despite the considerable numbers of vectors produced in the paddies. This finding supports other studies that indicate that irrigation in much of sub-Saharan Africa will not lead to increased malaria. Nonetheless, African governments planning irrigation projects need effective policies to encourage local communities to use personal protection measures, such as insecticide-treated bednets, and to ensure that these communities have access to effective antimalarial drugs and efficient health services.

Schistosome transmission, water-resource development and altitude in northern Ethiopia.

Schistosomiasis continues to be a major public-health problem, not least in association with water-resource developments. The impact of microdam construction in the northern Ethiopian highlands, in relation to possible increased risks of Schistosoma mansoni infection, has now been assessed. The results of incidence studies, carried out on 473 individuals sampled across eight microdam sites at altitudes of 1800-2225 m above sea level, indicated an overall annual incidence of 0.20 infections/person at risk. A multivariate Poisson regression model showed altitude and sex to be significant risk factors for infection, whereas proximity to a microdam was not significant, except possibly at very high altitudes. It was concluded that altitude was the major factor in this environment and that therefore, at least in terms of public-health planning, microdams should be sited as high as local geography permits.

Short-range attractiveness of pregnant women to Anopheles gambiae mosquitoes.

Malaria is a major cause of illness and an indirect cause of mortality in pregnant women. It can also cause stillbirths and low-birthweight babies. We have shown previously that pregnant women attracted twice as many Anopheles gambiae mosquitoes, the principal African malaria vector, as their non-pregnant counterparts over distances of about 15 m. In the current study (in 1998/99) we compared the short-range attractiveness of both pregnant and non-pregnant women sleeping under untreated bednets in Gambian villages. First, we measured the rate of mosquito entry under bednets and, second, we calculated the proportion of mosquitoes biting mothers under each bednet compared to their children. The feeding preference of An. gambiae collected under nets was determined by DNA fingerprinting blood samples from human subjects sleeping under each bednet and comparing these to fingerprints obtained from mosquito bloodmeals. Pregnant women were more attractive to An. gambiae mosquitoes than non-pregnant women under an untreated bednet. The number of mosquitoes entering bednets each night was 1.7-4.5 times higher in the pregnant group (P = 0.02) and pregnant women also received a higher proportion of bites under the bednets than did non-pregnant women (70% vs 52%, P = 0.001). This study clearly demonstrates that pregnant women are more exposed to malaria parasites than other women, which contributes to the greater vulnerability of pregnant women to malaria.

Malaria transmission risk variations derived from different agricultural practices in an irrigated area of northern Tanzania.

Malaria vector Anopheles and other mosquitoes (Diptera: Culicidae) were monitored for 12 months during 1994-95 in villages of Lower Moshi irrigation area (37 degrees 20' E, 3 degrees 21' S; approximately 700 m a.s.l.) south of Mount Kilimanjaro in northern Tanzania. Adult mosquito populations were sampled fortnightly by five methods: human bait collection indoors (18.00-06.00 hours) and outdoors (18.00-24.00 hours); from daytime resting-sites indoors and outdoors; by CDC light-traps over sleepers. Anopheles densities and rates of survival, anthropophily and malaria infection were compared between three villages representing different agro-ecosystems: irrigated sugarcane plantation; smallholder rice irrigation scheme, and savannah with subsistence crops. Respective study villages were Mvuleni (population 2200), Chekereni (population 3200) and Kisangasangeni (population approximately/= 1000), at least 7 km apart. Anopheles arabiensis Patton was found to be the principal malaria vector throughout the study area, with An. funestus Giles sensu lato of secondary importance in the sugarcane and savannah villages. Irrigated sugarcane cultivation resulted in water pooling, but this did not produce more vectors. Anopheles arabiensis densities averaged four-fold higher in the ricefield village, although their human blood-index was significantly less (48%) than in the sugarcane (68%) or savannah (66%) villages, despite similar proportions of humans and cows (ratio 1:1.1-1.4) as the main hosts at all sites. Parous rates, duration of the gonotrophic cycle and survival rates of An. arabiensis were similar in villages of all three agro-ecosystems. The potential risk of malaria, based on measurements of vectorial capacity of An. arabiensis and An.funestus combined, was four-fold higher in the ricefield village than in the sugarcane or savannah villages nearby. However, the more realistic estimate of malaria risk, based on entomological inoculation rates, indicated that exposure to infective vectors was 61-68% less for people in the ricefield village, due to the much lower sporozoite rate in An. arabiensis (ricefield 0.01%, sugarcane 0.1%, savannah 0.12%). This contrast was attributed to better socio-economic conditions of rice farmers, facilitating relatively more use of antimalarials and bednets for their families. Our findings show that, for a combination of reasons, the malaria challenge is lower for villagers associated with an irrigated rice-growing scheme (despite greater malaria vector potential), than for adjacent communities with other agro-ecosystems bringing less socio-economic benefits to health. This encourages the development of agro-irrigation schemes in African savannahs, provided that residents have ready access to antimalaria materials (i.e. effective antimalaria drugs and insecticidal bednets) that they may better afford for protection against the greater vectorial capacity of An. arabiensis from the ricefield agro-ecosystem.

Do untreated bednets protect against malaria?

Bednets are thought to offer little, if any, protection against malaria, unless treated with insecticide. There is also concern that the use of untreated nets will cause people sleeping without nets to receive more mosquito bites, and thus increase the malaria risk for other community members. Regular retreatment of nets is therefore viewed as critical for malaria control. However, despite good uptake of nets, many control programmes in Africa have reported low re-treatment rates. We investigated whether untreated bednets had any protective benefit (in October and November 1996) in The Gambia where nets, although widely used, are mostly untreated. Cross-sectional prevalence surveys were carried out in 48 villages and the risk of malaria parasitaemia was compared in young children sleeping with or without nets. Use of an untreated bednet in good condition was associated with a significantly lower prevalence of Plasmodium falciparum infection (51% protection [95% CI 34-64%], P < 0.001). This finding was only partly explained by differences in wealth between households, and children in the poorest households benefited most from sleeping under an untreated net (62% protection [14-83%], P = 0.018). There was no evidence that mosquitoes were diverted to feed on children sleeping without nets. These findings suggest that an untreated net, provided it is in relatively good condition, can protect against malaria. Control programmes should target the poorest households as they may have the most to gain from using nets.

Safety of the insect repellent N,N-diethyl-M-toluamide (DEET) in pregnancy.

The safety of daily application of N, N-diethyl-m-toluamide (DEET) (1.7 g of DEET/day) in the second and third trimesters of pregnancy was assessed as part of a double-blind, randomized, therapeutic trial of insect repellents for the prevention of malaria in pregnancy (n = 897). No adverse neurologic, gastrointestinal, or dermatologic effects were observed for women who applied a median total dose of 214.2 g of DEET per pregnancy (range = 0-345.1 g). DEET crossed the placenta and was detected in 8% (95% confidence interval = 2.6-18.2) of cord blood samples from a randomly selected subgroup of DEET users (n = 50). No adverse effects on survival, growth, or development at birth, or at one year, were found. This is the first study to document the safety of DEET applied regularly in the second and third trimesters of pregnancy. The results suggest that the risk of DEET accumulating in the fetus is low and that DEET is safe to use in later pregnancy.

Mapping distributions of chromosomal forms of Anopheles gambiae in West Africa using climate data.

The mosquito Anopheles gambiae Giles sensu stricto (Diptera: Culicidae), the principal vector of malaria in West Africa, comprises several chromosomal forms (e.g. Bissau, Forest, Mopti, Savanna) associated with climatic zones. Here we show how climate data can be used to map the geographical distribution of these chromosomal forms. The climate at 144 sites surveyed for mosquitoes in West Africa between 1971 and 92 was determined using computerized climate surfaces. Forest and Bissau forms occurred at relatively wet sites: median annual precipitation 1325 mm and 1438 mm, respectively, interquartile ranges (IQR) 1144-1858 mm and 1052-1825 mm), whilst the Mopti form was found at dry sites (annual 938 mm, IQR 713-1047 mm) and the Savanna form at sites intermediate between the wet and dry forms (annual 1067 mm, IQR 916-1279). Logistic regression analyses of the climate variables were carried out on a stratified random sample of half the sites. The resulting models correctly classified over 80% of the sites for presence or absence of each chromosomal form. When these models were tested against excluded sites they were also correct at over 80% of sites. The combined data produced models that were correct at over 86% of sites. Mean annual precipitation, evapotranspiration, minimum temperature and maximum temperature were the most important climate variables correlated with the distribution of these forms of An. gambiae. We used the logistic models to map the distribution of each chromosomal form within the reported range for An. gambiae s.s. in West Africa employing a geographical information system. Our maps indicate that each chromosomal form favours particular climate envelopes in well-defined ecoclimatic zones, although these forms are sympatric at the edges of their ranges. This study demonstrates that climate can be used to map the distribution of chromosomal forms of insects across large areas.

Human and other faeces as breeding media of the trachoma vector Musca sorbens.

The fly Musca sorbens Wiedemann (Diptera: Muscidae) apparently transmits Chlamydia trachomatis, causing human trachoma. The literature indicates that M. sorbens breeds predominantly in isolated human faeces on the soil surface, but not in covered pit latrines. We sought to identify breeding media of M. sorbens in a rural Gambian village endemic for trachoma. Test breeding media were presented for oviposition on soil-filled buckets and monitored for adult emergence. Musca sorbens emerged from human (6/9 trials), calf (3/9), cow (3/9), dog (2/9) and goat (1/9) faeces, but not from horse faeces, composting kitchen scraps or a soil control (0/9 of each). After adjusting for mass of medium, the greatest number of flies emerged from human faeces (1426 flies/kg). Median time for emergence was 9 (inter quartile range = 8-9.75) days post-oviposition. Of all flies emerging from faeces 81% were M. sorbens. Male and female flies emerging from human faeces were significantly larger than those from other media, suggesting that they would be more fecund and live longer than smaller flies from other sources. Female flies caught from children's eyes were of a similar size to those from human faeces, but significantly larger than those from other media. We consider that human faeces are the best larval medium for M. sorbens, although some breeding also occurs in animal faeces. Removal of human faeces from the environment, through the provision of basic sanitation, is likely to greatly reduce fly density, eye contact and hence trachoma transmission, but if faeces of other animals are present M. sorbens will persist.