Version classiqueVersion mobile
OpenEdition Books

Trente ans de lutte contre l’onchocercose en Afrique de l’Ouest. Traitements larvicides et protection de l’environnement

Laurent Yaméogo
Christian Levêque
Jean-Marc Hougard

Sélection d’articles du programme OCP / Selected OCP papers

Laboratory Toxicity of Potential Blackfly Larvicides on Some African Fish Species in the Onchocerciasis Control Programme Area

Laurent Yaméogo, Joseph-Marie Tapsoba et Davide Calamari

Texte intégral

1Received June 5, 1990

2The Onchocerciasis Control Programme of the World Health Organization uses larvicides to fight against the aquatic stages of the vector Simulium damnosum s.l., and thereby interrupt transmission of the disease. Since the appearance of resistance to Abate and chlorphoxim in certain cytotypes of the vector, the efficacy of many possible replacement insecticides has been tested and the impact of the best of them (permethrin, cyphenothrin, pyraclofos, and carbosulfan) on the aquatic fauna evaluated. © 1991 Academic Press, Inc.


3A vast Chemical control operation was launched in 1974 to interrupt the transmission of human onchocerciasis in several West African countries. The aquatic larval stages of the Simulium damnosum Theobald complex, the vector, are eliminated by the weekly application of degradable insecticides on the breeding sites. Three insecticides have been utilized in the Onchocerciasis Control Programme area (Fig. 1) : temephos, chlorphoxim, and Bacillus thuringiensis H-14. Aware of the potential environmental risk of this insecticidal treatment, the Programme pays particular attention to the selection of larvicides and has developed monitoring facilities and protocols. An ecological group, composed of independent experts, Controls and directs the Programme’s activities regarding environmental protection. The monitoring is geared mainly toward a regular study of the fish populations and the lotic benthic invertebrate populations on which many fish feed. In view of the results recorded during more than 10 years of larviciding, it can be said that the larvicides utilized have little effect on the nontarget fauna in the tropical lotie environments (Lévêque et al., 1988 ; Yaméogo et al., 1988 ; Lévêque, 1989).

4Since the appearance of resistance in some cytotypes of the S. damnosum s.l. complex to temephos and then chlorphoxim (Guillet et al., 1980 ; Kurtak et al., 1982), the screening of new blackfly larvicides has become one of the main fines of research in the Onchocerciasis Control Programme in West Africa.

5Among the many larvicides and/or formulations tested, permethrin, cyphenothrin (pyrethroids), pyraclofos (organophosphorus compound), and carbosulfan (carbamate) have shown sufficient efficacy against the vector and present a very little risk of rapid development of cross-resistance with the organophosphorus compounds.

6Trials were therefore carried out on the nontarget aquatic fauna to evaluate the toxicity levei of these larvicides. This paper reports the main results recorded during tank tests in laboratory on some fishes.

Fig. 1. Area covered by the Onchocerciasis Control Programme in West Africa (O.C.P.).

7These tests were performed in difficult periods and field conditions. Therefore the results are not presented as a standard comparison at a fixed time and for the same fish species. It was, however, considered relevant to publish the results because of the scantiness of data on aquatic toxicity for African fish species.


1. Fishes

8A number of species that are members of families which are well represented in the Programme area were used. They were Schilbe mystus, Pollimyrus isidori, Chrysichthys velifer, Chrysichthys nigrodigitatus, and Barbus macrops.

9S. mystus (Schilbeidae)—Linné 1762. This species is one of the most widespread African fish species, from the Zambesi and Kunene Rivers to the Senegal and Nile in the north including the east coast rivers. The maximum standard length is about 300 mm. The species feeds on insects, shrimps, small fishes, and fruits but is, above all, ichthyophagous and this tendency increases with size. The average length of the individuals tested was about 50 mm.

10P. isidori (Mormyridae)—Valenciennes 1846. Well known from the Niger, Gambia, Senegal, and Volta basins and from different Coastal rivers, P. isidori is a small species which feeds on insect larvae and on zooplankton. The average length of the individuals tested was about 47 mm.

11C. nigrodigitatus (Bagridae)—Lacépède 1803. This species is known from the Gambia, Senegal, Niger, and Volta basins and from most of the Coastal rivers. The maximum standard length is 475 mm but the average length of the individuals tested was 65 mm. Chrysichthys feed mainly on insect larvae (Chironomidae) but also on small mollusca, zooplankton, and Hemiptera.

12C. velifer (Bagridae)—Norman 1923. Reported only in the Bandama and Sassandra river basins, C. velifer presents the same feeding habits as C. nigrodigitatus. The average length of the individuais tested was 57 mm.

13Barbus macrops (Cyprinidae)—Boulenger 1911. Widely distributed in the Programme area, the maximum standard length of this small species is 98 mm but the average length of the species tested was 46 mm. It feeds mainly on insect larvae, small crustaceans (Copepoda, Cladocera, Ostracoda), and plant debris.

14Before the beginning of the trials the different fish species were kept in separate tanks in laboratory for a week. They were fed once daily and the water was oxygenated using diffusers.

15The fishes were of comparable size for each of the species. They were carefully put one by one into the test tanks so as to have 10 individuals of the same species in each solution. The water utilized was from the same source as the fishes.

2. Tests

16The acute toxicity tests were performed by means of the technique with periodic replacement of solutions (Ward and Parrish, 1983), which makes it possible to avoid the problems of decrease of the concentration of producís in aquariums during exposure time ; the solutions were replaced every 12 hr.

17The test solutions were 10 liters for each tank and were oxygenated. The water temperatures were 27 ± 2°C. The end point was death. The lethal concentrations were calculated according to the probit analysis (Finney, 1952).

18The Chemicals tested were : permethrin (3-phenoxybenzyl-(1 RS)-cis, trans 3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropanecarboxylate) ; cyphenothrin (cyano(3-phenoxyphenyl)-methyl-(1R)-cis, tans-2,2-dimethyl-3-(2-methyl-1-propenyl)-cyclopropa- necarboxylate) ; pyraclofos ((RS)-[O-1-(4-chlorophenyl)-pyrazol-4-yl-O-ethyl-S-pro-pylphosphorothioate]) ; and carbosulfan (2,3-dihydro-2,2-dimethyl-7-benzofuramyl [(dibutylamino) thio] methylcarbamate).

19The technical products were diluted taking into account the percentage of active ingredient and the results were expressed for each time, e.g., concentration of active principie (μg/liter). The larvicide solutions were prepared 1 hr before the start of the trials.


20The results are reported in Table 1 together with confidence limits and slopes. The toxicity curves are shown in Fig. 2. All the curves seem to be asymptotic and have normal trends ; therefore some missing data could be extrapolated. For example, the 72-hr LC50 for permethrin and P. isidori would be around 20 μg/liter, while the 72-hr LC50 for C. velifer exposed to cyphenothrin is about 150 μg/liter.


21The trials were carried out only with P. isidori. No mortality was recorded in the control tanks and the behavior of the individuais was normal up to the end of the trials.

TABLE 1. Median Lethal Concentrations (LC50) for Different Tropical Fish Species and Different Insecticides

TABLE 1. Median Lethal Concentrations (LC50) for Different Tropical Fish Species and Different Insecticides

22The fishes introduced into the permethrin solutions were a bit more active than those of the untreated tanks and less than 5 hr after the start of the trial some of them demonstrated disorderly movements followed by death.

23The 24-hr LC50 for P. isidori (40 μg/liter) is less than with pyraclofos and carbosulfan. The mean lethal concentration for a 48-hr exposure time is 26 μg/liter.

24A lot of data exist on the acute toxicity of permethrin in laboratory conditions on various fish species. However, they are quite scattered and different authors have given as 24-hr LC50 on rainbow trout values ranging from 8 μg/liter (Mulla et al., 1978) to 61 μg/liter (Coats and O’Donnel-Jeffrey, 1979) in comparable testing conditions. However, most of these values are in the low range : 12.5 ¿¿g/liter (Hill et al., 1976a), 13.7 μg/liter (Abram et al., 1980). Other species showed the same range of sensitivity. The 24-hr LC50 for Pimephales prometas is 15 μg/liter (Hill et al., 1976b), 6 ¿¿g/liter for Ictalurus punctatus (Buccafusco, 1976), 8.6-21 μg/liter (Hill et al, 1976c, 1977). Oreochromis aureus, a Tilapia species, has a toxicity of 6 ¿¿g/liter as 24-hr LC50 (Herzberg, 1988), while Oryzias latipes, Cyprinus carpio and Gambusia affinis have 41 μg/liter, 98 μg/liter, and 100 μg/liter, respectively (Miyamato, 1976 ; Hill et al., 1976c ; Mulla et al., 1978).

Fig. 2. Toxicity curves for the four insecticides tested. B.m, Barbus macrops ; P.i, Pollimyrus isidori ; C.n, Chrysichlhys nigrodigitatus ; C.v, Chrysichthys velifer ; S.m, Schilbe mystus. •, Pyraclofos ; O, carbosulfan ; X, permethrin ; □, cyphenothrin.


25During the trial no fish died in the control tanks, which were handled in the same way as the treated tanks. B. macrops and C. velifer were the species utilized for the trials.

26For B. macrops the 24-hr LC50 is the highest toxicity observed, while for C. velifer more than 40 times this dose is necessary to obtain 50 % mortality within 24 hr.

27For C. carpio, in a temperate climate, the 48-hr LC50 is 5.65 μg/liter (Sumitomo, 1983), a value that is very low compared with that recorded for C. velifer in a tropical environment.

28The two species tested, therefore, presented different susceptibilities to cyphenothrin. B. macrops was the most affected species ; for the same exposure time (48 hr), the LC50 of B. macrops was 18 times less than for C. velifer. However, according to the results of Sumitomo Ltd., the C. carpio susceptibility is two times higher than that of B. macrops.


29In the control tanks, only one Mormyridae died more than 48 hr after the beginning of the experiment.

30In addition to the mortalities recorded in the tanks treated, which made it possible to calculate the lethal doses (Table 1), the quite particular behavior of the Schilbeidae should be mentioned. In fact, 36 hr after treatment, while most of the fish in the control tank remained towards the bottom, the individuais that had received the larvicide but were sufficiently energetic tended to move to the surface. They showed signs of weakness towards the 72nd hr. The Schilbeidae then became very black, a characteristic of the survivors through the remainder of the experiment.

31The 24-hr LC50 of the species P. isidori for carbosulfan is less than that for S. mystus, being about half of the latter.

32Few data are available on carbosulfan acute toxicity. Rand et al. (1985) reported, in a review for bluegill, rainbow trout, and carp, 96-hr LC50 of 15, 42, and 55 μg/liter, respectively, while Bayoumi and Ibrahim (1988) observed a much higher toxicity on Tilapia zillii, 0.29 μg/liter, in contrast with the other data.


33Throughout the experiment no mortality was recorded in the control tanks for the two species used (P. isidori and C. nigrodigitatus).

34The behavior of P. isidori in the treated tanks did not change almost 10 hr after the introduction of the individuais into the larvicide solutions. Following the first change of solution, i.e., 12 hr after the start of the trial, the individuais showed agitation in all the tanks and mortalities were recorded for concentrations above 80 μg/liter. At the dose of 249 μg/liter, all the surviving individuais were moribund less than 24 hr after the beginning of the experiment and died within 48 hr.

35For C. nigrodigitatus, there were mortalities within 24 hr in the tanks treated with pyraclofos at 80 μg/liter. However, the situation became stable in all the solutions 72 hr after the start of the experiment and there were no additional mortalities up to the end of the trial 24 hr later.

36The above indicates a difference in susceptibility between the two species tested. The longer the exposure time, the more the least increase in dose results in considerable P. isidori mortalities, while C. nigrodigitatus becomes less affected by slight concentration increases. P. isidori therefore tolerates long-duration exposures less than C. nigrodigitatus even if the concentration of the pyraclofos solution is relatively low.

37The only available data for pyraclofos are those of the producer (Takeda, 1988), who gave the median tolerance limits (TL50) for an exposure time of 48 hr as 0.080 ppm for Salmo gairdneri and 0.044 for C. carpio.

Simulation Experiments

38Spraying simulations were carried out with Cyphenothrin on B. macrops. Solutions were prepared at the following doses : 0.1,0.05, 0.02, 0.01,0.001, and 0.0005 mg/liter. The individuais were then exposed to these doses successively for increasing times of 30 sec, 1 min, 3 min, 6 min, 3 hr, and 10 hr from the highest to the lowest doses. They were then put under observation in nontreated water. The experiment was repeated every week for 5 weeks and the tanks were visited regularly to remove dead individuais. The specimens were fed at least once every day in the observation tanks but this was stopped 24 hr before the start of each experiment.

39B. macrops was selected for this trial because of its relatively great susceptibility to cyphenothrin. Nevertheless, no mortality due directly to the insecticide was observed after 5 weeks of weekly treatment. The individuais manifested signs of agitation but it should be noted that even in the absence of any insecticide, Barbus are active by nature.


40As the first goal of this testing was to give an indication of the biological activity of these insecticides (which gave positive results on the target S. damnoswn) in terms of acute toxicity on fish, a comparison will be made between 24-hr LC50 for fish and the operational dose (OD).

41An operational dose is the concentration of insecticide that will kill Simulium larvae with a carry of 2.5 km at 10m3/sec and 10-15 km at 100m3/sec and it is usually expressed in mg/liter/10 min. Obviously the concentration at the spraying point in the river is 10 to 20 times higher than the operational dose but dilution quickly occurs since the Chemical insecticides are used at high water while during low water the biological insecticide B. thuringiensis is applied most of the time.

42It should be recalled as a guide that the 24-hr LC50 for P. isidori is 30.9 mg/liter with temephos (OD = 0.100) and that the lowest leveis found in literature data on other species are around 1 mg/liter. With the exception of this larvicide, which has been considered providential for the Programme for more than 10 years, all the insecticides currently utilized showed a high toxicity in laboratory on the fishes.

43The operational dose for permethrin, cyphenothrin, carbosulfan, and pyraclofos are 0.015, 0.015, 0.050, and 0.100 mg/liter/10 min, respectively.

44For permethrin, compared to the dose of 0.015 mg/liter utilized by OCP, the LC50 calculated are relatively higher while they correspond to those for cyphenothrin. Besides, the difference between the operational dose and the 24-hour LC50 is greater for permethrin than for carbosulfan.

45In laboratory, P. isidori is susceptible to the effects of carbosulfan. The susceptibility of S. mystus is less but the doses which caused mortalities are not too far from that utilized in the control campaign.

46For pyraclofos, the 24-hr LC50 of the species P. isidori is 1.5 times the operational dose and slightly greater than that of C. nigrodigitatus.

47In order to avoid mass fish mortalities and not considering any other factor, a first levei of a basal assessment is made comparing the 24-hr LC50 with the operational dose, bearing in mind that the exposure in river is much less (10 min versus 24 hr).

48Therefore, a worst-case situation is considered for the evaluation and a risk index for the ranking of the products is calculated.

49The following table compares the lowest 24-hr LC50 for fish with the operational dose :





Operational dose Mg/liter





24-hr LC50 fish Mg/liter





Risk index





50The highest score for risk was obtained by cyphenothrin ; it was medium for carbosulfan and pyraclofos, while the lowest was for permethrin. It can therefore be recommended to give low priority for further research to cyphenothrin in other experiments to be performed for more precise hazard assessments.


51The question whether African fish species are more tolerant to Chemical poisoning than fish from temperate areas has for a long time been a matter of debate among aquatic toxicologists, but experimental data were not available.

52Now, as a preliminary response, one can say from the above results that the toxicity of the insecticides tested is not extremely different for African fish species and fish from temperate areas and it seems that African fish react more or less like warm water fish with the exception of cyphenothrin, for which only a single record exists and, therefore, the possibilities of comparison are limited.

53From the few data available, it seems also that P. isidori and B. macrops are less tolerant than other species and this could be an indication for future toxicity testing.

54As a second resuit of this series of tests, one can give priority for further testing and investigation to permethrine, carbosulfan, and pyraclofos according to the ratio between operational dose and toxicity (0.37, 0.60, and 0.67, respectively).

55This second phase of the hazard assessment will include a literature search for metabolism, bioaccumulation potential, and persistence as well as experimental toxicity testing on nontarget invertebrate fauna in minigutter.

56The third phase will consist of the evaluation of the role of physicochemical parameters in the dispersion, taking into account river characteristics such as suspended solid/organic matters, flood, and carrying capacity, and in carrying out pilot-scale treatments on selected rivers. All these steps should be followed before an insecticide is allowed to be used in the Onchocerciasis Control Programme.

57Reports on the overall ecotoxicological evaluation are in preparation for pyraclofos (Yameogo et al., 1990b) and permethrin (Yameogo et ai, 1990a).


58This work was made possible by the technical collaboration of Mr. Nazaire Simpore (fisherman) who obtained specimens for us whenever necessary. We would also like to thank the drivers and laboratory assistants for their constant availability which permitted the smooth conduct of the experiments. Finally, we are indebted to the Chief of the Vector Control Unit (VCU) of the Programme and to the Programme Director for having approved the publication of this paper.



Abram, F. S. H., Evins, C., and Hobson, J. A. (1980). Permethrin for the control of animais in water mains. Water Res Cent. Tech. Rep. 145.

Bayoumi, O. C., and Ibrahim, H. Z. (1988). Determination of carbosulfan and itstwo major metabolites in water and their toxic effects against Culex pipiens and Tilapia zillii. Med. Fac. Landbouren. Rijksuniv. Gent 53(3b), 1471-1475.

Buccafusco, R. J. (1976). Acute toxicity of PP557 technical to channel catfish (Ictaluruspunctatus). EG&G Bionomics Report, June 1976.

Coats, J. R., and O’Donnel-Jeffrey, N. C. (1979). Toxicity of four synthetic pyrethroid insecticides to rainbow trout. Bull. Environ. Contam. Toxicol. 23, 250-255.

Finney, D. J. (1952). Probit Analysis, 2nd ed. Cambridge Univ. Press, Cambridge.

Guillet, P., Escaffre, M., Ouedraogo, M, and Quillevere, D. (1980). Mise en évidence d'une résistance au téméphos dans le complexe Simulium damnosum (S. sanctipauli et S. soubrense) en Côte d'ivoire. (Zone du Programme de Lutte Contre d’Onchocercose dans la Region du Bassin de la Volta). Cah. ORSTOM Sér. Entomol. Med. Parasitol. 17, 291-299.

Hersberg, A. M. (1988). Toxicity and accumulation of permethrin in the tilapia Greochramis aureus. Isr. J. Aquacult. Bamidgeh 40(1), 35-39.

Hill, R. W., Maddock, B. G., and Hart, B. (1976a). Determination of the toxicity of PP 557 to Rainbow trout (Salmo gairdneri). (Rep. No. BL/B/1700). ICI Brixham Lab.

Hill, R. W., Maddock, B. G., and Gilbert, J. L. (1976b). Determination of the acute toxicity of PP 557 to fothead minnows (Pimephales prometas) (Rep. No. BL/B/1726). ICI Brixham Lab.

Hill, R. W., Maddock, B. G., Hart, B., and Cornish, S. K. (1976c). Determination of acute toxicity of PP557 to mirror carp (Cyprinus carpio) (Rep. No. BL/B/1715). ICI Brixham Lab.

Hill, R. W., Maddock, B. G., Hart, B., and Bowles, P. G. (1977). Determination of the acute toxicity of formulation JFU 5054 to rainbow trout (Salmo gairdneri) (Rep. No. BL/B/1799). ICI Brixham Lab.

Kurtak, D., Ouedraogo, M., Ocran, M., Barro, T., and Guillet, P. (1982). Preliminary note on the appearance in Ivory Coast of resistance to chlorphoxim in Simulium soubrense/Sanctipauli larvae already resistant to temephos (Abate®). WHO unpublished document WHO/VBC/82.850.

Leveque, C. (1989). The use of insecticides in the onchocerciasis Control Programme and aquatic monitoring in West Africa. SCOPE 317-335.

Leveque, G, Fairhurst, C. P., Abban, K., Paugy, D., Curtis, M. S., and Traore, K. (1988). Onchocerciasis Control Programme in West Africa. Ten years monitoring of fish populations. Chemosphere 17, 421-440.

Miyamoto, J. (1976). Degradation, metabolism and toxicity of synthetic pyrathroids. Environ. Health Perspect. 14, 14-28.

Mulla, M. S., Navvab-Gojrats, and Darwazeh, H. A. (1978). Toxicity of mosquito larvicidal pyrethroids to four species of freshwater fishes. Environ. Enlomo !. 7(3), 428-438.

Rand, G. M., Deprospo, J. D., Robinson, R. A., and Fletcher, M. J. (1985). Carbosulfan : Aquatic toxicology. Toxicologist 5(1).

Sumitomo Chemical Co., Ltd. (1983). S-2703 Forte. Technical information. Osaka, Japan.

Takeda Chemical Industries, Ltd. (1988). Technical information. Pyraclofos (TIA-230). Tokyo, Japan.

Ward, G. S., and Parrish, P. R. (1983). Manuel des methodes de recherche sur l’environnement aquatique. Sixième partie. Tests de toxicité. FAO. Doc. technique sur les pêches. No. 185. FIRI/T185 (Fr).

Yameogo, L., Leveque, C., Traore, K., and Fairhurst, C. P. (1988). Dix ans de surveillance de la faune aquatique des rivières d’Afrique de l’Ouest traitées contre les Simulies (Diptera. Simuliidae), agents vecteurs de l’onchocercose humaine. Naturalist Can. (Rev. Ecol. Syst.) 115, 287-298.

Yameogo, L., Calamari, D., Elouard, J. M., Guillet, P., Paugy, D., and Walsh, J. F. (1990a). Ecological assessment of the OCP use of permethrin, in preparation.

Yameogo, L., Tapsoba, J. M., and Bihoum, M. (1990b). Short-term impact of pyraclofos (a new organophosphate compound) on the non-target fauna in Tropical environment, submitted for publication.

Table des illustrations

Légende Fig. 1. Area covered by the Onchocerciasis Control Programme in West Africa (O.C.P.).
Fichier image/jpeg, 162k
Titre TABLE 1. Median Lethal Concentrations (LC50) for Different Tropical Fish Species and Different Insecticides
Fichier image/jpeg, 128k
Légende Fig. 2. Toxicity curves for the four insecticides tested. B.m, Barbus macrops ; P.i, Pollimyrus isidori ; C.n, Chrysichlhys nigrodigitatus ; C.v, Chrysichthys velifer ; S.m, Schilbe mystus. •, Pyraclofos ; O, carbosulfan ; X, permethrin ; □, cyphenothrin.
Fichier image/jpeg, 58k


Onchocerciasis Control Programme, 01 B.P. 549, Ouagadougou 01, Burkina Faso

© IRD Éditions, 2003

Conditions d’utilisation :