Abstract.  The efficacy against mosquitoes (Diptera: Culicidae) of a bednet treated with carbamate insecticide [carbosulfan capsule suspension (CS) 200 mg/m2] was compared with four types of pyrethroid‐treated nets in veranda‐trap huts at Yaokoffikro near Bouaké, Côte d'Ivoire, where the malaria vector Anopheles gambiae Giles carries the kdr gene (conferring pyrethroid resistance) at high frequency and Culex quinquefasciatus Say is also pyrethroid resistant. Pyrethroids compared were lambdacyhalothrin CS 18 mg/m2, alphacypermethrin water dispersible granules (WG) 20 mg/m2, deltamethrin 50 mg/m2 (Permanet™) and permethrin emulsifiable concentrate (EC) 500 mg/m2. Insecticidal power and personal protection from mosquito bites were assessed before and after the nets were used for 8 months and hand washed five times in cold soapy water. Before washing, all treatments except permethrin significantly reduced blood‐feeding and all had significant insecticidal activity against An. gambiae. The carbosulfan net gave significantly higher killing of An. gambiae than all pyrethroid treatments except the Permanet. Against Culex spp., carbosulfan was more insecticidal and gave a significantly better protective effect than any of the pyrethroid treatments. After washing, treated nets retained various degrees of efficacy against both mosquito genera – but least for the carbosulfan net. Washed nets with three types of pyrethroid treatment (alphacypermethrin, lambdacyhalothrin, permethrin) gave significantly higher mortality rates of Culex than in huts with the same pyrethroid‐treated nets before washing. After five washes, the Permanet™, which is sold as a long‐lasting insecticidal product, performed no better than the other nets in our experimental conditions.

Only pyrethroid insecticides have so far been recommended for the treatment of mosquito nets for malaria control. Increasing resistance of malaria vectors to pyrethroids threatens to reduce the potency of this important method of vector control. Among the strategies proposed for resistance management is to use a pyrethroid and a non-pyrethroid insecticide in combination on the same mosquito net, either separately or as a mixture. Mixtures are particularly promising if there is potentiation between the two insecticides as this would make it possible to lower the dosage of each, as has been demonstrated under laboratory conditions for a mixture of bifenthrin (pyrethroid) and carbosulfan (carbamate). The effect of these types of treatment were compared in experimental huts on wild populations of Anopheles gambiae Giles and the nuisance mosquito Culex quinquefasciatus Say, both of which are multi-resistant. Four treatments were evaluated in experimental huts over six months: the recommended dosage of 50 mg m(-2) bifenthrin, 300 mg m(-2) carbosulfan, a mosaic of 300 mg m(-2) carbosulfan on the ceiling and 50 mg m(-2) bifenthrin on the sides, and a mixture of 6.25 mg m(-2) carbosulfan and 25 mg m(-2) bifenthrin. The mixture and mosaic treatments did not differ significantly in effectiveness from carbosulfan and bifenthrin alone against anophelines in terms of deterrency, induced exophily, blood feeding inhibition and overall mortality, but were more effective than in earlier tests with deltamethrin. These results are considered encouraging, as the combination of different classes of insecticides might be a potential tool for resistance management. The mixture might have an advantage in terms of lower cost and toxicity.

Resistance to carbosulfan, a carbamate insecticide, was detected in field populations of the malaria vector mosquito Anopheles gambiae Giles (Diptera: Culicidae) from two ecologically contrasted localities near Bouaké, Ivory Coast: rural M'bé with predominantly M form of An. gambiae susceptible to pyrethroids; suburban Yaokoffikro with predominantly S form of An. gambiae highly resistant to pyrethroids (96% kdr). The discriminating concentration of 0.4% carbosulfan (i.e. double the LC100) was determined from bioassays with the susceptible An. gambiae Kisumu strain. Following exposure to the diagnostic dosage (0.4% carbosulfan for 1 h), mortality rates of female An. gambiae adults (reared from larvae collected from ricefields) were 62% and 29% of those from M'bé and Yaokoffikro, respectively, 24 h post-exposure. Exposure for 3 min to netting impregnated with the operational dosage of carbosulfan 200 mg/m2 gave mortality rates of 88% of those from M'bé and only 12.2% for Yaokoffikro. In each case the control untreated mortality rate was insignificant. Biochemical assays to detect possible resistance mechanism(s) revealed the presence of insensitive AChE in populations of An. gambiae at both localities, more prevalent in the S form at Yaokoffikro than in M form at M'bé, as expected from bioassays results. Our study demonstrates the need to monitor carbamate resistance among populations of the An. gambiae complex in Africa, to determine its spread and anticipate vector control failure if these insecticides are employed.

The following study analyses the potentialities of the experimental huts built in M'be Valley (Côte d'Ivoire) where the evaluations of the insecticide products have been carried out for many years in line with the WHOPES protocol on the methodology of stage 2 assays. Starting a testing station first requires a good knowledge of the sensitivity of Anopheles gambiae to the main insecticide families. Then thanks to the experimental huts the efficacy of the various means of treatment can be compared with the one in untreated huts; this study focuses on house spraying using 100 mg a.i./m2 and bednets impregnated with lambda-cyhalothrin at a dose of 15 mg a.i./m2. The fipronil used in house spraying doesn't show any repellent effect, however it does have an irritating effect that increases the natural exophily of An. gambiae females entering the testing huts. The blood-feeding rate recorded in the treated huts was reduced to 24% and to 38% mortality rate consisting mainly of a 24 hours delayed mortality. The bednets treated with lambda-cyhalothrin have greatly reduced the contact between man and vector since the entry rate of An. gambiae females was cut down by 68% compared to the control. The exophily of this anopheles was twofold greater with the impregnated bednets and the blood-feeding rate reduced to 47%. Finally the global mortality rate, two thirds of immediate mortality, one third of delayed mortality, reached 35%. The experimental huts in the M'be Valley therefore provide essential information regarding the selection of the most efficacious insecticides against An. gambiae. This experimental method must be extended to other sites in order to finalize ever more selective and appropriate means of control against nuisance and disease-vector mosquitoes.

On the basis of sex chromosome variation, three cytotypes of Simulium squamosum (Enderlein) (Diptera: Simuliidae) are described from Cameroon and Nigeria. Simulium squamosum A is the typical form as originally described by Vajime & Dunbar (1975) with chromosome I as the sex chromosome. It occurs throughout most of Cameroon and south-east Nigeria. A second cytotype, S. squamosum B, is described from the river Sanaga (Cameroon). It also has chromosome I as the sex chromosome, but the nature of the sex differential region is different. Simulium squamosum C has no sex-linked chromosomal rearrangements. It is widespread in Nigeria and occurs near Mount Cameroon, where it seems to hybridize with S. squamosum A.

SummaryA new approach is proposed in the treatment of mosquito nets, using a ‘two‐in‐one’ combination of pyrethroid and non‐pyrethroid insecticides applied to different parts of bednets. The objectives are mainly to overcome certain limitations of pyrethroid‐impregnated bednets currently recommended for malaria control purposes. Apart from developing alternatives to pyrethroid dependency, we sought to counteract pyrethroid irritant effects on mosquitoes (excito‐repellency) and resistance to pyrethroids. The idea takes advantage of the presumed host‐seeking behaviour of mosquitoes confronted by a net draped over a bed, whereby the mosquito may explore the net from the top downwards. Thus, nets could be more effective if treated on the upper part with residual non‐irritant insecticide (carbamate or organophosphate) and with a pyrethroid on the lower part. Sequential exposure to different insecticides with distinct modes of action is equivalent to the use of a mixture as a potential method of managing insecticide resistance. We also intended to improve the control of nuisance mosquitoes, especially Culex quinquefasciatus Say (Diptera: Culicidae) that often survive pyrethroids, in order to encourage public compliance with use of insecticide‐treated nets (ITNs).Polyester bednets were pretreated with residual pyrethroid (bifenthrin 50 mg/m2 or deltamethrin 25 mg/m2) on the lower half and with carbamate (carbosulfan 300 mg/m2) on the upper half to minimize contact with net users. Unreplicated examples of these ‘two‐in‐one’ treated nets were field‐tested against wild mosquitoes, in comparison with an untreated net and bednets treated with each insecticide alone, including PermaNetTM wash‐resistant formulation of deltamethrin 50 mg/m2. Overnight tests involved volunteers sleeping under the experimental bednets in verandah‐trap huts at Yaokofikro, near Bouaké in Côte d'Ivoire, where the main malaria vector Anopheles gambiae Giles, as well as Culex quinquefasciatus Say, are highly resistant to pyrethroids. Efficacy of these ITNs was assessed in the huts by four entomological criteria: deterrency and induced exophily (effects on hut entry and exit), blood‐feeding and mortality rates (immediate and delayed).Overall, the best impact was achieved by the bednet treated with carbosulfan alone, followed by ‘two‐in‐one’ treatments with carbosulfan plus pyrethroid. Blood‐feeding rates were 13% An. gambiae and 17% Cx. quinquefasciatus in huts with untreated nets, but only 3% with carbosulfan ITNs, 7–11% with combinedITN treatment, 6–8% An. gambiae and 12–14% Cx. quinquefasciatus with pyrethroid alone. Mosquitoes that entered the huts were killed sooner by nets with combined treatment than by pyrethroid alone. Mortality‐rates in response to ITNs with carbosulfan (alone or combined with pyrethroid) were significantly greater for Cx. quinquefasciatus, but not for An. gambiae, compared to ITNs with only pyrethroid. About 20% of sleepers reported potential side‐effects (headache and/or sneezing) from use of ITN treated with carbosulfan alone. Further development of this new ‘two‐in‐one’ ITN concept requires a range of investigations (choice of effective products, cost‐benefit analysis, safety, etc.) leading to factory production of wash‐resistant insecticidal nets treated with complementary insecticides.

Surveillance for mosquito-borne viruses was conducted in Barkedji area from 1990 to 1995, following an outbreak of Rift Valley fever (RVF) virus in southern Mauritania. Mosquitoes, sand flies, and midges were collected from human bait and trapped by solid-state U.S. Army battery-powered CDC miniature light traps baited with dry ice or animals (sheep or chickens) at four ponds. Overall, 237,091 male and female mosquitoes representing 52 species in eight genera, 214,967 Phlebotomine sand flies, and 2,527 Culicoides were collected, identified, and tested for arboviruses in 9,490 pools (7,050 pools of female and 331 of male mosquitoes, 2,059 pools of sand flies and 50 pools of Culicoides). Viruses isolated included one Alphavirus, Babanki (BBK); six Flaviviruses, Bagaza (BAG), Ar D 65239, Wesselsbron (WSL), West Nile (WN), Koutango (KOU), Saboya (SAB); two Bunyavirus, Bunyamwera (BUN) and Ngari (NRI); two Phleboviruses, Rift Valley fever (RVF) and Gabek Forest (GF); one Orbivirus, Ar D 66707 (Sanar); one Rhabdovirus, Chandipura (CHP); and one unclassified virus, Ar D 95537. Based on repeated isolations, high field infection rates and abundance, Culex appeared to be the vectors of BAG, BBK, Ar D 65239 (BAG-like), and WN viruses, Ae. vexans and Ae. ochraceus of RVF virus, Mansonia of WN and BAG viruses, Mimomyia of WN and BAG viruses, and Phlebotomine of SAB, CHP, Ar D 95537, and GF viruses. Our data indicate that RVF virus circulated repeatedly in the Barkedji area.

Chikungunya fever is a viral disease transmitted to human beings by Aedes genus mosquitoes. From 1972 to 1986 in Kédougou, Senegal, 178 Chikungunya virus strains were isolated from gallery forest mosquitoes, with most of them isolated from Ae. furcifer-taylori (129 strains), Ae. luteocephalus (27 strains), and Ae. dalzieli (12 strains). The characteristics of the sylvatic transmission cycle are a circulation periodicity with silent intervals that last approximately three years. Few epidemics of this disease have been reported in Senegal. The most recent one occurred in 1996 in Kaffrine where two Chikungunya virus strains were isolated from Ae. aegypti. The retrospective analysis of viral isolates from mosquitoes, wild vertebrates, and humans allowed to us to characterize Chikungunya virus transmission cycles in Senegal and to compare them with those of yellow fever virus.

After an outbreak of Rift Valley fever in Southern Mauritania in 1987, entomologic studies were conducted in a bordering region in Sénégal from 1991 to 1996 to identify the sylvatic vectors of Rift Valley fever virus. The virus was isolated from the floodwater mosquitoes Aedes vexans and Ae. ochraceus. In 1974 and 1983, the virus had been isolated from Ae. dalzieli. Although these vectors differ from the main vectors in East and South Africa, they use the same type of breeding sites and also feed on cattle and sheep. Although enzootic vectors have now been identified in West Africa, the factors causing outbreaks remain unclear.

In two areas of Senegal where previous evidence of Rift Valley fever (RVF) virus circulation was detected, Barkedji in the Sahelian bioclimatic zone and Kedougou in the Sudano-Guinean zone, a longitudinal study of the enzootic maintenance of RVF virus was undertaken from 1991 to 1993. Mosquitoes, sand flies, and ticks were collected and domestic ungulates were monitored with serologic surveys. Rift Valley fever virus was not isolated in Kedougou. In Barkedji, RVF virus was isolated from Aedes vexans and Ae. ochraceus mosquitoes collected in traps near ground pools and cattle droves and from one health sheep. Sand flies were not involved in the maintenance cycle. Seroconversions were recorded in three (1.9%) of 160 monitored sheep and goats. The interepizootic vectors appeared to belong to the Aedes subgenus Neomelaniconion in East Africa, and to the subgenus Aedimorphus in West Africa. Epizootics in East Africa are associated with an increase in rainfall. However, factors associated with epizootics remain unknown for West Africa.

Ngari virus (NRI) (Bunyaviridae, genus Bunyavirus) was isolated first from male Aedes simpsoni mosquitoes in Southeastern Senegal in 1979. Then, it was recovered from several mosquito species in Senegal, Burkina Faso, Central African Republic and Madagascar. A potential pathogenicity of NRI virus in humans was suspected when the virus was isolated from two patients in Dakar in October and November 1993. The large diversity of Culicidae vectors and feeding patterns showed a large heterogeneity of vertebrate hosts. The wide geographical distribution of NRI virus in different bioclimatic areas indicated an important adaptability of the virus. Ngari virus epidemiology will need further investigations in order to approach the real pathogenicity of such emerging virus.

During the 1993 rainy season, 15,806 mosquitoes, including 14,304 Aedes ssp., were collected and tested for virus infection in 702 and 547 pools, respectively. Aedes furcifer (Edwards) was the most abundant species collected throughout the survey period. Yellow fever (YF) virus was detected in 187 pools: Ae. furcifer (123 isolates), Ae. taylori (Edwards) (41 isolates), and Ae. luteocephalus (Newstead) (23 isolates). A high prevalence of immunoglobulin (IgG) antibodies was found in human and simian populations. Results clearly indicated that increased sylvatic YF activity in eastern Senegal has the increased the risk of YF transmission among rural populations in West Africa. Our results showed that a minimal survey period may be effective in detecting the circulation of YF in the Kedougou area.

Thirteen strains of Rift Valley fever virus were isolated from Aedes vexans and Ae. ochraceus mosquitoes collected in October and November 1993 in northern Senegal. Entomologic and serologic data show that the risk of a new epizootic is increasing in this region.

Eleven phlebotomine sand Flies Species were captured in the Ferlo sahelian region during monthly surveys from April 1992 to June 1993. Around the Niakha ground pool, the most well studied location, the most abundant species were respectively : Sergentomyia dubia (40,71 %), S, schewtzi (25,54 %) and S. clydei (13,84 %). In total, 98,73 % oi the specimens captured belonged to the genus Sergentomyia, and only 1,27 % to the genus Phlebotomus. The sand flies community showed a peak oi abundance in January. The sand flies prefered habitats were termite hills, Followed by animals burrows and tree holes. In the Five communities studied, the most abundant species was S. dubio, the gecko leishmaniasis vector. P. duboscqi, the human cutaneous leishmaniasis vector, was captured each month, despite its low abundance, except in the ratel burrows. Chandipura and Saboya viruses, as well as two probably new viruses, were isolated From sand flies.

For the first time in West Africa, arboviruses were isolated from phlebotomine sand fly pools. One strain of Chandipura virus (a Vesiculovirus), four strains of Saboya virus (a Flavivirus), and one strain of a not yet identified virus were isolated. Three hundred twenty-two pools were established from a population of 33,917 sand flies caught in CO2 light traps in the Ferlo Sahelian region of Senegal from November 1991 to December 1992. This is the first isolation of Chandipura virus from any arthropod in Africa. Saboya virus has already been isolated from small rodents in Senegal; thus, its transmission cycle probably involves rodentophilic sand flies. No strain of Rift Valley fever phlebovirus, which caused an epizootic in this region in 1987, was isolated. During the same time at the same site, 11 sand fly species were identified from 4,191 specimens caught on sticky traps, including Phlebotomus duboscqi, a leishmaniasis vector.

During 1990, Dengue-2 (DEN-2) virus was isolated for the first time from mosquitoes (Aedes furcifer, six isolates; Ae. taylori, six isolates; Ae. luteocephalus, seven isolates) collected during an epidemic in which DEN-2 virus also was isolated from humans. Numerous isolations have been made previously from mosquitoes in the absence of human infection. In Senegal, DEN-2 virus appears to be maintained in an enzootic cycle and, therefore, plays an expanding role in human disease and increases the need for effective surveillance in mosquito populations.

To investigate past infection in and transmission of Rift Valley fever (RVF) virus to humans within an endemic focus, we undertook a retrospective cohort study of the seminomadic Peul people living in sub-Saharan northcentral Senegal. Residents of the rural settlement of Yonofere five years of age or older were studied during February-May 1989. Anti-RVF virus IgG was found in blood samples of 22.3% of 273 persons who responded to a standard questionnaire; none had IgM antibodies. Seropositivity was similar for males (25.4%) and females (21.1%), increased markedly with age for both sexes, and varied considerably among compounds (groups of huts) (0-37.5%). Risk factors for past RVF virus infection were nursing sick people, assisting animals during abortions/births, and treating sick animals. In all age groups, odds ratios (ORs) for RVF viral antibody among females who reported treating sick animals were three to six times greater than for those who did not. The ORs for males who reported assisting with animal births/abortions and nursing sick people were approximately five times those for males who did not. Serologic prevalence of RVF viral antibody among sheep averaged 30.1% overall (0.8% IgM), but varied among compounds (0-66.7%) in a manner different from that of humans. The seasonal abundance and relative density of potential mosquito vectors were estimated by monthly samples captured in Centers for Disease Control and Prevention-type traps. Mosquito abundance varied seasonally with rainfall (> 90% captures during four months). Species diversity was large (28 spp.), dominated by Aedes and Culex. Rift Valley fever virus was not isolated from 142 pools of 2,956 unengorged mosquitoes tested, although three other arboviruses were found. Results indicate that RVF is endemic in this region, people are at considerable risk of infection, and that a heretofore unrecognized mode of human infection under nonepizootic conditions may be transmission via contact with infected animals or humans.