The control of simulium damnosum in the river Niger and its tributaries in relation to the Kainji lake research project, covering the period 1961 to 1969i

@2.1 ,, 2.3 2.4 2.5 2.6 4.1 4.2 4.3 4.4 5.r 5.2 5.3 5.4 5.5 5.6 5,7 WORLD HEALTH ORGANIZATION ORGANISATION MONDIALE DE LA SANTE PDf 7o.4 ORIGINAL: ENGLISH RESTRICTED DISTRIBUTION a THE CONTROL OF SIMULIUM DAMNOST'M IN THE RI\TER NIGER AND ITS TRIBUTARIES IN RELATION TO THE KAINJI LAKE RESEARCH PROJECT, CO\TERING THE PERIOD 196I TO 1969I by J. F. Walsh Former Senior Entomologist to the Joint Consultant Engineers, Niger Dam Authority, Kainji, Nigeria CONTENTS t 2 t : t, , t I I I , Introduction . Description of the Kainji control area Location . Vegetation rrra fu.*ini Geology and drainage . C1 imate Population . . Resettlement J The prevalence and economic importance of onchocerciasis Prevalence . Economic imPortance 4. Distribution and prevalence of the vector Earlier surveys Control unit surveYs . Other species of Simulium Prevalence of adult S. damnosum 5. Treatments, materials and methods Choice of Iarviciding Choice of insecticide Concentration of larvicide ' Control season . . Larviciding cycles . . Larviciding Points . . Quantities of insecticide used The ieeue of thie document does not constitute formal publication. lt should not be reviewod! abstracted or quotod without the agreement of the World Health Organization. Authore alone are respon8ible for vicwa exprossed in signed articlcs. Page 3 4 4 4 4 5 6 tl 6 6 7 8 8 8 9 9 9 3.r 3.2 I 10 10 11 I1 1I t2) r Report prepared at the request of the worrd Health organization, Participating Agency 1n the Kainji Lake Research project financed by the united Nations Development Programme(special Fund), for which the Food and Agricultural Organization of the united Nations is the Executing AgencY. Ce document ne constitue pas une publication' ll ne doit faire I'objet d'aucun compte rendu ou r6gumd ni d'aucuno citation Sans I'autorigation de I'Organieation Mondiale de la Sant6. Les opinions exprim6es dans les articles signds n'engagent que leurs auteurs. 2CONTENTS Pag.e T26. Annua1 treatments Wet season control Dry season control Factors adversely affecting Annual summaries 7. Results Aquatic stages AduIts 8. Size. and composition of the Unit Activities and personnel Transport . . Costs . . 9. Discussion Preliminary investigations Dry season control Wet season control Organization and equipment Entomological investigations 1O. Conclusion Acknowl edgement s References . . Annexes: 6.r 6,2 6.3 6.4 8.1 8.2 8.3 9.1 9.2 9.3 9.4 9.5 control l2 t2 t2 13 a 13 I4 15 13 t6 7.t 7.2 15 15 I5 I6 l6 16 t7 18 18 t9 20 1 Indlcations of decreased susceptibillty the Kainji control area The use of traps . . AIl day catches Effect of larvicide on non-target organisms of Simulium damnosum larvae to DDT in 2. 3. 4. 22 23 25 26 27Tables and figures a 3- 1. INTRODUCTION During the period t96t-68 large -scale control of Simulium damnosum was undertaken in the vicinity of the Kainji Dam construction site. This scheme has highlighted some of the difficulties of successfully controlling S. damnosum especially in large areas with complex drainage systems and in which onchocerciasis has impalred the physlcal and economic weIl-being of the population. Since impoundment in August 1968, a smaller scale routine control programme has been introduced i.n which advantage is taken of the changed circumstances arising from the formation of the Kainji Lake to minimize the use of insecticides for control purposes. Feasibility studies were carried out during the late 1950's for a hydro-electrlc project on the River Niger in the Bussa area. The decision to go ahead was a complex one which depended on the fact that development coutd confer great benefits additional to the supply of electricity. Paramount among these was improved navigation, increased supplies of fish' protein and possibilities for irrigation being of lesser importance. Consequences to the health of the population owing to changes 1n the incidence of vector borne diseases were not considered. The consultant engineers sought advice on the necessity for vector control because: (a) they were aware that several major development projects in tropical areas had required insect controt operations for their successful completion. Apart from the classic case of the fight agalnst the mosquito vector of yelIow fever in Panama, which is weII known to civil engineers, the consultants at Kainji, with their considerable Afrrcan experience, were aware that tsetse flies had been controlled at Kariba and that controf of S. damnosum at Owen Fa1ls on the Victoria Nile had been of ma jor importance for the dam building operation. (Simulium control on the VoIta River at Akosombo arose out of the work at Kainjr.) (b) Reports by Drs Budden (1956) and Thomson (1959) had ind:.cated that the onchocerci-asis infection rate was very high in the Katnji-Wawa-Bussa area. (c) Three experts engaged on the feasibility studies had contracted the 61ss6se after only a few months exposure in the area. The engineers consequently asked Dr Waddy to vj.sit the area and report on the health aspects. Acting on Dr Waddy's advice Mr J. P. McMahon and Mr H. H. Goiny were appointed in March 1961, their brief being to establish a Vector Control Unit to keep the numbers of S. damnosum to such a level that: (a) work on the dam would not be interrupted or become hazardous owing to excessive biting nursance; (b) the project area would be sufficiently healthy and pleasant to live and work in for the costs of recruitment and retention of suitable key personnel to remain within reasonable Ilmits; (c) the labour force, especially the indigenous itj.nerant workers would not contract the disease.l 1 At the time it was feared that a heavily rnfected.work force, when disbanded on completion of the construction programme, might facilitate the spread of onchocerciasis to new areas. More recent information on the disease in West Africa suggests that tt occurs alrnost throughout the whole range of the vector (Gdckel, I966; Nnochi-rj-, 1964). Nevertheless there probably are isolated foci of S. damnosum in rnan-made habitats, especially at the norlnern edge of the range of the vector, in which the disease is absent (compare distrlbution maps in Budden, 1956, and Crosskey, 1960). Increasing civil engineering works are 1lke1y to increase the number of such sites. a 4- 2 2 DESCRIPTION OF THE KAINJI CONTROL AREA I Location The Kainji control area (Fig. I) was initially centred on the proposed Dam Site(g.53'N, 4.35'E) and covered an area within 15 miles (24 km) radius. In 1962 this was increased to 30 miLes (48 km) and later expanded to include the larger rivers to the south. However, the primary aim remained the protection of the Wawa-New Bussa-Kainji-Awuru district, ca. 3OO square rniles (775 krr2) in which the bulk of the work force lived and operated. Distant larviciding was intended to prevent the possibility of flies migratj-ng into the heart of the control area and also to provide residual protection to workers involved in road building, resettlement and navigational rmprovements. Most of the control area, Borgu District, Iies to the west of the River Niger and consists of slightly undulating country between 5oo-looo ft (I5O-3OO metres) with none over 2OOO ft (600 metres) above sea Ievel. Inselberg outcrops, an outstanding feature of much of the West African savanna' are virtually absent. 2.2 Vegetation and farming The entire district is classified by Keay (1953) as Northern Guinea Savanna but according to Ctayton (1957) the vegetation of the Nrger valley south of Awuru is Southern Guinea Savanna and that of the area north of a tine strrking approximately north-east through Dekala and across the northern tip of Foge Island Sub-Sudan savanna. Borgu is one of the most sparsely populated areas of Nigerla and apart from areas along the Niger and its northern border most of its vegetation is undisturbed by settlement or agriculture, though fires caused by man sweep through much of the region every dry season. Primary habttats have not been complicated by human actrvity as in much of the rest of the country and over hundreds of souare mtles Glossina morsitans submorsitans abounds. In 1ow-Iying sandstone areas and particularly rn tfr" Xo.rtugora-Maingy-ari district vegetation is dense with numerous thickets of @U!3_ sudanic.a etc. in wetter places. Such areas are known localLy as "kurimls" and tn"[E.rn *r."y-1,ftab1e habitats for Glossrna pa1pa1is. AIong the Nrger between Garaf ini and Bussa there were, before rmpound*u.rq "o.,aiderable stretches of well grown frrnging forest' Several species of trees were common and the tralling leaves of SvzvEium guineaeusis 1n particular provided the primary attachment sites for the aquatic stages of S. damnosum. AwaY from the rivers and areas of sandstone the vegetation is generally fairl y open savanna with patches of well grown "orchard woodland" rncluding almost pure stands of Isoberlinia doka a characteristic Northern Guinea savanna tree, and a preferred habitat o f G. morsitans. Farming was intensive only in the now rnundated alluvial- soils by the Nr ger River. South of Bussa away from the river the human population j's very low and cultiva tion confined to fairly narrow strips along the communrcatron routes' The main croP is Guinea corn, though miI1et, maize, cow pea' and groundnuts are also grown' Rice cultrvation is Practised in wetter places and root crops are produced near the vrI lages. Oil Palm and Borassus occur alongtheNiger,andbananasaregrownextensivelyatShaffini. Shea nuts and locust beans are harvested from wild trees and such economically valuable tree s are genera1IY left during the "slash and. burn" process of clearing new ground for the shifting cultivation. North of Bussa the picture is sim:.lar in essence but lncreasing population reduces the length of time land lies fallow and, at Ieast near the vlllages and main roads, there is littLe well grown timber. 2.3 Geology and drainage Most of the control area 1ies on the pre-Cambrian Basement Comprex of ord metamorphosed rocks. Sandstone sediments outcrop in the north west and south east of the area. The basic geo1ogygreatIyaffectstheSuitabi1ityoftheareafo"9=-@andindeedasCrosskey (1956) has pointed out the distribution of this species is probabty governed by geology more tiran by any other factor. Over the basement complex the rivers are rocky and fast flowrng' The sorl has little water holdrng capacity and extreme spate conditions are frequent' Flow -5- ceases during two or more months of the dry season. On sandstone the rivers tend to be deep and sluggish with much less tendency to spating and they are much more Iikely to be perennial. The drainage system is dominated by the River Nlger which flows roughly north to south through the area. It fLows over gneisses from Yelwa to north Foge Island thence splitting into two arms which embrace the island, a region of fertile aIluvlum. From Bussa south- wards to Bajlbo the river again flows across the basement complex into whlch it has cut a deep rock watled channel frequently divided by chains of long narrow islands. The flow of the Niger vaiies greatty at different times of the year. Figure 2 gives an idealised discharge curve for the river at Kainji based on 25 y."."t discharge measurements at Jebba, suitably corrected. The river is subject to two periods of flood each year, a sharp peak in the wet season known as the white flood or "farin ruwa" owing to the colour of the river carrying a large amount of sitt and the much lower btack fIood, or *bakin ruwa". This flood owes its existence to the release of the water of the previous wet season hitherto trapped in the swamps of the inundation zone of the Upper Niger. This water is much less turbid than that of the white flood. Minimum water levels are usually reached in June or July and at this time the Niger north of Kainji is not very suitable for S. damnosum and although a little breeding persists, S. griseicotle is much more numerous at this time. In addition to these seasonal variations there may be striking differences from one year to the next. Such differences had a marked effect on the costs of larvicide in the dj.fferent years (Tab1e 1). The basement complex to the west is drained by several major rivers flowing from the vicinity of the Dahomey border eastwards towards the Niger, these include the Menai, Timu, Doro, Wuruma and Moshi. The most important, the OIi, rises weII inside Dahomey. To the north the Swashi and Kpan are affected by the sandstone and tend to be perennial. A11 these rivers are very suitable for S. damnosum breedinB. Numerous smaller streams flow directly into the Niger or feed one of the principal tributaries. To the east of the Niger the only river Eystem of major importance to the control programme is that of the Kontagora which, being influenced by the sandstone, is perennial. Over much of the eastern side streams are not suitabLe for S. damnosum breeding. 2.4 Climate The climate is strictly seasonal with a dry season which usually extends from mid-October to early March, the peak season of ralnfall being between July and September. A summary of the rainfall figures is given for the region in Table 2. Annual variations are brought about by apparently non-periodic changes in the depth of penetration into West Africa of the inter- tropical zone of convergence. Rainfall is thus inconstant in quantity and pattern and in some years tends to be double peaked, in others single peaked, and sometimes there is an even distribution without a noticeable peak, but with the slgnificant rainy perlod being Ionger than usual. A heavy wet season has three main adverse effects on the Simulium control programme: (a) the number of streams likely to prove suitable for breeding S. damnosum is lncreased; (b) increased discharge rates require larger quantitles of insecticide; (c) travel to the treatment points may become very difflcult. Conversely, high flow rates help to: (i) ensure satisfactory passage of larvicide over considerable distances; (ii) increase the turbidity of the run-off with the probable result that the effectiveness of the larviclde is enhanced (Fredeen, 1962). In general the diurnal ranges of temperature and humidity are of greater magnitude than the variations rn the annual monthly means. The importance of the temperature and humidity Ievels lies in the fact that during the dry season conditions are too harsh for the fIy to -6 travel widely through the bush and at this tlme troublesome blting activity is confined to shel-tered areas cLose to perennial breeding sltes. 2.5 Population As Buchanan & Pugh (1955) have shown much of the "Middle Belt" of Nigeria has lower than average population and a fragmented tribal structure. The reasons are probably mainly polltico-historlcal but it is notable that vector-borne diseases are at their worst in the area. According to the 196O census the population of the Borgu divlsion in which most of the control area lies, was 81 942. These people are by no means evenly distributed, the northern part, Agwarra district, having about 38 persons per square mile (15 per km2), whereas the Bussa area has nine persons per square mile (3.5 per xm2) (rtlnkenberg, 1965). South of the Oli River the population is undoubtedly much lower. These figures compare with an over-all density for Nigeria of over 60 persons per square mile (23 per km2) in 1952' Most of the populatlon before the impoundment of the lake lived on the riverine strip or along the Kaiama-Wawa-Bussa-Agwarra road. This laterite road provides the only reasonable land communication route in the region, though there are a few other bush tracks whrch are motorable in the dry season. West of this road to the Dahomey border much of the land is totally uninhabited and it is rn this area that the 1460 square mile (3770 kn2) Central Borgu Game Reserve was estabLished in the earl-y lg60's, This was achieved without the necessity of resettling a single family. As already mentioned the tribal picture as in most of the Middle Belt iS somewhat complex, the most numerous group are the local Bussawa who are mainly engaged in arable farming, hunting and simple crafts. other sizeable groups are the Kamberis, a pagan tribe farming the higher tand, growing Guinea corn, m111et and beniseed; the Fulani as elsewhere predominantly semi-nomadic herdsmen; the Hausas mainly engaged in trading activities and the Nupe who penetrate up the Niger engaging in frshing and handicrafts' Along the Niger are a number of very small distinct groups who are engaged in fishing but also in farming some of the rrch alluvial sorl which under primitive systems of irrigation produces valuable crops of onions, rice, tomatoes, etc. Throughout the region fish and "bush meat" form important items in the diet of the people. 2.6 Resettlement Within the control area about 13 OOO people had to be resettled as a result of the formation of the Kainji Lake. The divisional capital Bussa was moved from a riverside site to an intand position within a few miles of the Dam and about 40 miles (64 km) from its original position. Most other villages have been moved away from the projected lake shore to higher ground without altering their relative posrtions' though there has been some amalgamation of small hanlete. The resettlement schcme, admirable in many respects' was carried out wlthout any regard to the public health aspects of the matter except at New Bussa where a water-borne sewage system was incorporated fotlowing the advice of Dr B. B' waddy' This has proved a great success. 3. THE PREVALENCE AND ECONOMIC IMPORTANCE OF ONCHOCERCIASIS 3. t Prevalence Budden (1956) in his maJor survey in Northern Nigeria has shown that onchocerciasis is endemic throughout the Kainjl control area. Indeed of the 11 most heavily infected riverine areas recorded by him, three lie with:.n it (Nrger-Bussa, Menai and OIr) and cover 2600 of the 6300 square mj.les (6716 of 16 273 km2) involved and 66 ooo of the 158 ooo peopte most seriously at risk. of this group Budden calculated that 5.7 per cent. were blind as a result ofonchocerciasis-aconclusionhebasedonastudyindepthofseveralcarefullyselected viltages including Bussa itself. In 1959 Thomson visited the wawa district and an examination of 27O persons revealed a 49 per cent. infection rate' 26 per cent' with nodules but only 3.4 per cent. with severe ocular symptoms. In 1963 a survey was carried out in the olr valley by Mr D. A. Danielson (Peace corps) working on behalf of the control Unit' He took a single skin snip from the forearm of each person and also subjected them to a cursory 7- examination for nodules. He made no attempt to assess the amount of blindness. The results by age-group are shown in Tab1e 3. Of the total of 844 people examined from alL age-groups 40.7 per cent. were proved positive with dermal embryos or onchocercal nodules, the proportlon of cases wlth nodules but having negative skin snips was 7.6 per cent. Clearly with multipte skin biopsies at more suitabte sites on the body the number of people found infected would have been considerably increased. Even with the simple method used 74 per cent. of alL persons over the age of 20 years in TamalaL (the nearest village to New Bussa) were found to be lnfected, whereas the figure for Wawa, which by this time had a substantial immigrant populatlon, was 30 per cent. In 1967/68 a more extensive survey was undertaken. In this two sktn snips were taken from each person, one from the hip region and the other from the lower Ieg, skin condition and presence of nodules were also recorded. Some older people who produced negative skin snips and did not present nodules were given an average 30 mg diethylcarbamazine - the Mazotti test. A person was judged blind if he could not distingrlish with one or both eyes the number of fingers held up by the investigator at a distance of three metres. People.whose blindness was obviously not the result of onchocerciasis were omltted but no claim j.s made that all the blind cases included resulted from the disease, though the majority probably did so. (See Budden, 1956, for figures for non-onchocercal areas.) Whole populations of the villages surveyed were examined as far as possible, though many of the under fives were missed and those who were examined usually onfy provided one skin snip. The basic results of thrs extensive survey are given in Tables 4-6. Over-aII prevalence was ca. 54 per cent. and in nearly every village between 45-65 per cent. There was surprisingly littIe difference between vr.Ilages sited cfose to notorious Simulium breeding pfaces and those 3-4 miles away. However, the percentages of bIindne." affip-p-.rt to vary in this respect. Thus in the Niger West Bank/OIi Va11ey district, the Niger East Bank/Kpatachi Island /to*", Kontagora-Maingyara district and the Auna distrj.ct (east of theNiger and north of the Kontagora) the percentages were 9.O, 8.4 and 3.3 respectively. Thus in the area least favourable to the vector although the onchocerclasis infectivity rate was as high as elsewhere the blindness rate was much lower. This blindness rate of 3.3 per cent. for a predominantly non-riverine population is remarkably srmilar to Thomson's(1959) figure for the Wawa district (aIso a non-riverine area). 3.2 Economic importance Doubts have been expressed about the economic importance of onchocerciasis but Budden(1956) showed how the areas of Northern Nigeria most severely affected by onchocerciasis were also among the most underpopulated, and such areas included not only rocky land but also fertile valleys, €.8. the Zamfara of Sokoto. Waddy (1969) has pointed out that in areas of severe onchocerciasis the human population tends to move out of the river vall-eys and settle on the higher ground aiding the spread of sorl erosion and increaslng the Simulium breeding potential of streams, thus ensuring their contrnued discomfort. From the Borgu surveys it is clear that onchocerciasis presents a major public health problem and therej-s strong evidence for its economic importance in the area. The population pattern in Borgu(at feast as regards the tributary rivers) resembles that described by l{addy for the Red and White Vo1ta Rivers in northern Chana, well watered river valleys tend to be deserted and heavy settlement is mainly oonfi.ned to higher ground in the north of the district. That this desertion of riverside sites may definitely occur as the result of excessive blindness is shown by the village of Yangba which is situated beside a notorious breeding rapid on the Ri-ver OIi. During the 1967 survey virtually all the adults were found to have onchocerciasis and of the population of 13O, 2I were btind (16 per cent.). This village was in a very depressed state with an unbalanced population and with many deserted and decaying huts. In 1969 the headmen decided to abandon the site completely owing to the amount of blj.ndness (pers. comm. Dr B. B. Waddy). The lower Kontagora vafley provides a more striking example. The river and its main tributary, the Maingyara, are perennial providing abundant c1ear, cool water throughout the dry season. This water is highly prized locally as a cure for Guinea worm. (Doubtless the settlements at Shaffini etc. are known to be free from the infection, and this is attributed to the quality of the water rather than to its constant flow. ) Apart from being well watered both valleys are known -8- to be quite fertile and suitable for the local forms of agriculture (pers. comm. Mr P. Mansfield, former Agricultural Officer). The lowest 1O miles of the Kontagora and five miles of the Maingyara ftowing across crystalline basement rocks are ideally sulted to the needs of S. damnosum. Further upstream, the area is one of sandstone debrls through which the rivers flow slowly in deep channels thus being in the main un that until very recently there was only one very smaI1 In 1968 this village, Dalau' had a population of 27 of suitable fo" [!g!!g. It is signif icant village situated in the lower reaches. whom 16 had onchocerciasis and four were btind, including the headman. The village is clearly in a bad state and durlng the period t96l-64 is known to have changed its position four times without avail. Since that time Simulium control operations have been very successful in the valley and these villagers have had some respite from the attacks of the fly. In addition a large new settlement (sabon Gari) began to develop in 1968 at the new Kainji-Mokwa road bridge. The inhabitants very successfurry farm the river banks obtainlng high yietds with primitive irrigation methods' Figure 3 shows the settlement situation in the area compared with the riverine system and the basic geology. It is Worth noting that some of the villages' e'g' Bambafu, have very precarious water supplies which sometimes fait during the dry season. This results in great hardship for the women who may have to travel as much as five miles to obtain supplies from the deserted river banks. Although the Val-Iey was once devastated by the slave raiding of Ibrahim Nag:wamatse, Emir of Kontagora (Muffett, I964) and rs the haunt of three species of tsetse flres' these factors seem very unlikely to be of importance in determining the present pattern of human settlement. Slave raiding ceased about the turn of the century and there do not appear to have been any maJor outbreaks of sleeping srckness in the last thirty or forty years' However, until the recent years of gmuli,u.m control onchocerciasis has been freely transmitted throughout the year by very tr"g. tI pop-.,fations, and the surrounding villages' even those situated away from the rivers, have suffered a heavy burden of disease and misery (Tabre 7)' The coincj.dence of the pattern of human settrement and the suitabitity of the rivers for the vector of onchocerciasis together with the new colonrzatjon of the varley during the period ofcontroloperationspointsfirmlytotheconclusionthatS.damnos'mandonchocerciasiswere instrumental in keeping this fertile va1ley uninhabited, if not in causing the original depopulation. It is worth recording that at least 60 square miles of fertile' well watered land remains idte at the present tLme' 4. DISTRIBUTION AND PREVALENCE OF THE VECTOR 4.r Earlier surveyS Simpson(I912)recordedman-bltingsimuliidsfromseverallocalitiesinNorthernNigerra, including Bussa, which almost certainty referred to s' damnosum but i: was not until I944 that this fly was first positively identrfied from materiaitottected in the Kudaru area by Dr T. A. M. Nash. There were no further developments until an entomological survey was undertaken by crosskey (1956) between 1952 and 1955. Crosskey showed that S' damnosum was a widespread and voracious man-biter and undoubtedly the main, rf not the only vector' of onchocerciasis in Northern Nigeria. He did however, record one other species "f EBlt"* biting man. This fly, s. bovr.s was somethrng of a pest in part of Abuja Emirate and several dissected specimens contarned developing filarial worms which crosskey found indistinguishable from onchocerca volvulus (crosskey, I95?a). Among the districts which crosskey found to be heavityinfestedwiths.damnosumwastheNigerValleybetweenJebbaandBussa,theKontagora andtheMalendorivers,resultSwhichparallelthefindingsofBudden.Hedrdnotpenetrate the hinterland of Borgu. 4,2 Control unit su rveys rn 1g6r McMahon and Goiny began thelr preliminary surveys in the Kainji area concentrating atfirstwithinfifteenmilesofthedamsiteandlaterextendingtothirtymlles.Theyalso lnvestigated the proposed line of the Kainii-Mokwa road' This survey work and subsequent studies have shown that the whole of the district lyrng on the basement complex is heavily infested with q. damnosu.m during the rainy season. The sandstone part, $/hilst being less suitable generall,y, provides most of the perennial breeding sites' Although' contrary to the 9- findings of Crosskey (1957b), S. damnosum has been found in very sma1l streams flowing at rates of less than two cusecs, there is no evidence to support the suggestion by Nnochirl(1964) that it may breed in sluggish, muddy streams. Nevertheless despite its abrlity to breed in sma1l streams, especially notable at the end of the rains when rivers are beginning to dry up, S. damnosum remains primarily a large river species. The f1y is welf known to the locaf people who accord it a variety of names. These names are Kusna at Bussa, Kunjl at Yangba and Makurgi at Awaru (pers. comm. S. A. Oyewole). They do not associate the bite with any disease and are sceptical of the explAnation that the fly's bite and the itching disease are in any way connected. Dlstributlonal information for the area is shown in Figure 4. Dry season survival in the area seems to be nunerous sites. The main period of breeding on March and breeding continues throughout the year rivers. Breeding may also take place for about Swashi and Malendo rivers. so1ely by means of perennial breeding at the Niger River occurs between December and on the Kontagora, Niger, Kpan and Bunsuru lO months of the year on the OIi, Menai, 4,3 Other species of Simulium During surveys of breeding places 17 other species of Simulium have been collected. These are alcocki, duodec imum diallonense, jgllrc, schoutedeni lmoukane. mcmahoni -' ::=:.::'cervicornutum, unicornutum ruficorne hirsutum, adersi, griseicolle, tridens hargreavesi and bovis, nearly all of which have been collected in conjunction with S. damnosum at some time. There is, however, no close tink between S. damnosum and any other species though thepresence of hargreavesi, cervicornutum and the large river species griseicolle and tridens indi.cates In general that a stream is likeIy to be suitable for damnosum, at least in its swlfter parts. alcocki. and its allies are rarely found with damnosum being typical of smaII, sluggish, muddy, and in some cases, poIIuted, streams. 4.4 Prevalence of adult S. damnosum Unfortunately there are no pre-control fly-catching data for the region. FIy rounds were set up in 1961 but catches were doubtless affected by experimentat 1arviciding operations. During the wet season the fly is found throughout the region in large numbers, though there is little doubt that most of the population remains cLose to the river banks and appreciable cross- country dispersal usually takes ptace only in the late wet season. At that tj.me very highfIy numbers may be found in completely uninhabited areas and this gives rise to speculation as to the preferred host in these circumstances. In many parts of the world Simuliidae are rated as important pests owing to their occurrence in vast swarms whose bites can have serious consequences for man and his lrvestock. Such outbreaks are well-known in Canada, Germany, Russia, and severaf other European countries, but rn Africa the only j.nstances hitherto reported are from the Sudan where the culprit is S. griselcolle. This fly breeds in vast numbers in the River Nile, and causes intense annoyance by crawling about the face. Though only a smalrpercentage bite, these are nevertheless very numerous (L,ewis, tg4g). Deaths of donkeys are reported by King (1923) and of turkeys by Garside & Darling (f951). In August 196g f1y numbers south of the OIi River reached unprecedented heights and using a man-baited skirt trap a landing rate of 3350 per man hour recorded on the 9th. This appears to be the greatest number S. damnosum ever taken and at this density of insects lifewas very unpleasant. Apartfrom S. damnosum no other species of Simulium was taken biting man in the Kai.nji area but .@, griseicolle, and .!]-$1]1141 were all taken on man. Special catches were performed at known S.. bovis breeding sites without success. Details of adult catches and catching methods are given in Section 7. TREATME}TTS, MATERIALS AND METHODS .1 Choice of larviciding Right from the outset it was decided to undertake larviciding procedures as the sole method of control. Thls decision was reached in the light of the knowledge of the success of this method in Kenya agaj.nst S. neqvei (McMahon et aI., tg5g) and in Uganda on the Victorla Nile 5 5 10- again st S. damnosum (Barnley, 1953) and more particularly at Abuja, Northern Nigeria (Davies flies precfuded anY schemeet a1., 1962). Lack of knowledge about the behavlour of the adult based on an attack on this stage and a cautionary note had been struck by the failure of the Mayo Kebbi, Chad scheme which had relied heavily on adulticiding (Taufflieb, 1956). 5.2 Choice of insecticide Throughout the period of control a 25 per cent. emulsifiable concentrate of DDT (Shell Arkotine D.25) was used. DDT in this formulation was chosen for trails in 1961 for the following reasons: (i) in the successful control- schemes al-ready mentioned DDT had been used; (il) a similar DDT formulation had proved quite effectlve and easy to handle at Abuia; (iii) DDT in this formutation was known to be exceptionally safe to m,.,;1 (iv) the insecticrde was readily available in Iarge quantities and the manufacturer undertook to deliver to the headquarters of the control unit. Larviciding trials carried out on the Niger, 01i, Kontagora and Maingyara were highly successful and throughout the perlod of control there was no reason to doubt the suitability of the chosen formulation except when some Iessening of susceptibility may have developed. Even at the present time such an insectlcide stilt seems, to the author, to be the best available product for Efig control with the sole possible exception of methoxychlor. A suitable non-chlorinatea fryarocarbon insecticide is a greatly needed alternative in case serious resistance develops in expo sed S. damnosum PoPulations. 5.3 Concentration of larvicide Treatment rates ranged from O.O33 ppm in the Nlger to 2.O ppm in some of the very smal1 streams. The amount of insecticide required was based on the standard thirty minute dose and calculated as described by McMahon et al. (1958). Normal delivery was by simple pouring from four or five gaIIon contarners, which wi-th a littIe experience can readily be emptied through thelr openi.ngs in 1O-I5 minutes without recourse to punching hotes in their bottoms etc. For small doses the insecticide was mixed with water before pouring. When possible suitable bridges or overhanging trees were utilized to ensure satisfactory spread of the insec- ticide across the breadth of the stream and the importance of spreading the insecticide was considered when treatment points were being selected. On larger rivers such as the Moshi, 01i and Nono use was made of dosrng devices constructed from scraps of water-piping etc. scavenged from the Dam constructlon site. In essence these devices consisted of a wire rope slung across the river ancl fastened high up rn convenient trees, from this was suspended a piece of metal piping vtfZ to 2 rnches in diameter which was welded to a 46 gallon oil drum at one end, the other end being closed. At intervals atong the pipe holes were drilled through which the insecticide could pour. It was necessary to ensure that the oil drum was firmty fi-xed well up the river bank, usually on a welded trrpod, weII clear of any possible flood water. A stop tap was incorporated at the joint of the pipe and the oil drum' The amount of insecticide required was ir-rtroduced into the oil drum, the top of which had been I Absorute safety of insecticide is vital when itliterate labour is being used' as it is extremely difficult to instil into such personnel adequate appreciation of the need to treat dangerous substances with sufficient caution. (To illustrate this point 'it may be mentioned thatthelocal"prdgin"forinsectrcideis"medicine"andthatonmorethanoneoccasionit was necess aty to dissuade workmen from using the insecticide as a disinfectant on some suppurating sore.) In addltion insecticide is likely to be stolen from stocks which for logistic reasons are sometimes left in the bush, for use during the rainy season' This pilfered lnsecticide is used in fishing operations' - 11 cut out, and the whole made up to 46 gallons, the tap was opened and the insecticide poured out through numerous holes, which dispensed the contents of the drum in about 20 to 30 minutes. These unsophisticated devices were very useful and easily produced given the raw materials. Care had to be taken to wash the apparatus out after use to prevent the holes becoming blocked up with crytattized insecticide. On the Niger treatments were always made from two or three large dug-out canoes and a single dug-out was used on the Kontagora. For routine treatments of small streams the amount of flow was usually estimated from experi-ence and consequentty the actual rate of larviciding was also an estimate. It was not considered worth while to spend the time necessary to gauge streams correctly and calculate the:.r discharge rates on every occasion. The situation was different on major rivers which were in any case beyond the capability of the Unit to measure. Fortunately these rivers had permanent gauges, and calibrated discharge curves were availabte to the entomologist. 5.4 Control season Owing to the difficulties of treating streams during the dry season, the size and complexity of the river systems involved, and their non-isolated nature, total control by dry season treatments was not considered feasible. Thus control was basically designed to start about the beginning of JuIy and to cease when the fly population began its rapid natural dectine (usualIy late October). However, some dry season tarviciding was carried out for two reasons: (a) to protect the workers on the dam site during the height of the construction period (1964-67) it was necessary to treat the Niger and the Kontagora system whenever f1y numbers in the site area showed signs of a sharp increasel (b) in addition to the above, several attempts were made to delay the onset of heavy wet season breeding by control in the Iate dry-early wet season period. 5.5 Larviciding cycles Larviciding was normally carried out once every IO days throughout the wet season, this being reduced to once a fortnight when control was exceptionally good. The cost of treating the Rlver Niger prohibited a tOday cycle and for this river either a Lbday cycle was used or Iarviciding was spasmodic and carried out only when deemed absolutely necessary. (The entomologist was alded by the fact that during the height of the wet season the Niger water levels were rlsing or falling so sharply that at this time the river was not a major producer "f s!g!!g. ) Each riverine system was treated as a unit and each larviciding cycle waseither: (a) carried out on one day, or (b) tributary streams and the upper reaches of the main river were treated on the first day and the lower reaches on the second day. In addition, aIl treatments in each fulI cycle were carried out within the shortest possible time which was usually three days. 5.6 Treatment points During the nine years of control, well over 60 treatment points have been used, some experimental, and some replacing earlier points which were found to be badly sited. Fj.gures 5-7 show the points in general use throughout from 7962-68. The points used in the wider area to protect the control zone from possible wind-borne fLies, and fina1ly, the 1969 control points which are the basis of future routine work. At the peak of control, 1964-66, over 50 points were in use each year on as many as 15 major rivers and 20 streams. After 1966 an attempt was made to reduce control but this was frustrated to some extent by lack of knowledge of the distrlbution pattern of the fly, and by the exceptionally heavy rains of 1968. Followj.ng -12- 6 6 range of the fly was about 10 miles (Davies et aI., 1962) and that breeding was confined to fairly large rivers. Thus after trials in 1961, control wa the impoundment of the laker in August 1968 no further treatments have been carried out to rivers entering the lake and at present treatments are only bej.ng made regularly at t2 points and less frequently at a further seven points (FfS. ?). In addition treatments of the lower OII at the start of the wet season may be very complicated and involve applying one or two gallons of insecticide every two to three miles along a 16-mi1e stretch of river. Control of the Niger i.s by means of atterations in the river level. Despite these changes the nucleus of control has remained in the control of breeding on the Niger, the Oli river system and the Kontagora river system. 5.7 Quantities of insecticide used The quantities of insecticide used annually have varied enormously with the type of approach to control and the state of the major rivers. Figures are given 1n Table 8. ANNUAL TREATI\MNTS t Wet season control InitiaIIy, plans for larviciding were based on the betief that the effective dispersal of S. damnosum s introduced within 15 miles of the Kainji Dam site. Later this was increased to withrn 3O miles and control extended to the head waters of the Kontagora. Major breeding rivers even more dlstant were incruded as a result of fears of wind-borne fries entering the control zone. From 1962 the programme was reduced as this hypothesis had been discarded. By 1968 the emphasis of the programme had shifted from the protection of workers on the dam site to the protection of the township and in 1969 a routine programme which could be operated without the presence of a specialist entomologist was introduced. 6.2 Drv season control Treatments of the Niger and Kontagora in attempts to eliminate the dry season population were made in 1962, 1964 and 1966. In 1964, 1965 and 1967 control had to be imposed during the dry season when fly populations at the dam site began to rlse. In other years rivers were treated during the wet season from the time they began to show a build-uP in breeding. The dry season attempts to etiminate the fly populations failed, though control for the benefit of the dam site workers was relatively simpte. During the dry season great drfficulty was experienced j.n treating rivers at a sufficient number of points to ensure that atI breeding was definitely extlnguished, and even when this was achreved it seemed to have no effect on the development of the wet season fIy population. Details of conditions obtal"ning in different years are set out in Table 8, together with a broad outline of the scope and success of the control measures. Figures 5-7 also indicate the scope of control. b.J Factors adversel af fect control TWo major factors affected the success and cost of the control programme. Firstly' the lack of a road network and adequate contour maps made it necessary to construct and maintain many miles of bush track whrch woutd remain passable to four-wheel drive vehicles at the height of the rains, and also footpaths and bicycle tracks. Owing to inadequate maps these tracks were sometimes badly positioned, and several had to be abandoned and reconstructed at more suitable sites as these were discovered. Surveying the area by trekking along river beds and across watersheds and constructing the necessary tracks was very expensive and consumed a great deal of the energies of the Unit. Fortunately, as the tracks "aged" they became more reliabte and less repair work needed doing. Nevertheless' falfure to reach far- ftung but vitar treatment points had serious repercussions on occasion. Thus in 1969 when the area NNW of Kaiama was extensively flooded, and the track to the Lower Nono treatment point became totally impassable, this had a disastrous effect on the fly numbers in that region, which soon spread to the townshrp area. -13 Secondly, lack of knowledge concerning the method of, distribution of the f1y, and its effective and maximum flight ranges caused doubts about the area throughout which control must be extended to protect fu1Iy the central area. Furthermore, the degree of control of the f1y necessary to ensure that transmission of Onchocerca volvulus ceases i.s not known. 6.4 Annual summaries 196I: Larviciding trj.als were carried out which showed that the chosen insecticide was very satisfactory, being tethal to S. dqmnoqgm at very low concentrations. The results of the trials are shown in Table 9 19622 Treatments began on the basis of control within 15 miles (24 kn) of the dam site. Larviciding was carried out on the Niger, Kontagora-Maingyara, 01i and some smaIl streams close to the New Bussa sj.te. In Septenber owing to very high fly numbers, control was extended to within 3O mites (48 km) of Kainji but it was not possibte at that late stage to carry out all the desired work. 1963: Control was carried out within 30 miles (48 km) of Kainji and several J-arge rivers further afield were also treated (see Fig. 6). 1964: Control was similar to that in 1963, though rather more attention was paid to smaller streams. 1965: Control was planned as for 1963-1964 but conditions were exceptionally bad and the entomologist was absent throughout June and July when heavy breeding had already begun to develop. Travel was extremely difficult and many treatment points could onty be reached by Iong and arduous marches over swampy ground and through tangled bush, 1966: Control was continued as in earlier years and was placed on a purely routine basis when the entomologist became seriously ill at the height of the wet season. 1967: The hypothesis of wind assisted migration had been tentatively abandoned and after the opening phase of the season, treatments were more restricted than in previous years. 1968: Rainfall in this year was quite exceptional, with several inches in Aprll (in the Kaiama district over 25 inches had fallen by 2L June), O1i reached unprecedented levels and exceeded 5OO cumecs for a month. Before impoundment of the take (3 August) dosing was based on the 3O-miIe radius scheme. Following impoundment all rivers discharging into the lake were omitted from the programme. Other planned reductions in the programme were abandoned owing to the exceptional conditions, and precautionary control of the Moshi and Ufa was reintroduced. Nevertheless tarvicidal expenditure was lower than in some earlier years. 1969: A Ilmited routine control scheme was introduced, designed primariLy to protect the township. Larviciding activities were concentrated in the 01i val1ey especially its southern tributaries. The Kontagora system and the Wuruma West continued to be treated(Fig. 7). Breeding on the Niger was controtted solely by flrrctuating the discharge leveIs from the lake. The intention was to kil1 eggs and larvae by lowering the water 1evel and causing them to become desiccated. The idea of flushing away the larvae was rejected as being less effective and too difficult from a dam operating point of view. By this means breeding on the Niger was completely extinguished. Control continues on the 1969 pattern under the direction of Mr S. A. Oyewole of the Niger Dams Authority. RESULTS I Aquatic stages Regular checks were made for breeding after most important Iarviciding operations, especially those of the Oli and Kontagora. Larval searches were simple on the Kontagora and could be relied upon to reflect the true situation. However, the OIi was very difflcult 7 7 t4- to search satisfactorily when in high flood. The Niger was always checked before and after larviciding, set collection stations marked out and each one visrted by canoe. Subjective index numbers of eggs, larvae and pupae present at each station were recorded. On some occasj-ons Iarvae on natural supports were placed in open ended tubes over which bolting silk was placed allowing free flow of water. Such tubes were also used in studies on length of life of the different aquati.c stages. Over the years the wet season tarviciding programme was entirely successful in that larval life on treated stretches of rivers was easily eliminated. Difficulties only arose in the early wet season when treatment points had to be very numerous owing to the intermittent flow of the rivers, and later when reaching the far flung treatment points was very difficult. Onty on the River Niger was the efficacy of the insecticide ever in doubt (see Annex f). Dry season work was more difficult, desplte much easier travelling conditions, as river's at that time consisted of large, near stagnant pools interspersed with short stretches of trickling water sufficient to support S. damnosum. Breeding sites on the O1r in January and February were often virtually hidden beneath boulders and were detected more by ear than sight. At low water the Niger also took on such an appearance and was no longer navlgable by canoe. Larviciding at such times required the spraying of each individual breed:-ng rapid, and satisfactory treatment from the ground was vrrtually impossible. The use of light aircraft for survey of possible breeding areas would have been of great value but much of the ground survey could not have been safely eliminated. 7.2 Adults FIy rounds were set up i.n I96I, though there had to be later modifications, especially owing to the changing nature of the dam site, clearance of vegetation in the reservoir zone and finally, impoundment. Initially fly rounds were based on a series of eight to 12 tube catches, of 15 minutes duration, with five minutes walking time between each catch. The catching team was normally four tube men and one recorder. In 1965 the skirt trap was introduced (Annex 2) and fly rounds were abandoned rn favour of statronary catches. Tube catches continued to be used for comparison wj.th earlier years and other schemes. Fly rounds and catching points are shown in Frgure 8. Catches were made regularly through the wet season rn the Kainji-NewBussaarea and addltional catches were made in the outer zone whenever possible. Unfortunately some of the details of the protection zone catches have been lost and the outer zone catches are rnsufficrently regular to be useful- for comparative purposes. A series of aII-day catches was made throughout the year at the Kontagora-Maingyara Junction' The results of these catches are dealt with rn Annex 3' Monthly biting rates for the Protection Zone and New Bussa township are shown in Figure 9. The histograms are based on tube collections only to make comparisons between years more vaIid. In Table 8 annual results are compared with rainfall figures and the states of the rivers, in parti.cular the OIi and the Niger. Without adequate baseline rnformatton, everything points to 196I berng a year of naturally low fly numbers and therefore it is not possible to pronounce on the degree of control achieved with any certainty' However, it is concluded that the type of control measures taken were adequate during years in which the oll River did not recerve much water from its upper reaches. There was no real evidence that treatments of the rivers far to the south or those much to the north of Kainji were vltal to the success of the control operations. Early fears of overland wind-borne movements of flies were not substantiated, but it was clear that movenents along river va11eys, especially in vatteys as large as the oli, coutd be greatly in excess of the 10 miles originally assumed. Overland movements were much more restricted and flres only became numerous i.n New Bussa several days after fly numbers in the lower oli valley, four miles away, had reached high levels. I5 8. SIZE AND COMPOSITION OF THE UNIT 8.1 Activitles and personnel Throughout, the Unit had over-atI responsibility for public h€elth mattcra including camp sanitation, water supplies, etc. In general this side of the Unit was dlrected by a health superlntendent who only took part in Simulium control during emergencies, e.g. illness of the entomologist. Clerical assistance was shared between the sections, and a lorry normally used by the heelth auperintendent was available to the Simulium Unit for carrying heavy loads. TVo mechanics who looked after the spray pumps, swingfogs, bicycles etc., were shared, as were the services of a senior staff transport nanager, during the periods of maximum activity. There was always an entomologist in charge, and in theory also an assistant, but 1n reality owing to leaves, sickness and difficulties in recruitment, on several occasions only one senior staff man was available to run the whole Unit. During the bulk of the control period the Unit operated with two permanent teams of six insect collectors each with a supervisor. These men carried out a1l f1y catching, Iarva1 searching, and larvicidlng. In the dry season they were engaged in repairing the bush tracks, carrying out drainage work for mosquito control and residual spraying for tsetse control. They were recruited locally and most were illiterate. No attempt was made to train them in the identj.fication of Simulium larvae and adults to species IeveI, though most of them could easily recogrrize S. damnosum, and also the pupae of several other species. The supervisors were not well educated, although by no means illiterate. They were conversant with simpte identification and recording procedures but were not given any laboratory training, therr marn qualities being knowledge of the loca1 terrain and the systems of tracks and paths, and ability to control the men. In addition to these collecting teams a gang of about 16 labourers with two headmcn was employed to construct and maintain the system of bush tracks which would remain passable to landrovers throughout the wet season and enable successful larviciding and inspection of the rivers. At its maximum the Unit had in use about 286 miles (460 km) of track which it had constructed, of these 119 m1les (191 km) were located in the OIi Valley. In 1969 when 126 miles (2O3 km) of tracks constructed by the Unit were stil1 in use the whole Simullum operation was carried out by six insect collectors and the two supervisors. Unfortunately this drastic reduction in numbers meant a marked curtailnent of the catchlng programme. A,2 Transport Transport of the Unit consisted of landrovers, bJ.cycles and one or two DKW mopeds, later replaced by BSA Bushmaster motorcycles. The road maintenance gang used a long wheel-based Iandrover pickup and the two collecting teams and the entomologist and assistant used stat:.on wagons. UsuaLly there were three or four vehicles, wlth drivers, avaj.lable at any one time. A Canadian tracked vehicle (a Muskeg Bombadier) intended for Arctic use was also available at trmes and this was used for clearing vegetation and regradlng tracks. It was afso used for travelling across boggy ground but was extremely unreliable and was usuall-y out of action. The Unit did not have its own river transport. 8.3 Costs During the dam construction period costs of the Simulium control operations were not kept separately from those of other services provided to the township. Vehicles were hired from the civil ongineerinB contractor as part of a general agreement between the contractor and consultants, on terms disadvantageous to the Unit. Undoubtedly costs were fairly high with insecticide, at about sterling tO.9 per gallon, and maintenance of the track system major items of expenditure. Simulium control activities during 1969 cost 17L66, excluding the salary of the entomologist. This figure was made up as follows: 16 (two landrovers - petrot and maintenance)Transport Labour: f3750 780 936 I500 200 two supervisors six insect collectors Material s 9. DISCUSSION 9. I Prelimj.nary investigations Experimental control operations were tried out in 1961, a year in which the OIi River was lower than normal and in which the rainfall pattern aided the controller. Clearly 1t wouLd have been much better if preliminary studies had been undertaken for at least two years before larviciding was attempted. This would also have provided baseline data on fIy numbers. IdealIy, before any Iarge-scaIe control scheme is undertaken several years data on rainfall and hydrological conditions should be obtained together with detailed entomological data from more than one year. 9.2 Dry season control In Borgu the short period during which major rivers cease to flow in the dry season and the multiplicity of perenniaL breeding sites, together with the large size of the rivers involved (especially the Niger), makes any attempt at dry season control or temporary eradication by ground based operations a very difficult proposition. Use of light aircraft might ease the logistic problems but it seems likely that the detection of the smaller breeding sites must remain the task of the entomologist on foot. 9.3 Wet season controf Wet season control, whilst presenting great logrstic problems and beinp costly in insectlcide, may offer the onty reasonable solution, especiatly in the Southern Guinea and Derived savanna districts. If long-lerm operations are envisaged, certain measures can be undertaken to ease the logistrc diffrculties and reduce costs. Initial construction of tracks to fairly high standards (aII tree roots carefully grubbed up, permanent drifts of rock and concrete, stones dumped in boggy places) wiII reduce travel problems and save on subsequent maintenance. Tracks thus constructed witL often, as happened in the lower Kontagora valley, become major routes for the passage of traders and market women, and in some cases the expense of track repairs may be shoul-dered by the villagers who benefit. Given adequate tracks and transport, ground control operations, even in the most difficult parts of the West African savanna, can be successful, though river systems which cut internatlonal boundarres (as the Oli) should be avoided. The ability of S. damnosum to breed in qulte sma1l streams makes aerial larviciding of doubtful utility. However, ln Iarge-scale operations the advantage of having a spraying aircraft would be considerable. Thus the exceptional conditions which prevented the treatment of the important Nono tributary and jeopardized control in 1969 could have been overcome wrth the use of an aircraft. A further advantage of aerial treatment would be the ease with which large rivers such as the OIi could be treated to ensure satisfactory dispersat of larvicide across their entire widths. Inltial hopes that control at Kainji would prove relatively cheap and simple were not substantrated. These hopes had been based on the apparent success of the Abuja scheme(Davies et a1., 1962) and the belief that effective flight range of the fly was about 1O show that although S. damnosum was generally unlj-kely to fly DDT and containers (ca. 1600 gallons) too1s, bicycle spares, cement, etc. mi1es. The Kainji results very far overland, movement exceeding 40 miles (64 km). rapid movement to a maximum movements of up to 40 km. along large river valleys was considerable, almost certainly This compares with the results of Le Berre (1966) who recorded of 41 km, and HaUserrnann (1969) who also reported riverine There j.s little doubt that the success of the control at Kainii- -L7- New Bussa depended upon control throughout the entire OIi catchment area, at feast in wetter years. Ear1y fears that wind-drifted flies might jeopardize control were not confirmed. However, there is some evidence that the pattern of fly movement altered in the late wet season, either in response to changing climatic conditions or to high populati.on density, and on occasion flies became more numerous away from their breeding sites than at them. From the parasitologj-caI findings it is clear that durlng the control period transmission of oncho- cerciasis has continued among the tocal population. The small numbers of flies to be found over four miles from a breeding sj-te were sufficient to act efficiently as vectors. Horvever the severity of individual infections and ocular effects may be reduced in such localitres. 9.4 Organization and equipment The success of large-scale control is primarily a question of loglstics. It is not reaIIy profltable to discuss the optimum area which can be adequately covered with particular numbers of men and items of equipment, as this will vary greatly according to the type of terrain, condi.tion of the existing road network, complexity of river systems, quality of available labour, and several other impondei'able factors. Thus a decision can only be made in the light of knowledge gained by surveying the proposed control area, and carrying out p11ot works. Nevertheless, it is possible to enumerate several points for consideration when the setting-up and equippingof a control unit is envisaged. No attempt has been made to evaluate the rote of aircraft, but the author feels strongly that this must be supplementary to that of adequate ground control operations and that, at least i-n the West Afrlcan Guinea Savanna, the aeroplane (or helicopter) cannot successfulty supplant the trekking entomologist, though it may weII make his task much easier. Based on experience at Kainji the followtng suggestions are offered (a) In very large-sca1e work the area should be split into zones whrch are manned by two entomologrsts, rather than having smaller zones with one entomologist in each. There are several reasons for this. Firstly, in the event of illness or accident dur:.ng the control season, there is stilt a man wrth expert knowledge of the area. Secondly, adequate leave can be taken without disrupting the smooth operating of the Control Unit. Thirdly, the presence of more than one person allows for cross fertiliza- tion of ideas and alleviates the real sense of intellectual isolation which can occur to the entomologist doing such work. (b) The work of the Simulium controller can be extremely arduous and in any scheme which is to continue for several years generous home leave allowance should be made of at least eight weeks per annum. In addition, opportunities to visit other Simulium areas for comparative work and discussions wrth co-workers should be available. (c) For very large operatlons a highly skilled sgnior staff transport and maintenance nanager should be employed. (d) The Unit should be equipped with four-wheel drive vehr.cles, and if used on longjourneys only properly upholstered statlon wagons should be considered. (The Long wheel-base Iandrover, IO seater-station sagon rs undoubtedly the most satisfactory vehicle for Simulium control work, ) (e) If possible vehicles in regular use on bush tracks should be withdrawn after two wet seasons. (f) Each vehicle shoufd have its own driver whose sole responsrbility is its driving, care and elementary maintenance. Supervisors and senior staff should not be expected to drive themselves long distances and then undertake supervision of control and survey work. -18- (g) Each vehicle should be fitted with a fuII length roof rack, a set of digging out tools, hydraulic jack, winch and rope, etc., as well aS two spare wheels, spare jerry cans and adequate frrst-aid kits. (Conslderation should be given tc the use of radio- equipped vehicles. These were used very successfulJ-y during Transmission Line survey work at Kainji.) (h) A ratlo of one vehicle to about lO-I2 workers should be reasonable, with extra transport (not ordinary motor cars) for senior staff. (i) Canoes and outboard motors should be part of the standard equipment of the Control Unit. A dinghy capable of being transported on the roof of a tandrover would be very useful. 9.5 Entomological lnvestigations Apart from difficulties 6f a logistic nature control by lack of knowledge concerning the biology of the fIy. need consideration are: of Simutium damnosum, is bedevilled The more important Po:.nts whlch (b) Resting sites: no satisfactory information exists concernlng resting behaviour. Such informatLon might tead to new control possibilities less likely to conflict wrth fishi.ng and conservation lnterests. In addition knowledge of resting sites would probably result in the coll,ection of engorged flies which might yield useful data on preferred hosts. (c) Aestivation and longevity: lack of knowledge on these topics seriously prejudices plans for dry season control which might be very successful in isolated foci north of ca. Il'30'N (at teast in Nigerta). Information on these fundamental polnts would probably be easiest to obtain in such northern foci. 1O. CONCLUSION Despite extremely adverse terrain, and several set-backs the Karnji control scheme achieved its basic aims in that: (a) at no tlme was a blting nuisance apparent at the Kainji Dam site' consequently work contlnued smoothly throughout each wet season; (b) only one of the senior staff was known to have contracted onchocerciasis (he had worked in Awuru for four Years); (c) the Large j.tinerant Nigerian work force was protected from the disease. This work force, now dispersed across the country, might have been instrumental in spreading or at least intensifying the dtsease in new areas; (a) Dispersion and flight range: in partlcular the fundamental reasons the f1y to disperse from its breeding sites, in addition to the distance this dispersal, need much more investigation. From Crosskey (1960) and it would appear that there are several isolated foci of Ljlry d"tp Savanna of Nigeria where studies on these problems could be undertaken. similar foci occur across West Africa. (d) life rn the construction township was never made unbearable and was mitdly uncomfortable during very brief periods in I968-69. Most of the no idea what a blackfly was. which cause and mode of Budden (1956) in the Sudan Doubtless indeed only inhabitants had -19- (e) although transmission of onchocerciasis was not interrupted, upwards of 50 OOO locaI inhabitants (not to mention those working on resettlement, roads, reservoir clearance etc.) have received some measure of protection from the bites of Simulium damnosum and the disease it carries. Wrth the OIi River passing within a few miles of the township and entering the Nigerjust below the dam, and the mighty Niger itself untreated, thls satisfactory conclusion to the Karn;i Dam development could hardly have been expected. ACKI\TOWLEDGEI\{ENTS The author wishes to acknowledge the efforts made by the many labourers, insect collectors, drivers and supervisors who have worked for the Control Unit since 1961 and also to thank in particular the entomologists and health superlntendents who have at various times directed its actj.vities. Messrs J. P. McMahon, H. H. Goiny, D. J. Robertson, J. S. Anderson, H. D. Davin, Drs J. J. MeIIink and Mr S. A. Oyewote. The work was made possible by the concern of Balfour Beatty & Co. Ltd., Joint Engineering Consultants, for the welfare of their workers and by the Niger Dams Authority who agreed to finance the operations. The Unit always received the fulI support of Dr C,. S. Hitchen and Mr J. D. Gwynne of Balfour Beatty and latterly that of Mr R. D. Nevison of N.D.A. (on secondment from Ontario Hydro). Dr B. B. Waddy origrnally advised on the setting-up of the Unlt and has continued his active interest throughout. I am particularly indebted to him for his encouragement and for suggesting that this report be written and aLso to the other members of the UNDP Kainji Lake Research Project and Mr D. Kelley of FAO, Rome. Dr D. J. Lewis and Mr J. P. McMahon gave helpfuf advice and Professor D. S. Bertrarn kindly aftowed me writing up facilities in the Entomology Department of the London School of Hygiene and Tropical Medicine. Finally I am grateful to WHO for sponsoring this report and to the staff of Parasitic Diseases unit, Division of Communicable Di-seases, especially Dr L. Kartman for advice. 20 REFERB{CES BarnIey, G. R . (1953) Control of Simulium damnosum (Theobald) on the Victoria Ni1e, Uganda. Unpublished WHO/Oncho/fS *ir"Jgraphed document Buchanan, K. M. & Pugh, J. C. (1951) Land and people in Nigeria, University of London Press Budden, F. H. (1956) The epidemiology of onchocerclasis in Northern Nigeria, Trans. roy. Soc. trop. Med. Hyg., 50, 366-378 Busvine, J. R. & Pal, R. (1969) The impact of lnsectieide-resistance on control of vectors and vector-borne diseases, BuIl. W1d HIth Org., 40, 731-44 Carlsson, G. (1967) Environmental factors influencing blackfly populations, 37, 139-tso BuII. WId Hlth Or CIayton, W. D. (1957) A preliminary survey of soil and vegetation in Northern Nigeria, Soil Survey Reports, Ministry of Agriculture, Samaru Crosskey, R. W. (1956) Trans. roy. Soc. The distribution of sirnufium damnosum Theobald in Northern Nigeria, trop. Med. Hye. , 5O, 379-392 Crosskey, R. W. (1957a) Man brting behaviour in Simulium bovis in Northern Nigeria, and infection with developing filariae, Ann. trop. Med. Parasit ,51r 80-86 Crosskey, R. W. (1957b) The Simuliidae (Di-ptera) of Northern Nigerj-a, 59-7 4 BuII. ent. Res. , 48, Crosskey, R. W Bu1I. ent (L958) First results in the control of SimuLium damnosum in Northern Nlgeria, Res. , P;, 715-735 Crosskey, R. W. (1960) Distribution recorris of the btackflres (Drptera: Simulij.dae) of Nigeria and the Southern Cameroons, with a key for their identrfication rn the pupal stage, J. W. Afr. Sci. Ass., 6,27-46 Davies, J. B. 1955-60, et al. Bul1. (1962) The control of Simulium damnosum at Abuja, Northern Nigeria, WId Hlth Ore 27, +Ot-StO Fredeen, F. J. H. Canad. Ent. , (1962) DDT and heptachlor as blackfLy larvicides in clear and turbid water, 94,875-880 Garside, J. & Darling, H. (1951) Death of turkeys by attack from Si-mul-ium griseicolle Becker in the Northern Sudan, BuIl ent. Res. , 3?, 583-s84 Gdckel, C. W. (1966) Mapplng of geographical distribution of onchocerciasis. mimeographed document WHO/Onch "fAA, al Unpubli shed ius Tschusi) a pest of Hausermann, w. (1969) on the biology of Srmulium dannosum Theobald, 19O3, the main vector of onchocerciasis in the Mahenge mountains, UIanga, Tanzania, Acta trop. (BaseI) 26, 29-69 Jamnback, H. & l{est, A. S. (I9?O) Decreased susceptibj-1rty of blackfly larvae to p, p'-DDT in New York State and Eastern Canada, J. econ. Entom., 63, 2L8-22r Keay, Krng, R.W.J.( 1953) An outline of Nr rlan v tation Lagos Govt. Printer H. H. (1923) The Spanish sparrow Passer hi iolensis tr grain in Dongola Province, BuIl. Wellcome trop. Res. Lab. (Ent. ), 20 2t Klinkenberg, K. (1965) I lorin Province, Report of the reconnaissance soil survey of part of Borgu Division, Soif. Surv. BuI1. (Samaru) 28 Le Berre, R. (1966) Theoba1d,19O3 Contribution i 1'6tude biologique et 6cotogi que de Simulium damnosum(Diptera Simufiidae) M6m. ORSTOM 17, p. 2O4 Lewis, D. J. (1948) The Simulildae of the Anglo-Egyptian Sudan, 99, 47s-e6 !.q1.. .oy. ent. So.. (L , McMahon, J. P. (1963) Kainji Development. Entomology Interim report 1961-1962. Unpublished Report Disease vector surveys and controf. McMahon, J. P. (1967) A review of the control of Simufium vectors of onchocerciasis, 8u11. Wld HIth Org. , !], 415-430 Highton, R. B. & Goiny, H BuII. WId Hlth Org. , 19, Muffett, D. J. M. (1964) Concerning Brave Captains, l,ondon. Deutsch Nnochirr, E. (1964) Studi.es on the epidemiology of onchocerciasis in the Ibadan area of Western Nj.geria, W. Af r med. J , f3, t3g-50 McMahon, J. P., from Kenya, Simpson, J. J. (I912) BuI1. ent. Res., H. ( 19s8) 7 5-tO7 The eradication of Simulium neavei Entomological Research in Briti-sh West Africa. 2, 30r-356 II. Northern Nigeria, & Harada, S. (1963) A record of blackfly larvae resistant to DDT in Japan, Med. , E, 4l-46 Suzuki, T., Jap. J. Ito, Y exp. TauffI i eb, Bu1f. R. (1956) Rapport sur Inst. Etudes Centrafr. la campagne antisimuli.dienne de 1956 au Mayo Kebbi, , lf, 53 Thomson, K. D. B. (1959) Report of a survey carried out in the Wawa District of Borgu(Ilorin Province), in connection with the proposal to dam the Ni.ger near Dogongari. Unpublished Report B. (1969) Prospects for the control of onchocerciasis in Africa, 843-8s8 Bull. WId Hlth Org. Walsh, J. F. (In press) The Simulridae and their control with speclaf reference to the Kainji control scheme, Proc. ent. Soc. Nigeria Waddy ,B. 40, 22- ANNEX 1 INDICATIONS OF DECREASED SUSCEPTIBILITY OF SIMULIUM DAMNOSUM LARVAE TO DDT IN THE KAINJI CONTROL AREA As mentioped previously control since I96t has been ent|rely by a 25 per cent' emulsifiable concentrateofDDT(ShellArkotineD.25).Tablegrecordsthesuccessobtainedin196lwith thisinsectici.Ie.Throughoutthecontrolperiodmostriversweretreatedatratesoftheorder of O.I-L.O parts per milIion, based on the standard 30 minutes period of applicatlon' These treatments were highly successful over a wide range of conditions' However' the river Niger itself, discharging at up to 22O OOO cusecs G2a n3fsec) was very costly to treat and the use of larvicide had to be as economicar as possible. It was found that under suitable conditions of flow treatments at the rate of o.o33 ppm were futty effective in the early years. standar- dized larval searches were carried out just prior to and 48 hours after larviciding' Treat- rnents were judged successful when larval life completely disappeared below the dosing poi'nt for a distance of go km (5o mires). These results "o*p.."d favourably with those of other control operations (Fredeen , tg62) and in several cases larval life was known to have been elimj-nated for at least 177 km (I1O miles). Larviciding rates varied rittle in the earry years but were higher than o.o33 ppm duri'ng periods of low discharge. The rates, which were known to be near the minimum for effective treatment, were increased slightly in 1964 and more sharply in 1966-67 when several treatments failed to eliminate larval life. The gradual increase in the successful larviciding rate is clearry shown in Fig. ro. For each year the upper number is the highest larviciding rate which was not fulIy effective and the lower is the lowest rate which was always ful1y effective' Table 10 gives 1n greater detail some of the key treatments made over the years' Fr:om Fj-g. lt) and Table IO it is clear that after eight years of regular larviciding simulium damnosum remained very susceptible to DDT emulsion. Nevertheless, in that time it became necessary to double the concentration to achieve complete destruction of the larval colonies and this had a very marked effect on the over-arl cost and feasibirity of the scheme' Thus to take an extreme example , on |2 october 1962 the river Niger was treated at a rate of o.o33 ppm, when discharging at 22O oOO cusecs (6244 n3fsec). This required 465 gallons of insecticide whj.ch at that time cost about Nt o.7 per gallon. Thus the total cost of this single treatment was over N€ 325. Doubling such a treatment would have been too costly' Despite many years of rarviciding against brackfries very few cases of reduced suscepti- bitity to insecticides have been reported. The only published cases appear to be from Japan(suzuki et aI. 1963) and very recently from canada and the united states of America (Jamnback & west, lg7o). So far no concrete evidence is forthcoming from Africa, though Busvine & PaI(1969) report the suspicion of DDT resistance from southern Ghana' Resistance was fulIy established in Japan after nine years of control but the breeding site was smalI and very isolated.InNewYorkStateirregularlarvicidinghadbeencarriedoutsincel94Sandthe Canadian breeding sites had also been subjected to a long period of treatment' Tropical species with their much shorter generation times might be expected to develop resistance faster than temperate species. I{owever, conditj-ons in Africa have not so far been very suitable for this. Despite consiclerable rnterest in African brackfly contror, Kainji appears to be the longest continued large control operation (McMahon, I967), and control at Kainji, as in other West African schemes, has taken place within a very much larger focus' the control area being constantly invaded by, presumably, highly susceptible flies' Also in the KainJi area breeding occurs on the Niger throughout the year and as egg to adult development takes only a fortnight over 20 treatments per year would have been necessary to keep the river compretery disinfested' As this was prohibitively expensive many generations of flies have been able to develop un- affected by control measures. Unfortunatery attempts to determine the susceptibirity revel to DDT of S, damnosum larvae at Kainji failed as it was not possible to keep the controls alive for 24 hours. 23- ANNEX 2 THE USE OF TRAPS Assessment of control must be by neans of moni-toring the fly populatlon. From the epi- demiological point of view man-f1y contact is the important considerati.on so that catching methods which assess this are vital to the controtler. Trapping methods which collect the non-hunting fly, whilst very useful in enabling the entomologist to build up a picture of the total behaviour of the insect, cannot replace the traditional catch calculated in terms of fIj.es (biting) per man hour (FPMH). Catching by means of a tube has several important limitations: (i) It is expensi.ve. (ij.) It is difficult to separate dlfferences in degree of attractiveness to the fIy and in skill of di.fferent collectors. (iii) Only exceptionally skilled tube catchers can obtain many more than 12O fpmh even when many more flies are biting. The skirt trap At Kainjj, a skirt trap was developed by Mr H. H. Goiny and subsequentty tried out by the writer. It has proved of great value. In essence it consists of a crinoline skirt of thick black cloth, supported by steel wire hoops, which fastens tightly at the waist. The collector, wearing only a pair of swimming trunks, dons the skirt and holds up the hem, (reinforced by wire), exposing his bare legs and feet. After one minute, at a signal from the supervisor, he drops the skirt. Any fl1es which have been attracted to hls legs are imprisoned in thedark except that there are two openings near the waist which are fitted with Kilner jars and no-return funnels cut from one litre milk bottles. The jars are treated with insecticide so that they function as kllling bottles. Inslde the sklrt are two twig brooms, with their handlesprotruding, which can be brushed against the legs to dislodge any flies whlch have already begun to feed. These traps were found to be qulte effective and had certain advantages over the tube collector: (1) By alternating dlfferent traps with different insect collectors it was possible to measure the dlfference in attractiveness of individuals to the fly, the element of ski11 having been eliminated. (i.i) No matter how many flies were biting a realistic catch was obtained. In August 1968 two trapmen recorded a rate of 3350 fpmh whilst three tubemen succeeded in recording 186 fpmh. (iii) At levets below I.o fpmh the trap becomes more efficient, probably owing to boredom infl-uencing the alertness of the tube collector, (Fig. 11). The trap would thus be of use in checking attempts at eradication of the fly when many hours of negative trapping would be inevitable. A disadvantage of the trap is that flies tend to get damaged and are difficult to extract from the bottles alive, thus when flies are needed for dissection tube catching is more suitabLe. The trap does not reduce costs as each still requires manning. The cost of each trap was under f5. When traps were brought into use the system of fly rounds was abandoned for stationary catches. The traps are cumbersome to carry through the bush and the material ls likely to be torn on thorn trees. 24 Annex 2 The Rothampstead Iisht trap A trap was constructed according to the instructions of lYilliams & Davies. It was powered by a l2-vo1t car battery and used a car headlamp bu1b. It was hoped to catch engorged flies wlth a view to determining their preferred hosts. The trap was operated during the 1967 wet season. Only nights after conventional catches had indicated exceptionally high fly numbers at the particular localities were used. Nevertheless, in 18 nights only five S. damnosum were captured, and no other species of Ellum were taken. The experiment was abandoned. 25- ANNEX 3 AIL DAY CATCHES AlI day catches were carried out in each month of 1967 at the junction of the Kontagora and l\laingyara rivers. Some further catches were undertaken in I968. Catchlng usually begarl between O8.OO and 09.OO hours and continued until dusk, 30 minutes catching being done in each hour. Some of the results are shown in Fig. 12.1 Although no reg:.rlar ageing of the captured flies was undertaken it seems untikely that the differences in various months could be due solely to the varying proportions of parous and nulliparous flies. The Kainji results were quite simiLar to those which HaUserrnann obtained j-n Tanzani-a (Hatlsermann, 1969), The contrast j,n pattern of biting behaviour between wet and dry season flies emphasizes the importance of very careful organization of the catching programme. IdealIy assessment of control operations should be based on a1I day catches. If in practise this proves prohibl- tively expensive, at least some regular catches should be carried out to determine the optimum blting period and catches arranged accordingly at this time. I I,\rll details will be published elsewhere. 26- ANNEX 4 EFFECT OF LARVICIDE ON NON-TARGET ORGANISMS Although knowledge of the river faunas in the Kainji area is rudlmentary, several collec- tions of invertebrates were made from the Kontagora and other rivers. In general streams ftowing across the basement complex suffer from two types of catastrophe each year. This makes generalizations about the effects of control operations, based on Holarctic experience, difficult to sustain. Firstty, savanna streams are seasonal and cease flowing for greater or lesser periods each year (in the case of large rivers about two months, and for small streams over six months). Secondly, whilst flowing they are subject to incredible spates which may result in sudden rises in water leveI, up to 15 ft (4.5 m) overnight has been re- corded on the Kontagora, a river which, owing to the influence of sandstone has a more regular flow than most. The result of these catastrophes seems to be that the rivers of the Guinea Savanna are relatively'impoverished and that Simuliidae form a large part of the total biomass(estimated at up to 80 per cent. - compare Carlsson' 1967). The larvicide used was very effective against most Simuliidae' though it was noticeable that S. griseicolle larvae survived on the Niger when treatments elimina possibly this was the result of slightly different feeding behaviour. ted aI1 S. damnosum. Presumably S. grisei- colle feed on smaller particles, and they may be less active feeders. After treatments *-" ".r.p.nded aII the species soon became re-established, except for S. bovis which was virtualty extinct in Borgu until after the great reduction in control in 1968 (see also Crosskey, I958 ) . Cotlections of invertebrates from the perennial river Kontagora in 1967-68 showed that this river was very impoverished and remained so throughout the dry season. Collections made in May, after slx months free from larvlcidlng, showed tittle difference from those made in January, except that about five species of Simuliids were re-established. Invertebrates, in descending order of abundance, consisted of the larvae of Coleoptera, Diptera, Ephemeroptera and Odonata. Molluscs and crabs were also reasonably common. At no time were Trichoptera and plecotera larvae taken from the Kontagora, though they were commonly found in the Niger north of Bussa and in the Malendo. Their absence from the Kontagora seens likery to have resulted dlrectly from the control operations. As both are predators of Simulium larvae this is unfortunate. There was no evidence that larviciding directLy caused fish mortality during normal wet season operations, but when treating intermittently flowing rivers 1n the dry season some mortality was inevitable. Thus streams in the Guinea savanna zone, especiarry those with seasonal flow, are pro- bably relatively unproductive. Animal species which are abundant are those with the best powers of overland dispersal or those which can survive in poorly oxygenated and still waters' With certain exceptions it is not thouBht that tarviciding had any marked direct effect on the non-Simutiid fauna. However, as Simuriids form a major proportion of the total biomass their elimination must have a marked indirect effect on the over-all productivlty of the treated streams. Dry season control inevitably results in direct flsh mortality. -27- TABLE 1 NIGER DISCHARGE LEYELS (CUSECS)1 ET TAII'T.IT 19 6r t962 1963 I 964 19 65 1 966 1967 19 68 ' , .3/".. = 3s.2 cusecs {tt3/sec) TABLE 2a. ANNUAL RAINFALL (IN mm) IN THE KAINJI DAM AREA 1961-69 Year Year Black flood maximum level Minimum flow White flood maxlmum level 69 690 64 060 91 280 69 000 69 000 80 240 ro 550 7 743 10 550 7 39r 5 63r 10 550 t4L 332 220 o,o0 115 rOO 184 000 126 000 109 100 149 600 Kainji harbour Rainfall Dry months 2 wawa/New BussaI 2Rainfall Dry months I 96I r962 1 963 r964 19 65 r966 I 967 1968 r969 Average 1 100 993. 6 835.2 876.8 986. 5 938. 7 901.O I 178 I I91 r ooo. l 4 4 5 5 6 5 6 5 936.3 r 168 L 2o,4 I O20 I r79 I O40 9c2.4 I 45r I 203 L 122.2 6 4 4 6 5 6 5 5 5 <,, 5. 11 1 t96t-64 Control Unit Office Wawa. 1965-69 Control Unit Office New Bussa. Months with less than 25 mm rainfall., I 28- TABLE 2b AVERAGE MOI{THLY RAINFALL (IN NM) IN THE KAINJI DAM AREA 1961-69 January February March Apri 1 May June July August September October November December TABLE 3. pREvALH{CE oF oNCHoCERCIASIS IN THE oLI VALLEY IN 1963I Age-group 1-_ ' E*p".sr"d in percentage of individuals with onchocerciasj.s in different age-groups. 844 persons examined. Month Kainji harbourRainfall Dry months Wa Rainfall *"/N ew Bussa Dry months o 7 .874 . 4.826 78.48 71. 61 138.5 216.9 t77 .3 2tL.8 87. 38 7 ,27 o 9 8 8 2 1 o o o o 1 8 9 o 5.44 LL.72 7 5.4 108. 1 168.4 2L5.4 2t8.7 24t.5 68. 8 8.4 o I 8 8 t o o o o o 3 8 q o-9 10-19 20-29 30- 39 40-49 50+ Females 8.2 9.1 29.2 4r.o 45.2 37 .4 MaIes t2.8 36.1 48.5 72.L 88.2 70.5 I -29- TABLE 4. PREVALENCE OF ONCHOCERCIASIS IN THE OLI VALLEY IN 1967 38 79 15. 8 2.4 37 23 2,7 1.8 Not necessarity the result of onchocerciasis. TABLE 5. PREVALENCE OF ONCHOCERCTASTS IN THE NrGER WEST BANK/ OLI VALLEY REGION 1967-68 Percentage 76 79 30 L4 7t I5.7 5.4 54 1 1 Age-group o-5 6-10 t1-15 1 6-30 3o+ MaIes No. examined to +ve biopsy lo +ve nodules le +ve ocular defectsl Females No. examined /e +ve biopsy 16 +ve nodules ls +ve ocular defectsl 9 o o o I o o o 3I 16 6.5 44 o o o o 26 11 3.9 18 o o o Populat ion Number1548 Positive biopsy only Positive biopsy + nodules Nodules only Positive Mazotti test only Bl indnessl s75 2L5 )) |fl" 46) I9 140 37 ) ) ) ) ) ) 2) L4 55. n" J 1 e.o) Not necessarily the result of onchocerciasis. o 12 30- TABLE 6. PREVALENCE OF ONCHOCERCIASIS IN THREE DISTRICTS NEAR KAINJI 1967-68 For localities see F1g. 3. Not necessarily the result of onchocercj-asis. TABLE ?. ONCHOCERCIASIS IN THE LOWER KONTAGORA VALLEY 1967-68 I Fo, localities see Fig. 3. ' *o, necessarily the resurt of onchocerciasis' Di st ri ct1 No. of peopleexamined Infected No. % BIind2 No. Niger W. nant/ori Valley Niger E. Bank/Kpatachi Is.f Lowe r Kontagora-MaingYara Auna I 548 841 848 855 445 449 55 53 53 140 69 28 9 8.1 J.J Vi t lage1 No. of peopleexamined % rnfection % Brirrd.r"."2 DaIau Shaffini Fel lagi Patiko T. Anfani 27 186 59 135 11r 59 51 46 51 51 16 l2 l2 6.2 1.8 3I d o o !-{ , '-'l EQtdaiH.Ecz e .Fl -i\ v(sd cB(d = (r) o N -4 F{ tr)$ Fl ^lo o$o Fl ot- Fi '-.,| oNo Fl \fl oN r-{ @(o F{ r-l (o cr)o ti ooh0E bo'.{arzEd .Fl .c ^xoodo0) ='.{ o o oo(o o$ .{ oo '{ oo .l o (o N F{ ooo !$ @ F{ oo F{ to r'{ F{ ooo o NN c\t co '-{sf F{ 'Flo,iE b00a,E td^ .''{ -C Oxoc)dooE'rl ,15o ooo F.' F{ F{oooooooc'oooooooroooFroo@o Ct$F{FlrO'-{@rO N F{ ri r.{ '-{ F{ N 0)! ..4 Ootr .''r r5 O+r O-{O o,-{Or(do60 Cv H $lo$ F{ r-{(oAIOo)F-NCr) ,_{@r,)rJ)(oNoo(oroD-$(oN@o) cfjcD(oF-(I)srN Fl o tr +J o qr o op F q) Oti e)OF{ {J ..{ od >+{ o i(Bo!trtro+Job0c > - o .,.{YQrr rco d.o>z .r.l .-.1 -{IQooo +{h00 .n .ctr >+)OOk +),ioCHHtr d Otr -fr 0)oob0 , >.r{ ..1 .A Z1' [rdotr OlCP..rE.- .E--- o q O{J.il@\\ tr cdsl Ca .'-{ V o .'-{ dL dp @ <r l{ o .r-) d E dCE(dJ1 oo ,@ .riooCdP o Eb @ .F,l$L tr!trooo '-) .,-) .r_) ddcdrrEEE oP!1rddtrECECEr{ (d,14 dll d,ll '-{ o oO oO oOd , ,@ 5@ ,@P .'.{ .d .'.{ .F{ tr13!OOO!O o d dP d+r d+)Ftrtrtitr .F{U)OOb !E.Eb Eb !b X rf@ @.Fi @.l{ O.d f4 Nsl sl L i$ tr <t lr .cd =>o.otuz a .m to tr- F{ o cf) Fi tr) F.l F{ @ o o C\t ci (0 o ci .o .+) C) =otr. oAPN .o frh q (o @ (o co co ro ro (?) '{ (o F{ F( @ @ F{ o ^l CJ A F- o t- @ :t{ @ r.- lr) Erd o d H a HA H() H Ert) trla zH : tsJo Fzoo'tHG.<o2 14 ts1o.< >EL Hrztr< H =O r,l r,l O 4Zq =Qo2CJ< r{H ^Fl.o .p< =oA r,1 . F-rH<vF aaFhl a r,1i, Fl zz L o d E t{ oo bo .F{ .Fl It o 'F{ F{ 14 @ E] Fl tq F O)(o o d co(o o/ F-(o o F{ (o(o o .l ro(o o r{ $ o F{ C'(o o Fl c!(o o .l F{(.o oA .-l -32- TABLE 9. L,ARVICIDING TRIALS 1961 ResuI tTreatmentin ppn Length of river inspected Discharge ( cusecs ) DateRiver A few larvae survived over 27 km downstream Complete success Complete success Complete success Complete success Complete success Complete success Total failure Complete success Complete success Conplete success 80 lsn 80 km 33 lsn 33 km 33 km 96 km 64 lsn 64 km 13 km 13 kn 13 }m o.025 o. 033 o. 10 o.05 o. 03 o. 50 o.20 o. 10 o. 50 o. 10 o. 05 27 .9 .6L 11.10.61 8. 8. 61 13.9.61 4.10. 61 30. 8. 61 24.10. 61 16.11.61 17.8.61 24.rO .6L 15.11.61 r4r 332 LL6 372 2 r18 354 236 t77 118 500 3 531 2 706 t t77Kontagora Kontagora Kontagora Maingyara Maingyara Maingyara Niger Niger ori o1i oli 33- TABLE 10. DETAILS OF THE MORE IIIIPORTANT TREATTvIENTS OF THE RIVER NIGER AT BUSSA, USING A 25% DDT EMULSION Date River dischargein 1OOO cusecs Larviciding rate in ppm Effectiveness over 80 km Remarks 27 .9 .6r 11.10.61 14'et 1962 (6) 29. 8. 63 13.9.63 23.9. 63 19. 8. 64 27 .8.64 L4 .9 .64 5. 10. 64 13.rO.64 Mar 65 (2) 14.8. 65 25.9. 65 29 .3 .66 14.9.66 9. ro. 66 11.12. 66 5.1.67 I .2 .67 t8.2.67 19. 9. 67 30. 9. 67 24.1O.67 6.4 . 68 16.4.68 26.4 .68 141 116 30-220 95 l02 85 L27 95 r84 r68 130 7t 52 126 55 78 9I 57 6I 68 68 L24 135 93 70 62 78 o.o25 o. o33 o. o33 o. o33 o. o28 o. o40 o. o40 o. o50 o. o40 o. o40 o. o40 o. o50 o. 067 o. o40 o. 067 o. o58 o. o50 o. o50 o.046 o. 046 o. 046 o. o50 o. o58 o. o67 o.o71 o.o77 o. o91 xx xxx xxx xxx xx xxx xxx xxx xxx xx xxx xxx xxx xxx xxx xxx xxx xx x x x x xx xx xx xxx xxx Effective to 177 kn A few S. griseicolle survived One effective to II2 km Effective to 177 lcn Effective to 177 km A few S. griseicolle survived Mainly final instars survived Mainly final but also some earlier instars Many final instars A few older larvae survived Nine larvae only xxx xx x Fu11y effective, all larval life extinguished. Very smaIl numbers of larvae survived. Substantial numbers of larvae survived.

Fig. 1 THE KAINJI CONTROL AREA YELWA Con t rol A7ar NIGERIA Fog. la BUSSA \ tlJ '\._.) .)I ,i (. o T a wAwA AINJI .K IAMA g o , .TOKWA . KlsHt .sHAXt Ibb ht --/, ILOR IN \ o oo ? o.(, o, o E - a!- o o z =G,Y (! 0) O) 2 o) : G. q) E q)(,) (I, E o o o o o ofg o u- o) o o o. o tr) =-) c a -) (E o (o o z o o o. o) a O) a o o) u- c(o -) C\li II N(f, fcase) scaurn3 ooo't + E .v, rJ)N g E ! E oa.oPL5 03E >oo; i!OCat Otr) l o)(E f (!Il E o C) o f o CE ct a u,x o o F a ,6 E) E ao a c o a e e G o- a c o! o U' (u (I,l zl a a a \ \-O \ oAo o- ll- el t I B# uroOz) @o U)f co 3 E (r l 3 -l U'F 9 G F @ o z l CDI (b i E, o z_ = E, o0 Fz oY TL uJ =oJ ! a I(J F o-Y I Y z 6 F U) UJ (L uJI z c{l t UJ J J =o I Y z CD Fa LU B (r LU(9 z el ;(, o o UJ(9 & z G UJF o- F z UJ =luJ F UJ(t, iltlJJ G o0 F z o Y (v) E)ii Fig.4 - DISTRIBUTION OF SIMUL]UM DAMNOSUM BREEDING SITES I Scrsonrl Broodrng Sr tca lralor Br..ding Rivort O Percnnitl Ei.ading Si t.t KH \ I ! ! \. ? 50o I I -'! Fig.5 - INNER LARvtCtDtNG pOtNTS KAINJ O Regular Points 1963 -Aug.'69 o Occasiona I ,, Fis.6 - OUTER LARVICIDING POINTS TREATMENT POINTS z Expcrmanttl L Long B.ng. P Rcaul.r 1963 - Aug 'C8 L a P ;oCL '= ^9(,Lc -YO ovoctr LO l-c(! .L(D =,fEaLoi_)roFoc ocd);E:9E:-:ao) -fG,-c:o,o!O:"o()o) rll I tllI ,l E sz. lI) N o -o (g co & o Qo ') rfNrv) l)v"l \(E 6n<z) trt (E 3(u 3 ) 3 ---Y' r o) @ o, U;Fz 6 o- (9 z o a E J r\ (9 L I et_ 33 LIJ -zii O< a a 3 a ; o Cou)9ol- .3!s3 (o ! :! = o :o) a ox o P oEo- C39)o.= -JurE- coe(-);s3i;OFrlU-(J iUl. EI u? N aF z o(L (9 =-(J F o (b a oz f o G, J lJ- I @ ct)iI PROTE CTION ZON E 1961 3 JJASON NEW BUSSA 15 10 5 o JJASON . No Catches Mo nth s o 3 NEW BUSSA 3 0 3 1 963 o o 3 1964 1965 1966 1967 1968 1969 0 3 0 3 0 3 0 f oI tr o o .9 E 2 oo 20 5 3 ) Fig. 9 - DENSITY OF SIMULIUM DAMNOSUM JJASON 7 5 2 o 0 toO+)LrO hl! .ri O ltN B o .b0 od;odoO! o ool .r{ *; Po0)E{J .cdPO .r{ 'Fl (sC .Ft cdo O .-l ,+d o >0)o>io .ood .r{ O +J C) dlrP, .n h04r 'F{ d d.^ o od+)> .r{ lr dC.{I +) o@acc.Aod o.b0OcS o trOO 'r{OlJ(doo]J.dd '-lKQq b0d 'd 0)d> .Fl .F{OP 'r{ O>c)Lq6qd0) ocld IJ 0)q+j od o C 'r{o€ .ac P .r{ .A ooocq 'r{i0)o>ld'.{F+J3()o0) FIHOH .oo o(o O) F{ ?-l F-q o t\ tr- o F-(0 r{ D- o o (o(o O) F{ o roq o @lr)q o lr) o .{ sq o srr(o oq o roq o co(o oA @ C\o c; cr) c,)q o N(o o Fl CfJ o A(o 6, F{ ro No d o o EE I o 0)P ...1 E o(?) 0) o .r{ o o +J(d ti 6I .A .A o 'F{ tr d rl a @rl a H 94t) El h frl z H o a 14tr @(o I .-{ o .{ =l;r CAIot E]I = H F] =Ha tuo FlofrF2 () Eo tu z Ha H O H & S d r,1 Hd frHo H z Fl fr1frDo H 14 Fig.11- TRAP & TUBE CATCHES COMPARED (Ovcr l,OO0hours in 1966) a o x a 300 200 100 70 50 30 20 t5 3 to .at c rtch.l x ( TT x O.3 O.5 0.7 1 1.5 2 rlr x x II I 3 57tO15203050701OO a c o a x T o c G,F o o E3 o I an u,t =lt a e ro, 7 5 3 2 1.5 I 0.7 o.5 o.3 I a I FLrE3/HOUi by ru!E I Fis.12 - ALL DAY CATCHES, KONTAGORA-MAINGYARA JUNCTION r8 16 11 12 'to -y-1- April (tO8 mtn houTa cetching ) -o-- -o- Scpt.mbcr t197 " ) V 2 -r-x- Fcbruary - o --1-Augucl A \ --(-V \o .0 \ I8 6 1 \ -\ \,-___ o/ 2 = =o G f o - o UJ =r 8 7 6 a 3 (272 (164 .n- -J fr \ \ \ \ \ \ \ 5 / ? 2 o to 13-14 la-15 15-16 16-17 17-lE 18-19 HOURS t 8- 9 -1O 1O-11 1l -12 1 -13 \