~ ~ p%51
I. ~* ~'T
* .T*l- *~I~
r xfr.iA r ~
'v "ll -,V:T_
1 9 Vfr4.
SIR WILLIAM HALCROW & PARTNERS
(LATE C. S.MEIK & HALCROW)
CHARTERED CIVIL ENGINEERS.
SIR WILLIAM HALCROW, M.I.C.E.
H. D. MORGAN, M.Sc, (ENO.), M.I.C.E.
PETER A.SCOTT, B.Sc., M.I.C.E.
C. M. ROBERTS. M.I.C.E.
J. A. JOHNSTON, M.I.C.E.
V. A. M. ROBERTSON.C.B.E., M.C., M.I.C.E.
R. H. ADAMS, M.I.C.E.
B. D. RICHARDS, B.Sc., M.I.C.E.
R. D. BROWN, M.I.C.E.
TELEPHONE NC. WHITEHALL 8700 & 8283-4
TELEGRAMS "PROPULSION. SOWEST. LONDON"
Dr. Lewis Berner,
University of Florida,
GOVERNMENT OF THE GOLD COAST
RIVER VOLTA BASIN
LEWIS BERNER, B.Sc., M.Sc., Ph.D.
3oth NOVEMBER, 195o
UNIVERSITY OF FLORIDA,
Wlghtman Mountain, Ltd., Westminster, S.W.1
Part I INTRODUCTION:
Objectives...................... .. ...... ......... 1
Methods.. ...... .. ..... .. .. .. .. .... ... ... ...... ..
Part II MALARIA. ONOHOCERCIASIS. TRYPANOSQMIASI8
and the VOIA RIVER PROJECT:
THE V012A RIVER SYSTEM.............................. 6
Major tributaries of the Lower Volta............. 8
VEGETATION OF THE VOLTAIAN BASIN,................... 10
CLIMATE OF THE VOL2AIAN BASIN....................... 11
DISTRIBUTION OF POPULATION IN THE VOI1AIAN BASIN.... 12
THE DEVELOPMENT OF THE VOIMA RIVER BASIN ........... 12
MALARIA. .... .. ..... .. . .. .... .... ....... ..... .... a3
Malaria vectors.. ......................... ...... 14
Anopheles collections............................. 17
General discussion and recommendations........... 19
Harbour areas............ ....................... 26
The insect vectors ....... ..... .................
Simulium damnosum, collections and observations
in the Volta drainage....................... 32
Other species of Simulium found in the Volta
drainage...... ............... ......... ... *8
General discussion and recommendations.....,,.... 40
TRYPANOSOMIASIS.................................. . 44
Collections and distribution of Tsetse........... 45
Present methods of control ....................... 47
General discussion and recommendations ........... 48
Part III MALARIA, ONCHOCERCIASIS, TRYPANOSOMIASIS
and the ACORA PLAINS IRRIGATION SCHEME:
ACCRA PLAINS IRRIGATION SCHEME...................... 52
IRRIGATION AND MALARIA.............................. 52
Irrigation malaria................... ...... ..... 52
Anopheles in ricefields....................... 54
Effects of defective irrigation................. 55
Irrigation and anophelism in West Africa......... 58
Technique of control............................. 59
ONCHOCERCIASIS. ... .......... ...... ........... 62
TRYPANQSOMIASIS............................... ..... 63
Part IV OTHER INSECT-BORNE DISEASES;
LOASIS..... ..... ..H.... ..A................ ..... ..... 66
ACANTHOCHEILONEMA PERSTANS.......................... 67
YELLOW FEVER................... ................... 68
DENGUE. .......... ....... ........ ..... ....... .... 70
Part V SCHISTOSOMIASIS and GUINEA WORM:
SCHISTOSOMIASIS (BILHARZIASIS)............... ....... 71
GUINEA WORM. .......................... ...... .... 73
Part VI RECOMMENDATIONS ON THE VOILA RIVER PROJECT:
RECOMMENDATIONS........... ..... ..................... 75
MALARIA ...... ..... .. .... .... .......... ..... 75
TRYPANOSOMIASIS............................ ....... 77
Part VII RECOMMENDATIONS ON THE ACCRA PLAINS
MALARIA......... ... ... . ...... ............. ... ... .79
ONCHOOERCIASIS............ ....................... 79
TRYPANOSOMIASIS ..................... ......... .. 79
Part VIII RECOMMENDATIONS ON CONTROL OF OTHER
LOASIS......... ...... .... ............ . .. ...... 80
AOANTHOCHEILONEMA PERSTANS........... .............. 80
YELLOW FEVER...................... ................. 80
DENGUE. ..... ........ ... .. . .. ....... ....... .... 80
Part IX RECOMMENDATIONS ON SCHISTOSOMIASIS AND
SCHISTOSOMIASIS (BILHARZIASIS)...................... 81
GUINEA WORM ........................... 81
MAPS AND FIGURES
PART I : INTRODUCTION.
1. The approaching reality of the construction of a dam
on the Volta River for the production of hydro-electric
power necessitated studies of factors far removed from the
immediate engineering problem. One of the major considera-
tions was that of health. In West Africa it is a well
known fact that many of the most important human diseases
are insect transmitted. Therefore, it was recommended that
an entomological survey be conducted to evaluate the
possible effects that the impoundment of the river might
have on the insects of medical importance,
2. Through the Economic Co-operation Administration of
the United States my services were obtained to conduct this'
survey. I was granted leave of absence from the faculty of
the Department of Biology, University of Florida, for a
period of six months beginning 1st June, 1950, as requested
by the Colonial Office. On 14th June, I arrived in the
Gold Coast to begin the survey, which I continued until the
end of October, at which time I stopped field work in order
to collate my data and write my final report.
3. During the first three and a half months of my investi-
gation, I was seriously hampered by lack of entomological
equipment and it was not until mid-September that some of
the required items were received from England. Nevertheless,
my field work progressed rather well and I was able to
conduct much of my work in a reasonably satisfactory manner.
4. I would like to express my sincere appreciation to the
Colonial Office, the Gold Coast Government, and to all the
local officials who were so very helpful and kind to me
during my stay in West Africa. Without the exceptionally
fine assistance of Mr. Eric S. Lomax, Assistant Secretary
for Development, Secretariat, Accra, Gold Coast, my task
would have been far more difficult. Dr. M.H.Hughes, Medical
Research Institute, Accra, has been most helpful in providing
me with laboratory facilities and-equipment,' acquainting me,
through his wide experience, with onchocerciasis in the Gold
Coast, and making his library on Simulium available for my
use. The Director of Medical Services, Dr. R.L.Cheverton,
and the Deputy Diroctor of Medical Services, Dr. W.S.Ormiston,
have both given me every assistance. Major B.K.Hartshorne,
engineer for Sir William Halcrow and Partners, has worked
closely with me and has provided me with a clear picture of
the. entire project. Major Hartshorne's predecessor,
Mr. R.G.Taylor, was also of considerable help to me in the
initial stages of my work. Dr.Paul Freeman, British Museum
(Natural History), has kindly identified several of my
species of Simulium. To all these people and others too
numerous to men-fio individually, I would like to take this
means of expressing my thanks.
5. After my arrival in the Gold Coast, I felt that I
must first clearly define my problem. With that in mind,
the following objectives of the survey were proposed:-
(1) To determine the species of insect disease
vectors present in the Volta river system,
(2) To determine the distribution of these insects
in the river system.
(3) To determine the seasonal distribution of the
(4) To correlate adult and immature stages by rearing
(5) To investigate the biting habits of the insect
(6) To determine, by dissection, the infection rate
of the insect vectors.
(7) To determine the effects of the rainy season on
the spread of insect vectors.
(8) To evaluate the possible effect of impoundment on
the present habitat and the ecological changes
that might be produced.
(9) To determine future problems that would be posed
by impoundment of the Volta.
(10) To determine the relationships of insect vectors
to the present human populations.
(Xl). To determine the relationships of vectors to mass
movements of populations caused by rise of water
level and the flooding of present villages.
(12) To determine problems that will be created by
bringing in large groups of non-Africans.
(13) To determine the effects of the Accra Plains.
Irrigation Scheme on the spread of onobhocerciasis,
malaria, guinea worm, schistosomiasis, etc,
(14) To make appropriate recommendations,
6. Admittedly this was an ambitious programme for one
man to carry out in the short space of four and a half months
of field work; however, I felt that each of these objectives
was of sufficient importance to merit consideration in my
report. As each insect transmitted disease is considered,
I shall endeavour to give my impression of its relation to
the, objectives set forth above. I feel extremely presumptious
attempting to cover such a broad field and so many objectives.
Certainly the problems posed here would form the basis for
many years research by several authorities. In the press
of time, I realise that an opinion must be presented now, so
that an intelligent approach can be made in planning disease
control along with the planning of the construction of the
Ajena dam and the Accra Plains Irrigation Scheme. For that
reason, I am giving opinions now that may, at a later date,
.have to be reassessed in the light of future developments.
7. In carrying out the objectives set forth above,
it was necessary for me to travel extensively throughout
the Volta river system. As time was a consideration, I
confined my attention primarily to the lower Volta and its
tributaries. This was considered advisable because it is
the section that will be affected by impoundment. In
working in a particular area, I chose most conveniently
sited rest houses as my headquarters and worked out from
there. My vehicle, used almost exclusively in these travels,
was a "Land Rover". The car proved to be admirably suited
to travel on the primitive roads and trails of the Gold
Coast. At times it was very tiring, but this was compensated
for by knowing that its four wheel drive could get me through
difficult terrain. Unfortunately, on my last trip my
African driver became intoxicated and wrecked the vehicle
making it impossible for me to complete my projected tour.
8. In establishing my plan of travel, I first worked
the Volta below the proposed dam site at Ajena, then the
Ajena section. From there I worked the eastern banks of
the river and its tributaries. Next came the Afram and its
tributaries, thence to the upper part of the Volta on the
west shore. In detail the places and dates on which they
were visited are listed below (Map 1):-
Volta south of impoundment
Ada ... July 3 -4
Akuse ... July 8 11; 13 15; 18 21.
Amedica Ada (via motor launch) ... July 21 22.
Senchi ... July 31 August 5.
Ajena ... June 16; 22; August 4;
September 26; 28; October 30.
Volta north of impoundment (east shore):-
Anum ... August 8 10.
Kpandu ... August 15 19; 25 26.
Kete Krachi August 19 25.
Tafo ... September 5 6.
Begoro ... September 6 7.
Kawhu Tafo September 11 13.
Mankrong September 13 15.
Agogo ... September 19 20.
Kumasi .. September 20 22.
Maipong .. September 22 23.
October 5 9.
.Pru drainage :-
Ejura .., October 6.
Prang ..* October 11.
Yeji ... October 9 16.
9. I considered that one of the best methods by which
to evaluate the effects of the building of a barrage on
the Volta on the medically important insects was to get
into the field and examine the main river as well as its
tributaries. Not only would such an approach give me a
reasonably complete picture of the terrain but several of
my other objectives could be attacked at the same time.
10. After studying available maps of an area in which
I was planning to work, I attempted to reach all tributaries
or main streams within a reasonable distance of the rest
house at which my temporary headquarters were located.
Although many small streams are shown on the maps, the
great majority of them on both east and west sides of the
river were completely dry during the months in which I
was able to visit them.
11. At each stream I sampled various habitats to determine
species of Simulium (black fly) present and their prevalence,
as well as collecting and noting the presence of other
stream inhabiting insects.
12. Since tsetse-fly boys were not available until late
September, my observations and collections of tsetse were
confined to noting and collecting those which attacked the
natives, my assistants or myself. Other biting insects
were recorded in the same way. The fly-boys, who were
trained by the Tsetse Control Organization at Lawra, were
put to work in the Ajena-Misikrom area with instructions
to collect not only tsetse but also any other biting insects
that might attack them.
13. At Kpong, where Simulium damnosum, the vector of
onchocerciasis, is a serious pest, I planned to try to
obtain some quantitative data regarding its prevalence.
With this in mind I employed one of the local men, who
was literate, to collect the insects three days per week
during definite hours of the day. Within a week, I
discovered that this man was also employed at the Kpong
water works and that hid data on Simulium catches were
being faked. The results were so discouraging that I
decided to abandon the use of unsupervised collectors.
14. In places where Simulium adults were found, I
collected specimens easily by utilising the local children
as bait. It was simple to get as many assistants as needed
by giving the children glass vials and showing them how to
catch the insects as they came to rest and started biting.
The children considered it to be a game, and to stimulate
even more interest a small coin given to each child
resulted in good catches of the insects.
15. During my stay in the Kpong-Senchi section of the
Volta, the black fly adults were easily collected in
sufficient numbers for a limited number of dissections to
be made for the determination of infections with the worm,
Onchocerca volvulus. The dissections were carried out on
the stage of a binocular dissecting microscope. A live fly
was chloroformed, transferred to normal saline solution on
a microscope slide and then the head, thorax, and abdomen
were cut apart and each section was placed in a separate
drop of saline solution coloured with aqueous methylene blue.
16. The mouthparts were first removed from the head and
torn apart in the search for 0. volvulus larvae; next the
rest of the head was macerateT. The thorax was then
opened and the wing muscles teased apart to release any
larvae that might be lying in them. Finally the abdomen
was opened and mid and hindgut examined. The use of
methylene blue in the normal saline was of considerable
help in quickly identifying the larvae of the worm.
17. Mosquito larvae were collected by the standard dipping
method; however, the larvae were not retained after the
species present were noted. Adult mosquitoes were collected
in native villages by the use of a technique I found very
satisfactory in the Accra region in 1944. A sleeping room
of a native house was chosen, preferably a darkened one with
a thatched roof, the floor and "furniture" were covered
with white -bed sheets, all doors and windows (if there were
any) were closed and the room was then thoroughly sprayed
with a pyrethrum insect spray. After being sprayed, the
room was left closed for ten minutes, then opened and the
stunned or dead mosquitoes that had fallen on the sheets
were collected by means of an aspirator. Admittedly this
method is open to criticism since it neglects all species
of mosquitoes which rest outdoors; however, so far as the
medically important species, exclusive of the yellow fever
forms, are concerned, I feel that this method is entirely
satisfactory for sampling populations of West African
species, as the important ones tend to rest in houses during
18. In examining the Volta itself, I employed native
canoes to take me into the stream. At several points I was
able to obtain canoemen who were willing to take me into the
rapids. Admittedly it was a risky undertaking in the flimsy
native-built boats, the bottoms of which were built by
simply nailing two or three boards together and calking with
cotton. As the canoe entered rough water, each movement of
the water resulted in some flexing of the boards; however,
I considered that I was fortunate in not having the boards
buckle up completely, and all rapids were navigated safely.
As often as possible I had the canoemen stop the canoe where
rocks or islands projected from the water and I made
collections from them. The canoemen also ofton entered the
water to bring up loose rocks from the river bottom for my
PART II : MALARIA, ONCHOCERCIASIS, TRYPANOSOMIASIS
AND THE VOLTA RIVER PROJECT.
THE VOLTA RIVER SYSTEM.
19. The Volta comprises the chief river system of the
Gold Coast (Map 1). Its headwaters originate in the
semi-arid French territory to the north and drain south-
ward. The Black Volta forms the northwestern boundary
of the Gold Coast until it turns eastward at Ntereso to
join the White Volta thirty-five miles northwest of Yeji.
The White Volta enters the country from the northeast and
is soon,joined by the Red Volta. The Sissili and the
Kulpawn also draining south, join about forty miles below
the northern boundary of the Gold Coast as the Kulpawn,
the stream now enters the White Volta where it turns
south, From the confluence of the two main rivers to
the mouth, the Volta stretches 340 miles to enter the
Gulf of Guinea at Ada.
20. Other principal tributaries of the river system
are the Daka, Oti, Asuakawkaw, and the Dayi entering the
Volta from the north or east and the Pru, Sene, Obosum,
and Afram from the west. With the addition of the waters
of these rivers, the Volta becomes one of the major streams
of West Africa.
21. In its course, the Volta derives its water from the
semi-arid northern reaches of its tributaries, from the
Ashanti rain-forests', and from the relatively dry grasslands
of the southern section of the country. Because of the
nature of the rainfall (discussed elsewhere in this report),
some of these streams are intermittent while others are
perennial. The Volta is influenced by tidal changes as
far upstream as Sopwe; below that point the water becomes
brackish or salty in the dry season when the flow of fresh
waterdecreases (Junner & Bates, 1945).
22. According to Kitson (1925) the drainage area of the
Black Volta, the White Volta, and the Volta to the Akwapim
range near Ajena is about 139,000 square miles. Xitson
estimated the rainfall in this drainage to be roughly
11.6 million million cubic feet annually.
23. Junner and Hirst (1946) point out that the Voltaian
Basin, which covers an area of some 40.,000 square miles in
the Gold Coast and which is drained by the Volta River, is
mostly low lying with the greater portion being less than
500 feet above sea level. The basin is surrounded by a rim
of highly dissected highlands through which several
tributaries enter from the north and west. The eastern
rim of the Voltaian Basin is broken only where the main
river leaves, it as it passes through the gorge near Ajona.
The southern rim is breached by the headwaters of the Pru,
the Afram, and the Ongwam.
24, Much of the central basin is 400 500 feet above
sea-level. The White Volta enters the Voltaian Basin at
about 400 feet and the Black Volta at 300. The confluence
is roughly at 250 feet and the Volta at its exit from the
basin is less than 100 feet above sea level.
25. The eastern rim of the Basin rises from 1500 feet
near the Volta to 2000-3000 feet near the international
boundary and then decreases gradually to the north. The
southern rim has a maximum altitude from Koforidua to
Mampong varying from 2000-2200 feet, except for a few
places that rise to 2500-2730 feet. Proceeding from Mampong
westward to Techiman the altitude is only 1200-1400 feet.
The highest points on the western and northern rims of the
basin are between 1500-1700 feet above sea level.
26. Xitson (1925) p.44) points out that along the course
of the Volta from Yeji to Ajena there are 22 rapids. "Most
of these rapids, however, are below the tops of the banks
of the river, so are obliterated when the river is flowing
a banker. During the dry season the river, 400 to more
than 1000 yards wide, flows at 20 to 40 feet below the tops
of the banks, but in the rains it overflows, and in many
localities inundates many square miles of country on each
side to a depth of several feet. The enormous discharge of
the river can, therefore, be realized."
27. The rapids at Kpong are extensive, being formed of a
bar of granite which stretches diagonally across the river.
The width of the rapids is about 4000 feet. The innumerable
channels coalesce some 51Y miles from the upper end to form
the main channel of the river once more. The biggest
drop occurs in the first mile where there is a difference
of some ten feet.
28. In June, 1950, when the level of the river was
relatively low, the rocks at Kpong were fully exposed. At
this time the water was rather shallow and the flow was not
excessively rapid so that over most of the area the water
simply coursed around the rocks and the surface was not
strongly disturbed. By 1st August the level of the river
had risen to such an extent that it was impossible to cross
onto rocks, on which I had previously walked with ease,
without wading into the river. In late September, the
rocks of the rapids were almost completely submerged and
the rate of flow was much greater than in June and July.
During 1950 the river did not rise to such a level that the
rapids were ever completely obliterated, but I have been
informed that such is the case in years of normal rainfall.
29. The Senchi rapids, located a few miles north of the
Kpong rapids, are similar in most respects to the latter.
These rapids also consist of a bar of rock crossing the
river diagonally and causing the water to spill sideways
over a distance of about 9000 feet. At low water there is
a difference in level of about 15 feet above and below the
rapids. Inspection of these rapids revealed that they too
become covered as thot river rises.
30. During the course of my travels over the river
system, I visited several other points on the river, which
are indicated on the existing one-inch contour maps as being
rapids. Most of these were either small rock outcrops or
islands present in the main river channel, but they can
hardly be compared in extent or importance with the Kpong
and Senchi rapids. Undoubtedly these smaller rapids are
submerged at high water.
31. The notable exception to the description of the
rapids above Ajena is that at Kete Krachi. I arrived in
the region in mid-August. Of all the rapids in the Volta
river system, these were far and away the most vigorous.
The water literally boiled through the river channel and
beat against the large boulders which are strewn in mid-
stream. The water appeared to be heavily silt-laden and
when viewed from a distance, the rapids resembled the ocean
shore during a violent storm. The current was far too
swift for canoes to pass through. The river is spread
over the rook basin and to the east and west it could be
seen spilling in a broad, flat sheet as it came pouring
across the rock bed toward the main channel. While standing
at the river's edge, I could see an appreciable rise to the
north giving the water the appearance of running downhill.
The stretch of rapids seemed to be about % to /4 mile in
length. The width at the rapids was not much greater than
the width of the main river channel.
32. The minimum flow in the river occurs in April, but
the level rises gradually as the rainy season begins. The
main flood rise occurs in July and the level continues up-
wards until late September or October when peak flood level
is reached. In October and November the water quickly
recedes and from then until April the flow gradually
diminishes (figure 1). The regularity of the peak flood,
by measurements over 20 years, has been shown to be very
constant, varying only a few days before or after the end
of September (Bird Report, 1949).
Major Tributaries of the lower Volta.
33. The Dayi river joins the main stream approximately
25 miles north of the projected dam site. The stream is
rather small, at the intersection of the Kpandu-Huime road
being only 50 75 feet in width and varying from 3 6
feet in depth. In August, the river ran about 20 feet below
the steep, clay banks. About 200 yards upstream from the
road crossing the river, it was somewhat wider and shallower -
not more than three feet deep. The water was relatively
clear and the bottom was sandy. Another 100 yards upstream
there is a large rock outcrop that traverses half the stream
bed and deflects the water to the west. In the main
channel the water seemed to be deep and swift flowing;
however, I did not investigate this because of the reported
presence of a large crocodile. I also visited the Dayi
just north of Hohoe where it was much shallower than at
Huime. Its character otherwise was much the same.
34. The Asuakawkaw is about 75 miles due north of Ajena.
The river is small at the ferry crossing on the Kete Krachi
road, where it is about 100 feet across, six or seven feet
deep and slow flowing. The slow flow was probably due to
the proximity of the Volta.
35. Continuing northward, the Oti is the next river met
on the east side of the Volta. It joins the main stream
about 20 miles north of the Asuakawkaw. The Oti is large
and deep and, in August, had a strong current strong
enough to be used to carry the ferry across the river.
At the ferry crossing it is,about 300 400 feet wide. I
also visited the Oti at Atafie, which is almost due east of
Kete Krachi. In this area the banks were 15 20 feet above
the river level. The water was flowing swiftly in a channel
that is about the same width as at the Otisu ferry crossing.
The water was very turbid and even at the banks was so
cloudy that I could not see the bottom. Of all the
tributaries of the Volta which I examined, the Oti was the
largest by far.
36. The Daka river was visited on 22nd August at the
point where it intersects the Chindri Salaga road at
Grubs. The river-was about 150 200 feet wide at the
Grube drift, a bridge which is open only at low water.
It seemed to be rather shallow, but had a swift current
in some parts, particularly over a primitive, native-built
rock dam that was built just below the drift. No other
rocks were noted in the stream. The banks were about 20 -
25 feet above the river level.
37. The river Pru enters the Volta just south of Yeji.
When I saw it at Prang, in October, the Pru was running
well above its August level, but because of the low 1950
rainfall, it could not be considered as being in flood.
The stream was about 75 feet wide and apparently 6 to 8
feet deep, based on estimates made when it was observed in
August. The current was swift to turbulent at a rock out-
crop, but mostly the flow could only be considered as
moderate. The water was very murky and gray in colour;
during the low stage, on the other hand, it was entirely
38. I also visited the Pru where it crosses the road
between Ejura and Nkoranze (Kintampo road). The stream
was very small here, being not more than 20 feet wide and
3 feet deep. There was a heavy rainfall the day before my
arrival and so the current was strong. Inquiry revealed
that just prior to the rain the stream was only ankle deep
and about 3 feet wide.
39. Because of their inaccessibility during the rainy
season, I was unable to travel to either the Sene or the
Obosum; however, I feel reasonably certain that they
strongly resemble the Afran, described below. The Sene
joins the Volta west of Kete Krachi and the Obosum enters
from the West a few miles north of Kpandu.
40. The Afram river is the southernmost tributary of
any importance. It rises in the hills forming the southern
boundary of the Voltaian basin, and flows into the Volta
some twenty miles north of Ajena. At Mankrong, a village
of three houses, located on the banks of the Afram, the
river is about 100 150 feet wide. A large outcrop of
boulders, apparently of a coarse conglomerate, extends
completely across the stream and is used as a footpath by
the natives. At some points it was ten feet above the
water level (on September 13) and it was breached at
several places to allow the passage of water. Just above
and below the rock outcrop, the river is deep enough for
canoe traffic, but about 75 yards upstream a geAtle cascade
stretches for some 300 400 feet. The water at this time
was pouring over the rocks rather swiftly, but it was clear
and contained little suspended material. The river banks,
either rocky or sandy, stood about 12 15 feet above,
the river level. At low water, the stream is shallow
enough for lorries to cross; whether flow ceases completely
in the dry season, I was not able to ascertain.
41. I also saw the Afram at Achi, a small village about
one mile from Adwaso. The river was not more than 75 feet
in width and about knee deep. The bed of the stream was
of shale and the water flowed smoothly over it at a moderate
rate. The southern bank was about 15 20 feet high while
the northern bank stood no more than 5 8 feet above the
42. On 14th September I was able to get to the Afram
at Nkatepa. From the village to the river the descent
was rather steep, the path leading over limestone outcrops.
The river is about 200 feet wide and apparently deep in
this region. The current was very slow so that a
- completely different picture was obtained from that at
Mankrong and Achi.
43. Most of the smaller streams have little vegetation
in them. Usually in the slower sections some grasses may
become established but for the most part the beds are barren.
Of course when the streams are in flood, some of the marginal
vegetation may become partially submerged, but these plants
must be considered as only transient elements of the stream
flora. The only small stream in which I found vegetation
of any consequence growing in its bed was the Koloe, a
small tributary of the Dayi river. Dense clumps of a fern-
like plant were flourishing above a rock outcrop which
partially dammed the stream.
44. The Volta, itself, has marginal grasses growing
sparsely near the banks and in the shallower water of the
rapids. Fringing trees and bushes dip their branches into
the water and in the protection thus offered water lettuce
becomes established, but I have not encountered it growing
with the success it enjoys in small ponds. At places in
the rapids, wherever the grasses have an opportunity of
becoming rooted, thick mats may be formed, As the level
of the river rises with increased rainfall, some of the
hydrophytic trees are partially submerged, and, at times
submergence may be extensive, as can be seen by the presence
of debris caught in the branches many feet above the low-
level stage of the river.
45. At Kete Krachi I found a plant resembling Myriophyllum
growing in.the quieter water near the rapids. Because of
the turbidity of the water I was unable to determine the
extent of the beds of this plant. Also where the water ran
shallowly over the river bottom a species of moss was
closely adherent to the entire upper surface of the rocks.
The same moss was found on some of the rocks taken from
water 3 5 feet deep in the rapids at Kpong. Apparently
the plant grows well on the rocks at low water when
sufficient sunlight can penetrate the water for normal plant
growth, but at the time of my examination the turbidity of
the water was so great that I suspect that the plants were
suffering from a lack of sunlight. .This was reflected in
the rather limited and small growths in the Kpong rapids as
compared with the success of the plant at Kete Krachi where
only a few inches of water covered the rocks.
VEGETATION OF THE VOLTAIAN BASIN.
46. Nash (1948) classifies the vegetation of the Gold
Coast in three categories, namely: the secondary coastal
thicket which may be either the dry, coppicing thicket of
the littoral, or thicket, transitional to forest; the
forest, of which there are two types, the evergreen and the
mixed deciduous; and woodland savannah.
47. Typical fringing, riverine vegetation lines the.
shores of the Volta except in a few places where it has
been slashed by farmers, or where villages are located.
Back from the river on the east side above Senchi, the main
road passes through open country that can be classed as
woodland-savannah. Baobabs are a conspicuous element of
the landscape. Continuing northward the woodland-savannah
gradually merges into mixed deciduous forest and at the
Asikuma-Agu road there is an even more pronounced change
wiTh the vegetation becoming much more dense. In spite
of the fact that the June-July rains had only recently
ended, every stream valley in the region seemed to be
absolutely dry during my visit in August. From Agu north-
wards to Anum the road passes through heavy forest inter-
spersed with cocoa farms.
48. The road north continues through mixed deciduous
forest, but just north of Kpeve, the country again becomes
more open and the vegetation takes on the appearance of the
woodland-savannah. Closer to the Volta, however, the
typical heavy forest is present. The. lower section of the
road from Kpandu to Kete Krachi again traverses mixed
deciduous forest as far as the Asuakawkaw river. Above the
river typical woodland-savannah makes its appearance and is
characteristic of the country all the way to its northern
49. Other than the forest conditions described above,
the greater part of the Volta and its tributaries are
situated in woodland-savannah. Only the head-warters of
the Afram and some of those of the Pru originate in the
mixed deciduous forest. The remainder of the streams
derive their waters from the woodland-savannah. All of the
small streams feeding the Afram which were examined by me,
rise in the uplands which are covered by the heavy forest
50. One of the most striking vegetation changes that I
observed during my travels over the river system was that
between Kwahu-Tafo and Adwaso near the Afram. There is an
extremely sharp break from the mixed deciduous forest of the
Kwahu plateau to the woodland-savannah vegetation as soon as
the road drops below the crest of the plateau on the north
side. On top of the plateau the air is moist and cool, but
with the change in vegetation there is also a change in the
atmosphere and the increase in temperature is immediately
CLIMATE OF THE VOLTAIAN BASIN.
51. The report by Junner and Hirst (1946 p. 8) describes
the climate of the Voltaian basin as follows:
"The southern part of the basin, south of the
latitude of Kintampo, is within the Guinea zone of equatorial
climate, which is characterized by a marked double rainy
season, high and fairly constant temperature and humidity,
and a high degree of wetness; and the northern part is in
the Sudanese zone which is characterized in this region by
a single rainy season followed by a long dry season, and by
a lower rainfall and greater range of temperature and
humidity than in the Guinea zone.
"In the Guinea zone the average rainfall ranges from
54 73 inches, January is generally the only month with
less than one inch of rain, and the average number of days
per year with a rainfall of 0.01 inch or more is 80 125.
The mean shade temperature is 750 800 and the mean
relative humidity is over 80 per cent ......
"In the Sudanese zone the average rainfall is
41 49 inches and the average number of days with a rain-
fall of 0.01 inch or more is 70 94. The rainy season
lasts from May to September or October and there are three
months (December to February) with less than one inch of
rain in the south and five (November to March) in the
north. The average annual mean temperature is 82- 830, the
average range of temperature 200- 25, and the average
humidity at 9 a.m, 60 75 per cent.
"The value of the rainfall in the central and
northern parts of the basin is much less than the rainfall
figures suggest; run-off during the rainy season is high
and the desiccating effect of the harmattan, which starts
shortly after the end of the rains and lasts for two or
three months in this part of the basin, is high. Evaporation
is at a maximum at this time of the year and from meapure-
ments at Tamale dam, the only place where such measurements
have been made, it would appear to average 9 12 inches
per month during January to April compared with about 4
inches per month during August to October. The total annual
evaporation at Tamale is approximately 7 8 feet."
52. The rainfall, temperature, and relative humidity in
various areas of the Voltaian basin are shown graphically
in figure 2. These graphs are from the 1949 edition of
the "Atlas of the Gold Coast".
DISTRIBUTION OF POPULATION IN THE VOLTAIAN BASIN ,
53. Map 3, adapted from the population map of the
"Gold Coast Atlas", 5th edition, 1949, shows the present
distribution of the inhabitants of the Gold Coast. It is
rather clear from a glance at the map that great areas
which will be affected by impoundment at Ajena, are either
uninhabited or so lightly populated that new or aggravated
problems will be inconsequential, in them. Such is the case
in the great Afram plains and the valleys of the Obosum and
the Sene, where the average density of population varies
from 0 10 per square mile.
54. North of Kpandu there are only scattered and
isolated areas in the main Volta system where the population
density is sufficiently high to warrant consideration in
proposing measures for dealing with extensive control of
disease-bearing insects. As new villages are established
on the reservoir shore they, too, must be considered in
the overall plan.
THE DEVELOPMENT OF THE VOLTA RIVER BASIN.
55. It has been proposed that a dam be constructed
across the Volta river in the vicinity of Ajena for the
purpose of producing hydro-electric power. Much of the
power would be used in the production of alumina and
aluminium at factories situated near the power station or
the harbour. The plan includes the transportation of
bauxite from the Gold Coast mines, the building of a
harbour, the extension of existing railways, and the
possible development of river traffic.
56. After impoundment of the river is completed, a
reservoir area of about 1850 square miles will be created
(Map 2). It will extend north from Ajena approximately
to the confluence of the Black and White Voltas; however,
the upper limits still remain to be determined accurately.
The reservoir will follow the course of the river and the
lower parts of its tributaries, and will form a long narrow
lake with its greatest width where the Afram joins the Volta.
At this point it is estimated that the lake will be about
twenty-five miles in width. Northward it-will vary from
one or two miles to more than five miles in width over much
of its length. Near Kete Krachi the lake will become much
narrower and from this region north to the end of the
reservoir area, will narrow still more until it is confined
to the present river channel.
57. Malaria is one of the most serious diseases of West
Africa, where almost the entire population is infected.
This section of the world has long been noted for the
virulence of malarial attacks and for the high mortality
rate produced by the disease. All four species of human
malaria are known to occur in the Gold Coast, but Plasmodium
falciparum is far more common than the other species. To
the best of my knowledge, no comprehensive surveys of the
prevalence of malaria have ever been carried out throughout
the entire country; however, many local surveys of the
incidence of the disease have been made and have shown
repeatedly that malignant tertian malaria is almost
universally present. It is most readily seen in young
children, but in older individuals is more difficult to
demonstrate. It is also well known that in the Gold Coast
the population has developed some immunity to the disease
and that only an occasional light fever is experienced once
the immunity has been acquired. It is also well known
that malaria is one of the most important contributors to
the high infant mortality rate in the Gold Coast.
58. Dr. L.Chwatt (1949) says "it can safely be assumed,
on the basis of past malaria surveys, that Africans,
living in a hyperendemic area, where the transmission of
malaria is practically perennial on account of favourable
climatic conditions, acquire a good deal of immunity to
the main strains of the two most common parasite species
(P. falciparum and the much less frequent P. malariae).
TTs immunity, acquired at the cost of the supposedly high
mortality of infants and small children, is sufficiently
high to consider malaria disease to be a relatively un-
important cause of morbidity and mortality of the adult
population and to make disastrous malaria epidemics a most
"Whether this acquired immunity is absolute or not
is still debatable. There is little evidence of malaria
causing an appreciable morbidity of the adult population
in the southern parts of West Africa. The out-patients at
the hospitals with malaria hardly ever exceed 5 per cent (in
the south) and the in-patients hardly ever exceed 2 per cent
of all admissions."
59. Chwatt further says, "One more point, justly
emphasised by Milligan (in litt.) must be very seriously
taken into account. Any large scale engineering project
in a malarious country is invariably followed by the
'tropical aggregation of labour' and mixing of people
living in different and often widely separated parts of
"The acquired immunity to malaria is not only
species specific, but also strain specific. This means
that a person immune to one or several strains of a given
species of malaria parasite may not be immune to a different
strain of the same species and consequently may develop
clinical malaria after the infection with a foreign strain...'"
60. The importation of a large non-immune European
population into a hyperendemic malarious region in West
Arica definitely imposes a large amount of risk, particularly
as this group, composed of essential personnel, can be
completely immobilised by an epidemic of malaria. Because
of this possibility and the great danger of exposing migrant
African labour forces to a strain of malaria to which they
have no immunity, this dangerous disease must receive con-
siderable attention in any plans for impoundment of the
61. Malaria has long been associated with the impoundment
of water. The most comprehensive and complete summary of
the disease as related to the construction of dams is the
recent work "Malaria Control on Impounded Waters" prepared in
1947 by the Health and Safety Department of the Tennessee
Valley Authority. There is no need here for me to review
the previous problems of this nature. Suffice it to say that
any impoundment of water in a tropical or subtropical country
has an influx of malaria as its corollary, unless intelligent
anti-malaria planning goes hand-in-hand with the construction
62. Only three species of Anopheles mosquitoes are of
any real importance in the transmission of malaria in West
Africa. These are Anopheles gambiae gabiae Giles, A.
gambiae melas Theobald, and A. funestus Funestus Giles.
That other species are potenFTal vectors in the region is
well known and A. hancocki Edwards, A. hargreavesi Evans,
A. nili Theobald, A. pharoensis Theobald, which are present
n- T~e Gold Coast,"Eave all been shown to be capable of
harbouring natural infections.
63. Of the species of mosquitoes listed above, A. gambiae
gambiae is the most important malaria carrier. It T- also
the most ubiquitous and the most homophilous. A, gambiae
melas, the coastal form, is generally similar to gambiae
in its ability to transmit malaria and in its adult habits.
A. funestus is likewise a very effective vector and also a
species which can be considered as a domestic one. The
other species are of such limited importance that I shall
not consider them further in this report.
Anopheles gambiae gambiae Giles
64. This species of mosquito is one of the most efficient
vectors of malaria in the world. Its ability to establish
itself in new areas is well illustrated by its transfer to
Brazil, and the effectiveness as a malaria transmitter by
the epidemic of the disease which resulted after its
establishment in that country. At the present time,
A. gambiae is one of the most feared insects of medical
"Tportance and many precautions are taken in this day of
rapid air travel, to prevent its spread into areas where it
does not now exist.
65. The remarkable adaptability of A. gambiae lies,
for the most part, in its ability to in-abit a great
variety of breeding places. These have been adequately
described by Evans (1938, p. 312). ".....in most parts of
its range it is characteristically a small pool-breeder;
its breeding-places are in general at least partially
exposed to direct sunlight. The word 'pool', as here used,
includes puddles, shallow ponds; borrow (or murram) pits;
animal footprints; isolated collections of water in road-
side and other drains, ditches, irrigation furrows, seepage
areas, and the drying-up beds of streams or near the
edges of lakes and swamps. The shallow overflow from
streams or drains may also be an important source of the
66. Dr. Evans continues, "It has often been observed
(Anderson, Demeillon, Leeson, and others) that many prolific
breeding-places of this species are created by human
activities. Such sources include: pools on surface or in
drains of badly constructed roads; borrow pits; puddles,
etc., caused by cattle, villages, or even gold miners at the
edges of streams and water holes; cutting down bush shading
streams; damming of streams without keeping banks deep and
67. Normally in temperate zones larviciding treatment for
Anopheles control is given every seven to ten days, depending
on the climatic factors and rate of development of the
insects. The rate of development for A. gambiae has been
studied in several parts of Africa and in Brazil and highly
variable results were obtained ranging from 6 days to 15 days.
However, in 1944, while I was serving in the American army as
entomologist for the Interallied Malaria Control Group at
Accra, I found that under exceptionally favourable field
conditions it was possible for the species to complete the
cycle egg to adult in as little as 5 days and that a 6 day
cycle was frequently the case.
68. Along with the ability of the species to adapt itself
to such a variety of breeding places, its preference for
human blood meals enhances its potentialities for transmitting
malaria. It has been found by several investigators that
when offered the choice of animal blood or human blood,
almost invariably the A. gambiae females preferred that of
humans. Precipitin tesls on stomach contents of wild A.
gambiae have shown that they, too, in the majority of cases,
have fed on human blood.
69. Dissections of A. gambiae females in Africa have
proved that actually -it is a highly dangerous malaria
vector. According to Demeillon (1947) natural infectivity
rates ranging from 0.7 to 30.2% have been found. I do not
have the figures available, but I recall that in our work
at Accra, we found infectivity rates of 4 to 5%.
70. Because temperature is subject to such small fluctua-
tions in We'st Africa and is well above any temperature that
might inhibit breeding, mosquito production occurs throughout
the year. AA might be expected, the adults of A. gambiae
are far more numerous during and shortly after 7Ee rainy
season because of the greatly increased amount of water
available for breeding. It is not the water area in itself
which is important, but the tremendous increase in-shore
line, for the larvae tend to remain very close.to the edge
of the water.
71. Experiments conducted by P.C.G.Adams (1940) at
Nkana, Northern Rhodesia, demonstrated that A. gambiae
could fly 4,25 miles down-wind. According to Evans (1938),
Symes working in East Africa, discovered the species would
fly 3 miles. At Lagos, Barber and Olinger (1951) concluded
that.the adults were flying over a mile. Again at Acora in
1944, in our control project, we began the control
programme with the assumption that the adults would certainly
fly no more than 2 miles. In spite of what we.felt was a
very effective control programme, our catching stations
were still producing adults. Before we were completely
preventing mosquitoes from entering the area to be protected
(army camps), it was necessary to extend our control to a
radius of 5 miles. The country in the Accra region is very
open, and it is true there is little to impede the flight of
the mosquitoes. In heavily forested country, on the other
hand, it is very likely that the range of flight.would be
Anopheles gambiae melas Theobald
72. The melanic form of A. gambiae is primarily a coastal
species and its breeding is associated with brackish water.
There have been instances where a melanic form has also been
taken inland, but these are relatively few. Whether the
coastal form and the inland form are conspecific still
remains to be proved.
73. Many of the remarks about A. gambiae will also apply
to melas; however, its selection-of breeding places is
rather different. A. gambiae melas breeds in brackish tidal
marshes where Paspalum grass is abundant, as well as in man-
grove swamps (Avicinna sp.). Dr. C.R.Ribbands (1944) has
studied this species extensively at Freetown, and there he
found that the larvae were able to complete development in
water with a salt concentration of 150% sea water. On one
occasion, when I was working in Accra in 1944, I obtained
some larvae of melas to ascertain whether they could live
in sea water. A sample of water was collected 10 miles from
shore and the larvae were transferred to it. Every one of
the larvae which was so treated completed its development
in the sea water, and adults were reared. I do not claim
that it is normal for the species to develop in sea water,
although larvae are often found in rock pools along the sea
shore; however, these examples do show the tremendous range
of breeding sites that are available to A. gambiae gambiae
and A. gambiae melas.
Anopheles funestus funestus Giles
74. This species is next in importance to A. gambiae
gambiae as a malaria vector in-West Africa, were it is
widely distributed. A. funestus is a less ubiquitous
mosquito than A. gambTie and the latter species is often
found in the ab-ence of the former. Evans (1938) states
"characteristically in bodies of clear water that are either
large and more or less permanent, e.g. swamps (near edges,
if deep) weedy sides of streams, rivers, furrows or
ditches, protected portions of lake-shore, ponds, etc.,
especially when weedy, or water such as seepages, which are
fed from underground permanent sources."
75. In addition to the above listed habitats can be added:
disused and overgrown borrow pits; hoof marks among grass
in marshy land; and flooded rice fields. Exceptionally,
they are sometimes found in deep wells. However, in West
Africa, the most important breeding sites are those
mentioned in the quotation from Evans.
76. The larvae are not sun-loving as are those of
gambiae and usually require some shade. They are
frequently found in ponds and at lake shores where Pistea
stratiotes (water lettuce) is growing, and the larvae can
also be seen in ponds and swamps, partially covered with
77. Just as in the case of gambiae, funestus is addicted
to human blood, and is likewise found in considerable
numbers in African houses where many adults tend to rest
during the day. That it is an efficient malaria vector
has been proved by dissections of many adults. Results
of these dissections showed a natural infectivity rate of
from 0.24 to 27%; however, I believe the last figure must
certainly be exceptional and that of 4 to 5%. the same as
gambiae, is more typical in West Africa. The greater
importance of gambiae in West Africa lies in the fact that
it tends to be more widespread-and adaptable to a greater
number of habitats.
78. Adams (1940) recorded flights of funestus of 4/2
miles with the wind, and Demeillon in 1954 captured 2 females
at a distance of 43 miles from the nearest breeding place.
He differentiated, however, between the maximum flight
range and maximum effective range, and concluded from a
careful study of adult populations that 80 per cent of the
adults occurred within a half-mile radius of the breeding
79. The abundance of funestus is most certainly linked
with rainfall in West Africa. According to Evans (1938,
pp. 158-159) it was observed by Garnham at Eisumu that
population levels are high throughout the year because the
species breeds in permanent bodies of water. At Kampala,
Uganda, "two peaks of maximum density of this species occur,
'the first in the dry season, the second shortly after the
onset of the long rains.' Symes (1932) distinguishes two
types of seasonal fluctuation of the species in Kenya; in
the first type the prevalent breeding places are surface
swamps and small trickling streams, and in such localities
'the greatest output, occurs during the height of the rainy
season'; in .the second type the species breeds mainly in
bigger streams and 'the peak of the numbers occurs after
the main rains, when the rivers have settled down to a
80. During the course of this survey I made many
collections of mosquitoes at villages located on the banks
of the Volta. Collecting was carried out from Ada to Yeji
by the use of the spraying technique described in the
section on methods above.
81. Results that were obtained were not quantitative
because I did not feel that I could devote sufficient time
to the problem to derive such data. My main object was
to establish the presence of the vector species, and this
was done without question.
82. At Ada, as expected, only A. gambiae, probably
melas (many with 4-banded palps) among the anophelines was
taken. Away from the coast at Kpong, both A. gambiae and
A.funestus were collected, the former in faF-larger numbers
than funestus; some nili were also collected. The same
held true at Akwamu West, Ajena, and Misikrom, except that
the proportions of funestus and nili were somewhat greater
in relation to gambiae.
83. In spite of the fact that the houses at Misikrom were
built in a primitive manner with neither doors nor windows,
but only openings where the doors should be, and the fact
that there was no ceiling to the rooms, the thatched roof
being several feet above the wall, by the use of a single
aerosol sprayer, I was able to collect 181 adult A. gambiae,
32 A. funestus, and 12 A. nili from a room only 8--by 10'.
Without doubt at least 3~-or-4times this number of mosquitoes
must have been hiding in the room and the great majority
escaped or died outside the walls following spraying.
Collections at Misikrom were made in early August.
84. At Kete Krachi the picture was repeated with A.gambiae,
A. funestus, and A. nili occurring in great numbers,
Sgambiae being mech more numerous than the others. In one
small room more than 290 adult Anopheles were collected,
while in small, round, thatched huts the average ran at
nearly 150 adults. Similar results were obtained at Yeji.
85. Iarvae of the various species of Anopheles were found
breeding in the expected places. Along the shaded, or
partially shaded, banks of the Volta in the backwaters
where the water was relatively quiet, A. funestus larvae
were thriving. At Kete Kraohi, where Ehe V1lta passes over
a great expanse of exposed rock, the river had been a foot
or more above the level at the time of my visit in late
August. Isolated pools, left behind by the receding river
furnished ideal places for the production of innumerable
A. gambiae. I found a similar picture at Kpong where
depressions in the rocks projecting from the rapids were
filled with rain water. A.gambiae larvae were present.
The same was repeated at Atikpeta.
86, Although I knew that some marginal swamps and
marshes existed along the course of the Volta at certain
times of the year, I was able to examine only two. It so
happened that my travels took me to the areas shown on the
maps as being swampy at times when they were completely dry.
However, the two I did examine produced larvae of
A. funestus as expected. The largest of these is the
extensive marsh between Akuse and Amedika. In July when
I visited the area, although the heavy rains were
supposedly just ended, the marsh consisted only of isolated
pools scattered through the grass. It is likely that with
the rise in the level of thq river in late September that
the marsh is very extensive. A second marginal swamp was
located just north of the village Akwamu West. I collected
A, funestus larvae there in early August at which time the
villagers informed me that they were draining the swamp by
digging a ditch to the Volta, a short distance away. I
returned to the swamp late in September, when the river was
almost at flood level, and found it completely dry.
87. My searches for mosquito larvae did not always-
reflect the same picture as adult collections, and I am
relying more on the adult collections to draw my conclusions
as to the prevalence of Anopheles. Wherever water was
reasonably accessible, I did look for larvae and almost
always my search resulted in finding the insects.
88. In my opinion, it can certainly be concluded that
A. gambiae and A. funestus are widespread and numerous at
all times of the year throughout the Volta river system,
except at the mouth, where only A. gambiae melas is common.
Therefore, -hese two species mustbe considered as one of
the primary problems in dealing with insect vectors in the
Volta river project.
General Discussion and Recommendations.
89. Following impoundment of the Volta, ecological and
biological changes in the river will be profound, as the
riverine flora and fauna will be converted into lacustrine
types. Naturally, mosquito populations will be affected
favourably and there should be an even greater number of
the pests produced than is now present. With the cessation
of flow, the banks of the new lake should become overgrown
with fringing trees and bush and the littoral zone will be
invaded by vegetation characteristic of standing water.
This will be particularly true in the areas where the lake
banks are not steep and where trees are present in the
90. Conditions such as may be expected following
impoundment were observed just north of Kete Krachi where
the Volta is very wide and relatively static because of the
damming effects of the rock outcrop which produces the
shallow rapids near the town. In this section, trees were
growing in profusion down to the water's edge. Many of
these trees characteristically had low hanging branches
which even dipped into the water. Grasses had invaded the
quiet shallow water and near shore, water lettuce was growing
in clumps. Such conditions are conducive to the production
of mosquitoes, particularly some of the anophelines, as has
been demonstrated repeatedly in the United States as well
as in Africa.
91. In reaching this expansion of the Volta I followed
a native trail which ended at the river bank. Trees had
been cut away from the area and only a small amount of
emergent grass persisted in the water for about fiftty feet
on either side. Although there was little wind, wave action
was noticeable and no mosquito larvae were seen in the
92. Granted that ideal conditions for breeding of
Anopheles funestus will be produced by impoundment and the
consequent invasion of the shore zone by vegetation, need we
be concerned with A. gambiae, a species not associated with
vegetation in the Iarval stage? I feel that breeding of
this mosquito species will be a major problem. With the
change in the character of the river and the flooding of
extensive areas of land, the problems created by the
impoundment at high water will be no greater than they are
at present, which of course are considerable. On the other
hand as the level of the lake falls with the drawdown, the
countless. depressions at the shore zone will each retain
water, probably for some time, especially since the ground
will be saturated. Each of the pools will potentially be
able to turn out hordes of A. gambiae.
93. As the reservoir fills toward its high level,
marginal grasses will become established and will flourish
for several months until the drawdown proceeds rapidly
enough to prevent germination of the seeds. That means
A. funestus will be favoured from September through
Tcember, but with the fall in the water level becoming more
rapid in January and continuing to June, A. gambiae will
find ideal-breedifg grounds. There is no doubt that the
number of A. gambiae, which elsewhere are at their lowest
level of aB'uEdance during the dry months, December to mid-
May, will, in the region of the Volta reservoir, greatly
increase in numbers, and that the malaria potential will
thereby also be tremendously boosted.
94. In the villages along the Volta, I have observed
that the people, through a lifetime of association, have
become so accustomed to living with A. gambiae and A. funestus
that they are no longer conscious of-iheir presence"-uring
the dry season. I have already pointed out above that
the non-migrant African has developed a partial immunity
to his local strain of malaria. During the rains, with the
influx of hordes of culicines as well as anophelines, the
Africans do become aware of the presence of the insects and
are anxious to obtain any relief from the pests.
95. It is rather amusing to note the surprise on the
face of an African householder when he is shown the number
of mosquitoes caught in his bedroom. Almost invariably when
I approached a village for the first time and requested
permission to look for mosquitoes I would be told that there
were none, or very few. It was always with reluctance that
permission was granted me to investigate a bedroom. However,
when I could demonstrate the presence of 50 300 mosquitoes,
mostly Anopheles, in a small bedroom, requests were then
showered on me to spray every house in the village.
96. Then, will impoundment modify, aggravate, or
change the malaria picture in the villages along the Volta
reservoir? Available information indicates malaria
infections are universal and the immunity of the African
permits him, as an adult, to live with the disease with
only occasional discomfort. On the other hand it has been
demonstrated that malaria is particularly serious in
infants and is a major contributor to the high mortality
rate among them. No accurate statement as to its real
effects can be given for no accurate vital statistics are
maintained in the less accessible villages, and these make
up an important segment of the population under consideration.
Assumed that cognizance is taken of the price of acquiring
an immunity by the African, and measures are taken to
reduce the disease; two major additional problems will
then be created:
(i) The Africans will lose their immunity to
malaria, and, consequently, if they travel to a
non-malaria free area, they may be subject to
severe or fatal attacks of the disease. They
will also be seriously affected if a malaria
epidemic were to occur.
(ii) With the reduction of the infant mortality rate,
populations will increase with all the attendant
The question of malaria control is not simple, but one with
decided complexities and with great social implications.
97. Assuming that the villages which are to be flooded
will be relocated in the same general area in which they
are now sited, I do not believe that there will be any
change in- the malaria picture. But, as construction of
the dam gets under way, large numbers of migrant labourers
will be moving into the Ajena area. If these migrants are
all immune to the local strains of malaria, there will be
no problem, but if there is no immunity, many man-days
will be lost because of illness. A second consideration,
and one which needs much study, is the possibility of
migrant workers bringing foreign strains of malaria into
the Ajena area. Since the local inhabitants will not be
immune to this strain, serious malaria epidemics may result.
It is likely that the same strains of malaria exist through-
out West Africa, but, as far as I can determine, this is not
known with any degree of certainty.
98. Chwatt (1949) points out that when two battalions
of African soldiers were brought from the Congo and were
stationed for about one year in Central Nigeria, malaria
cases (P. falciparum) were three to four times as high as
in Nigerian troops and "the number of severe cases was very
striking. And yet most, if not all of the Congolese came
from a part of Africa where malaria, is endemic and most of
them must have been immune to the Congo strain or strains
of P. falciparum.....,
99. The importation of a large European technical and
supervisory force into the Ajena area will create a problem
of some magnitude from the malaria standpoint. None of
these persons will have any immunity to the disease, and they
will be living in an hyperendemic area where it is a
certainty that they will be bitten by an infected Anopheles.
For that reason it is imperative that all possible steps be
taken to protect the workers from -the bites of mosquitoes.
100. I have indicated above something of the situation
with regard to malaria and malarial mosquitoes as it now
exists in the lower Volta and the problems that will be
created following impoundment, provided no steps are taken
to control the insects. What steps should be taken, to
eliminate the menace of malaria and the malarial mosquitoes?
It is a very serious problem and must involve a considerable
amount of study and planning. Fortunately, planning for
malaria control can go hand in hand with the planning of the
reservoir preparation and the construction of the dam.
101. The Tennessee Valley Authority, in constructing
reservoirs, has used the scheme of planning malaria control
at the inception of the programme, and such farsightedness
has paid tremendous dividends in improving the health of the
local inhabitants, raising their standards of living, and
reclaiming land. There is no reason why such planning should
not be used on the Volta project.
102. The T.V.A. recommends that the following staff be
employed for malaria control:
an engineer (experienced in malaria control-
on impounded waters)
This staff would be an ideal one for planning and executing
malaria control in the Volta reservoir and below the dam.
All of the men must be experienced in the work and preferably
with experience in the tropics.
103. One of the first items that must be determined by
the malariologist is the exact limits of the area in which he
plans to control malaria. It is perfectly obvious that it
would be impractical and unrealistic to attempt to control
mosquitoes throughout the entire reservoir area. In fact
those areas which will need to be considered will make up
only a small part of the whole, provided mass migrations
into the marginal lands do not occur. In my opinion, it will
be practical to employ modern mosquito control measures only
in those parts of the reservoir bordering on the larger
centres of populations. By that I mean villages and towns.
It would not be feasible to try to control A. gambiae around
each farm house, many of which are widely separated from
104. Without question, the first area that must be
considered is that zone where construction is in progress.
It is imperative that the staff be given every possible
protection. For that reason, I recommend that the malaria
control staff listed above be employed without delay.
105. I would strongly recommend that the experimental
programme of mosquito control by residual house spraying
now being conducted at Ilaro, Nigeria, by Dr.Leonard Chwatt,
be closely watched. His technique may prove to be ideally
suited to mosquito control in more remote areas and offers
promise of being effective and cheap. He is continuing the
experimental spraying of Ilaro for a period of three years
but I believe that by March, 1951, his preliminary results
may permit him to draw tentative conclusions as to the
effects on A. gambiae and A. funestus. These two species
of mosquitoes are ideally suited for control by residual
sprays as both are so strongly domesticated.
106. The search for new drugs which act as malaria
prophylactics has made great strides in recent years. Until
adequate mosquito control measures have been instituted, and
are effective, all European personnel should be required to
take the proved brands of these prophylactics or suppressives.
107. In the construction of houses for Europeans, I
strongly recommend that they be completely screened with
eighteen mesh screening, through which the local species of
Anopheles cannot pass, and that mosquito-trap doors be
installed at all outside entrances.
108. The T.V.A. Manual (1947, p. 35-36) points out that
"adequate preparation of the reservoir basin is fundamental
to the achievement of satisfactory mosquito control after
impoundage......The objective of reservoir preparation
measures is to clear, drain, and otherwise prepare the
basin so that a clean water surface and a clean shore line
will exist throughout the reservoir, and so that all marginal
pounds, sloughs, and depressions will fluctuate freely with
the main lake." The malaria control engineer should,
therefore, take steps to prepare the reservoir area in
accordance with presently established and proved practices.
109. In addition to preparation of the reservoir area
in controlling mosquitoes, larviciding must still play. an
important part, if it is shown that mosquitoes cannot be
controlled by residual house-spraying alone. In such a
vast control area, the use of an aircraft will be a most.
valuable adjunct to ground crews in spreading the
insecticide. The malaria control engineer will require
boats for his gangs to get to those parts of the
reservoir inaccessible by road. He will need store houses,
as well as a completely adequate supply of equipment. He
should be given every assistance in a difficult job.
110. The malariologist and the biologist (primarily an
entomologist) will require laboratory space, preferably in
the region of Ajena, or wherever permanent construction
headquarters may be established. They should be supplied with
the latest types of equipment for their work, which of course'
is absolutely essential in any well planned malaria control
111. A common practice in reservoirs, when it is feasible,
is to fluctuate the level of the water periodically as a
mosquito control measure. The purpose of this is to
eliminate or minimise the intersection line (intersection
of water surface by vegetation and floatage) in mosquito
breeding areas along the margins of the reservoir. Such
fluctuation may draw the larvae out from the protection of
the plants or may strand them on mud banks, or it may strand
the floatage, thus exposing the larvae to predators. This
method of control would be applicable against A. funestus,
but I do not believe it would be particularly effective for
control of A. gambiae. At any rate, Sir William Halcrow and
Partners have pointed out that only seasonal recession will
be used on the Volta reservoir and that it would be impossible
to employ the principle of periodical fluctuation as a
mosquito control measure.
112. In the United States where Anopheles quadrimaculatus
is the mosquito of major importance, it has been found that
a recession of water level of 0.2 to 0.3 feet per week,
without periodical fluctuation, is adequate to control the
mosquito in the Tennessee river reservoirs. Where periodic
fluctuation is used, a rate of recession of about 0.1 to 0.2
foot per week is generally satisfactory.
115. The drawdowns now being considered for the Volta
reservoir are shown in figure 1. From the peak level of
nearly 260 feet, the water may exceptionally recede to a
minimum of 250 or 205 feet,. depending on the final decision.
The daily and monthly drawdowns are also shown in figure 1.
It can be seen that the fall in the reservoir level from
December to May would fulfil the requirements for control
of the American mosquito, but, in my opinion, this small
drop would not be sufficient to stop the breeding of
A. gambiae which is independent of vegetation and shade.
114. A study of the one inch contour maps shows that in
some sections the reservoir will spread over great areas,
particularly in the Afram plains. The land here is low
and gradually shelving. If a drawdown to the minimum of
205 feet is employed for the production of hydro-electrio
power, then, in some sections of the Afram plains the
water will, in receding from the 260 foot level, be
drained from the land, exposing a zone extending from 205
foot contour some four to five miles on each side of the
reservoir. In much of the lower Afram valley the water
will be withdrawn for at least one mile on either side of
the lake. Admittedly, a drawdown of 55 feet will be a rare
occurrene, Normally, using either 230 or 205 feet as a
minimum, the reservoir level is not likely to fall much
below the 245 foot level. Under these circumstances, the
water will be withdrawn not more, and usually less than a
mile in the lower Afram. It has been calculated that with
the maximum drawdown from 255 O.D. to 205 O.D. about 1230
square miles of the reservoir area will be exposed. With
the drawdown to 230 O.D. the figure is only 640 square
miles. The major problems thus created in this vast area
will be in the residual pools left behind in the minor
depressions and the vegetation that will form a great marsh
in the shallower parts of the reservoir. It is fortunate
that the Afram valley is so lightly populated, for this
means that one of the major potential mosquito producing
areas will be of relatively minor importance.
115. In the Volta itself, the banks are normally rather
steep, rising well above the future reservoir level so that
in uch -of the reservoir the water will be well confined.
At certain points, however, the banks are bordered by
temporary marshes. With impoundment, these marshes will
become more numerous and more permanent.
116, From the malaria standpoint, do the dangers of the
Afram valley, then, overbalance the gains to the hydro-
electric programme in increasing the minimum drawdown level
by 25 feet? As far as mosquito control is concerned
there is little to choose between the two. The slight
differences in heights of the water will be of no significance
in the overall picture as the drawdown rates are roughly the
same and the normal low water level for both the 205 and
the 230 minima will be about 245 feet. The small fraction
of a foot in which the drawdowns differ can hardly make any
difference to such an adaptable animal as A. gambiae. I
can see no objection, therefore, to using *he greater draw-
down for the production of power.
117. In dealing with A. gambiae, footprints, hoofprints,
and other slight marginal-depressions must be considered.
If the. ground is saturated, these imprints will hold water
long enough to produce one or several generations of
A. gambiae. On the other hand, if the recession rate is
o rapid that the footprints retain water for less than
five days, such habitats will not constitute a problem
during recession, but this is not likely with a maximum drop
in the water level of only 0.3 foot per week.
118. At Yeji ferry, I was fortunate enough to be present
when a large herd of cattle was brought across the river.
This permitted me to observe the length of time that water.
would be retained in footprints during the falling stage of
the river level. The cattle crossed the river on the morning
of 12th October. Shortly after, I noted that water had
seeped through the saturated ground and had filled those
footprints that were within three to five feet of the
water's edge. On the night of 12th October there was a
rather heavy rain and the next morning the level of the
river was the same as that of the previous morning. All
the cattle tracks were still filled with water. However,
when I again visited the river's edge on the morning of
14th October the river had dropped about 6 inches and now
nearly all the hoofprints were dry, or well on the way
to becoming dry. This is fairly indicative that a
rapidly falling water level in the reservoir will dry
the marginal land-so rapidly that where the slope is
sufficiently great and the soil satisfactory, footprints
will be no problem.
119. The footprint problem will be a real menace during
the periods of constant level, no matter what the drawdown.
If the minimum 230 foot level is used, the reservoir will
remain relatively constant from mid-May to mid-July. At
the 205 foot minimum level, the constant stage will last
from early June to mid-July. Another bad period will be
the constant high level, which will be maintained from
late September to late October. Any malaria control scheme
involving A. gambiae must take cognisance of the habit of
this specie of breading in small, transient, water holes,
and the multitude of footprints along the shore of the
reservoir will be one of the major considerations.
120. During the constant level periods and slow recession,
grasses will undoubtedly invade the marginal zones. In the
tropics, growth is rapid and it will take only a short time
for the emergent vegetation to intersect the water line.
When these conditions are produced, particularly in back-
waters, A. funestus larvae will appear. In view of the
close relationship between A. funestus and aquatic vegetation,
it would be highly desirable-for a botanical survey to be
conducted in the Volta and its tributaries so that a
foundation can be provided on which measures can be planned
to prevent or control the establishment of undesirable
species of plants.
121. A. gambiae is known to breed prolifically in borrow
pits. Whnever it is necessary to employ these pits, they
should be dug well above the 260 foot contour and in such a
way that they can easily be drained. If the pits are so
made, much useless draining and filling will be avoided in
122. As the water level drops during the period of
recession, considerable floatage will be stranded on the
reservoir shore. Such material should be removed at low
water to prevent its refloating as the water rises in the
reservoir. The floatage has a large intersection value and
provides an excellent habitat for breeding of A. funestus.
Floating plants such as water lettuce may often be removed
successfully by stranding.
125. Below the dam, the water level in the river will
be held approximately at the present July level, which
during the time I observed it, was sufficiently low so that
marginal marshes and swamps were practically non-existent,
except in the Ada area. All such marginal danger spots
should be properly drained and the section of the river
below Ajena should receive attention equivalent to that
given the reservoir area. This is essential not only for
the protection of the Africans but also the greatly increased
number of Europeans that will be travelling on the river.
124. I was asked to look over the Ada region, at the
mouth of the Volta river, where construction of a
harbour has been proposed. Early in July I went to Ada
and spent two days examining the area. Later in the
month I again visited Ada.
125. The town is located at the coast on a sand spit
bordered on the south by the Gulf of Guinea, on the east
by the Volta river, and on the north and west by an
extensive mangrove swamp and Paspalum marsh. I noted
that the mangrove trees, which are characteristically
brackish water plants, were growing as far north as the
Angaw river. The road to Ada traverses the salt marshes,
which do not appear to be mich influenced by tidal effects.
Undoubtedly, there are some fresh water pools in these
marshes which surround Ada on the north and west.
126. I have indicated in my discussion of malaria vectors
that A. gambiae melas is closely associated with brackish
water,-and- par f-icuTarly with those marshes where Paspalum
grows in profusion. Conditions, then, on the north andT
west sides of Ada are ideal for the production of vast
hordes of this species. My collections of adults in the
town of Ada proved that the marshes are, in fact, prolific.
127. Close to the ocean side of the sand spit, the wind
is strong enough to discourage the movement of large numbers
of the mosquitoes into the area; however, this is an un-
dependable factor in mosquito control, and, though it is
.important in locating the quarters of personnel, it should
be given little weight from the standpoint of malaria
128. If Ada is selected as the harbour site, extensive
mosquito control measures will be required to protect the
permanent personnel, as well as the crews of ships which
dock in the harbour. The location in a veritable marsh
will mean that control will be costly, as considerable
drainage and larviciding will be required.
129. At Major. Hartshorne's request, I have also looked
over the Teshi and Tema areas, in both of which I have
previously worked during the war years. These two towns
are located at the southern extremity of the Accra plains
in a rather dry section of the coastal belt. Near each of
the towns there is a small lagoon, which presumably will
play a part in harbour development, provided one or the
other of these towns is selected.
130. At the upper ends of the Tema and the Teshi lagoons
there is some Paspalu.m and some mangrove (Avicinna sp.).
Both of the lagoons are far smaller than the marshes in the
Ada region a.nd their mosquito producing potential is much
more limited than that of the Ada marshes. Furthermore,
away from the lagoons themselves, there is little water in
which mosquito breeding takes place, except during and
immediately following the rainy season.
131. The control of mosquitoes in either the Teshi or
Tema lagoons would be relatively easy and far less
expensive than at Ada. There would be few drains
required; only marginal clearing of vegetation would
have to be done around that part of the lagoon rot being
used in the harbour development; and it is ntbirely
probable that after the lagoons, which are now sealed off
from the salt water, are opened and are free to fluctuate
with the ocean tides, no other mosquito control measures
will be required, except during the rainy season when
larviciding or residual spraying in houses must be
132. The recent researches of Drs. M.H.Hughes (1949)
and B.B.Waddy (1949) showing the importance of onchocerciasis
in the Gold Coast have aroused great interest in the
disease, both on the part of professional workers as well
as laymen. Dr.Hughes has repeatedly insisted that the
disease is one of the major problems of the Volta river
basin anid that the presently contemplated impoundment
project must take the. disease into account, particularly
during the construction phase. It is not the immediate
effects of onchocerciasis with which he is impressed, but
that collateral blindness which so frequently develops in
cases of long standing. Hughes' findings in the lower
Volta clearly point to the conclusion that onchocerciasis
must be attacked at an early date and that humanitarian
considerations require that control of the vector should
be undertaken, even if the Volta river development scheme
is not brought to fruition.
133. Hughes has pointed out that there is a close
correlation between onchocerciasis and depopulation in
rural areas. Working in the Navrongo District of the
Northern Territories, he noticed that people in declining
villages were heavily infested with Onchocerca volvulus, and
that many of the adults were blind. In fifteen villages
which he surveyed, he found the percentage of blind
inhabitants ranging from 0.1% to 6.5% of the total. It is
his feeling that most of the blindness is associated with
infections of the worm.' Waddy, working in the adjacent
Tumu Diqtrit found similar conditions and he demonstrated
that m6st of the blind persons there had the microfilaria
of 0. volvulus in their conjunctive. Of the population
decreases, Waddy believes that, contrary to previous
suggestions, infections do not seem to impair, fertility,
nor did he find that the disease affected longevity. He
was also of the opinion that the apparent well being and
energy of the population was uot affected. On the other
hand, he was most impressed.with the high percentage of
blindness in adult males.
134. At one village where Waddy examined 974 persons,
204 were men in the 30 50 years age group and of these,
39, or nearly 20% were blind.. Labour was so short at this
village that the blind men were actually led out to the
fields to work. Waddy does not see a direct causation of
depopulation in the blindness of onchocerciasis, but he
believes that its indirect effects are most important. In
his opinion the disease is at present the greatest single
cause of blindness among men of the Northern .Territories.
135. Puyelo (1949) examined 88,289 Africans in the Mossi
country of the Upper Volta (French territory) for the
presence of onchocerciasis. His survey revealed 8,172 with
filarial cysts and 2,066 were blind due to the disease.
Blindness was so common that it was not unusual to see
caravans of blind persons holding sticks end to end. In
'one village, 7.6% of 1564 persons were blind; in another
region 20% of the population was found to be infected, but
in Puyelo's opinion it is more likely that 60 80% of the
people have the disease.
136. In 1949, Dr. Hughes began a survey of some of the
smaller villages of the lower Volta in the vicinity of
Senchi, and he discovered that onchocerciasis is widespread
in the population. His studies in this area are being
continued in an effort to obtain a clear picture of the
disease and its manifestations. At Agbotia, a small
village on the east bank of the Volta near Senchi ferry,
Hughes, in a survey of 62 people., found an overall infection
rate of 82%. At Atimpoko 163 persons were examined and an
infection rate of 68.1% was obtained. In the two villages
the men had the heaviest percentage of infections -- 94% in
Agbotia and 92.5% in Atimpoko. Hughes explains the higher
incidence in men as probably being due to the fact that, as
boys, they start fishing with cast nets at an early age and
to do so remove their clothing; the same applies to adult
fishermen. In so disrobing, they were more exposed to the
bites of the insect vector. Girls lead a more sheltered
life and as they grow older wear clothing which partially
protects them from the bites of the flies. Whether this
supposition is a satisfactory explanation or not remains to
be proved. Certainly 'most young girls, prior to puberty,
wear no more clothing than the boys, and as they come to
the river shore to bathe, wash clothes, or play they are
just as susceptible to bites by Simulium as the boys.
137. Wanson, Courtois and Lebied (1949) report that
onchocerciasis is very widespread at Leopoldville, and
that in the past few years the disease has grown in
magnitude both in the native and European populations. They
estimate that 100% of the riparian natives, who dwell near
the rapids, show microfilaria on examination, and 5% suffer
from ocular complications, including blindness. They point
out, however, that blindness is typically found only among
the old men. Surveys of the European population revealed
that 45% were infected.
138. The disease, caused by the filarial worm, Onchocerca
volvulus, is widely distributed throughout much of Central
Africa, as well as in parts of Central America and Southern
Mexico. So far as is known, man is the only host, though
it has been suggested that other animals might possibly
harbour the worm. In West Africa the only proven vector
is the black fly or buffalo gnat, Simulium damnosum.
Microfilariae, which are present in the skin, are ingested
by the fly when it bites an infected person. Within the body
of the fly, the worm undergoes a metamorphosis requiring
6 to 10 days or more. Before transmission to a new host,
the infective larvae are coiled in the base of the labrum
of the infected fly. Infection of man takes place through
contamination of the bite (Belding, 1942).
139. Because the disease is not a debilitating one,
it has received little attention in the past. With the
recognition of associated blindness, far greater effort
is being made to control or eradicate it. Unfortunately,
results of these efforts are not immediately apparent as
the-infections are long lived. Results will only manifest
themselves in the generation produced after eradication of
the insect vector.
140. These surveys of the incidence of onchocerciasis,
with percentage of infection often calculated on the basis
of. a single examination, adequately point to the fact that
the disease is almost universal along West African rivers
where there are rapids of swift water present. If repeat
surveys were performed, the incidence might easily
141. I have indicated that there is' a close association
of the incidence of the disease and a riverside habitation
in the vicinity of rapids. That such is not invariably the
case is well known, but so far as I have been able to
ascertain, all infected individuals, wherever they may be
when the disease is detected, give a history of having been
at or near a swift river at some time in the past. Dr.
Hughes (1949) discusses the prevalence of blind men in the
region of Ada, where the Volta is tidal and where the vector
of onchocerciasis does not occur. He explains the high
incidence of this disability by the fact that the major
occupation of the inhabitants of Ada is fishing. In
pursuing this activity, the men migrate far up the Volta,
some to the vicinity of the Kpong and Senchi rapids and
others even farther upstream. It is here that infections
are acquired. later these men return to Ada and as they
become older, the ocular complications appear.
The Insect Vectors.
142. In 1926 D.B.Blacklock observed the larval
metamorphosis of 0. volvulus in the black fly, Simulium
damnosum. Subsequently, this has been confirmed by other
workers and the larval stages of the worm have been described
in some detail. S. neavi, a second proven vector of
onchocerciasis, is-iown to occur only in east-central and
143. The Simuliidae are widely distributed throughout
the world and are recognized as one of the major insect
pests associated with streams and rivers. For a short
period during the summer, Simulium is such a nuisance in
Canadian forests as to be virtually unbearable. In both
Europe and America the insects have been known to kill
cattle, horses and other domestic animals. The bite of
the fly is painless at first, except for a slight prickling
sensation. Later on ulcer-like sores may be produced due
to the salivary toxin. In susceptible individuals there
may. be marked inflammation and local swellings (Belding,
144. Nearly all species, of the Simuliidae require
running water for their development, except S. adersi, the
larvae of which has been found on the edge o--"a lake where
wave action was sufficient to thoroughly aerate the water.
The pupae are found in the same location as the larvae,
even if the water is torrential, but they are protected by
a shoe-shaped cocoon.
Similium damnosum (Theobald)
145. S. damnosum is the only prove vector of 0. volvulus
that Qccurs in the Gold Coast, but even in this small
country the distribution of the fly has never been studied
in detail. It is known to occur in only one stream out-
side the drainage of the Volta river. Hughes (1949)
recorded the species from a tributary of the Densu river
that flows under the Adawso-Koforidua road at the 47 mile
146. The chief factor which appears to influence thQ
distribution of the fly is the rate of flow in streams.
Wanson ew Henrad (1945) found that in the Congo a current
of 3.5 4.5 kilometers per hour was the most suitable rate
for the development of larvae. No larvae were found when
the water flowed faster than 5 kilometers per hour and the
minimum rate was found to be 3 kilometers per hour. Their
findings, taken at face value, seem to be an excellent
example of Liebig'-s law of the Minimum -- that essential
factor .which exists in an environment at a minimum level is
the factor which limits the distribution of a species. The
same holds for maximal quantities. I have examined many
streams in the Volta system where conditions, other than
current, would have been suitable for S. damnosum! yet the
absence of this one element appeared to be sufficient to
prevent the species from inhabiting the streams.
147. Whether Wanson's and Henrad's conclusions are valid
is now open to question. Since their results were published,
Lewis (1948) has found larvae of S. damnosum breeding on
hard mud in comparatively placid s-Tretches of the Nile in
the Anglo-Egyptian Sudan. His larvae were taken from water
flowing at a rate of about 2 kilometers per hour. It is
very possible that in West Africa, this mud-dwelling habit
is developed by the larvae but has not yet been discovered
by entomologists investigating the insects.
148. Very little specific information is available on the
biology of S. damnosum in West Africa. Wanson and his
colleagues working in the Belgian Congo, have, however,
contributed greatly toward a knowledge of the species in
that region. Taxonomically, their species appears to be
identical with damnosum of the Gold Coast, and it is likely
that no essential biological differences occur, even though
the areas are widely separated. Briefly, they have found
that the eggs, which are glued by the ovipositing female to
some underwater objects in swiftly flowing water hatch in
a few hours. The newly hatched larvae attach themselves
to aquatic plants, detached leaves, sticks, or rocks which
are firmly anchored in rapid water. The larval stage lasts
five days, when pupation occurs in the same type of habitat
in which the larva develops. The pupal stage takes four
days, about nine days being req-uired in all for the complete
egg adult cycle. The duration of the larval stages in the
Congo appear to be constant regardless of the season as
there is little variation in the temperature of the water of
the rapids, where it ranges from 23.50 2400. Eggs are
not matured in the female until after a blood meal is taken,
but if the meal is adequate and the external temperature
high enough, the eggs will develop in 3 or 4 days. About
200-250 ova are laid at a time.
149. Probably the controlling factor in the speed of
the current is the amount of dissolved oxygen in the
water. Where the stream is turbulent, the water becomes
thoroughly mixed with air and saturated with oxygen. The
fact that Simulium larvae in the West African streams have
Been found only in such places bears out this assumption.
Food is obtained by the larva from the flowing.water by
straining out the plankton with its two, long oral fans
which it holds outstretched against the current. Lack of
food does not seem to be the limiting factor for damnosum
in slower streams, as other species of Simulium may be
found in them in abundance.
150. After completing its pupal development the insect
emerges as an adult during the early hours after daybreak.
Soon after emergence, there is a concentration of the
new adults on the vegetation in the vicinity of the rapids.
In the Congo, the males concentrate around flowers of the
leguminous tree, Baphia bossenge, on which they feed, or
on the under surfaces of the leaves to rest. The females
search for a host from which to obtain a blood meal.
Mating probably takes place after the maturing of the ova.
Longevity of the adults has not yet been determined.
151. Adults of S. damnosum can be found at the banks of
perennial rivers Throughout the year. Hughes (1949) has
summarized the available information about seasonal
distribution in general terms. He points out that in
some years they are more plentiful than others. In the
smaller, seasonal rivers the greatest number occur at the
height of the rainy season, when the streams are at their
peak levels. In the large rivers such as the Congo and
Nile, the insects are most abundant at low water.
152. The maximum distance which the female S. damnosum
can fly has never been demonstrated, but circumstantial
evidence has led some observers to credit the species with
remarkable powers of flight. Gibbins in 1936 reported
that swarms of adults were found at Kampala, Uganda, and
that they had flown from the Nile, a distance of 72
kilometers. He considered that streams near Kampala
were not suitable for breeding of the species. Wanson
et al (1949) estimated that the insect was migrating 40 45
kilometers, and in 1945 captured adult 15 20 kilometers
away from breeding places. Lewis (1948) reports that at
Hamdab, Anglo-Egyptian Sudan, several kilometers from a
large breeding area, biting adults were common in the
153. The effectiveness of the ability of Simulium
to transmit onchocerciasis has been proved in Central
America as well as in Africa. The infected flies form a
rather large percentage of the populations, and even
those which show infective larvae are common. Considering
the great number of bites sustained by people in a Simulium
area, an infectivity rate of 1% is far more than sufficient
to insure infection with the disease. Wanson et al (1949)
reports that 2.9% of the female S. damnosum taken in
villages on the banks of the rapiTs of the Congo, on
dissection, were found to have infectious larvae in the head.
13.35% of the adults dissected had incipient infections in
the muscles of the thorax. It is hardly surprising, then,
that every human living in such villages shows infections
of the worm as they are bitten by the flies countless times
each year resulting in repeated re-infections.
Collections and Observations
in thej-olta Drainage.
154. Map 1 shows those parts of the Volta where
collections of insects were made. Almost invariably
wherever rapids were present in the main river, larvae
and pupae of S. damnosum were found in large numbers.
In only a singTe tributary of the Volta, the Daka river,
did I find S. damnosum breeding commonly, but here'they
were more numerous .tEhn in the Volta itself.
155. The village of Kpong stretches for nearly a mile
along the west bank of the Volta at the edge of the
largest rapids in the river. The rapids were studied
over a period of nearly two weeks, beginning 8th July
and ending 20th July, with a view to determining the
presence of S. damnosum, its prevalence and distribution,
and the poss81le effects of impoundment of the river on
the stabilisation of the population of the insects.
156. Just above the rapids, the river is relatively deep
and slow flowing, but as the outcrops of rock and the
numerous islands in the region of Kpong make their appear-
ance, the depth lessens and the river spreads out in a
broad, flat sheet. Over much of the area, the water is
rather shallow, being not more than 2 6 feet deep. In .
the shallower, rocky channels of the rapids there are
smaller, loose stones strewn over the stream bed.
157. In the shallower parts of the river, and in particular
those places which had only recently become covered by the
rising water, grasses were rooted and the blades streamed
out in the direction of flow just below the water surface.
Also, along the banks of the river and the shores of the
islands, where vegetation was plentiful, an occasional branch
dipped into the water and it, too, was covered or partially
covered by the flowing water.
158. A native canoe was utilised in examining the rapids.
The canoe navigated into the swiftest water and where out-
crops occurred in mid-stream, we stopped, pulled the canoe
onto the rocks and investigated the area for stream insects.
The canoemen went into the water for rocks, bringing up
large as well as small ones from depths ranging from 1 5
feet. The rocks in midstream, which were in water 3 to 4
feet deep and which were large enough not to be rolled
about by the current, had their upper surfaces covered with
a closely growing network of moss (probably similar to the
species of Podostemacea which Wanson found in the Congo).
Sufficient sunlight was able to penetrate the turbid water
to permit the plants to grow reasonably well.
159. Some of the larger rocks were inhabited by Simulium
damnosum and S. griseicolle. On one rock, measuring about
18" 1" x 4", more than O0 pupae (mostly griseicolle)
were attached. In spite of the large number of blackflies
on that rock., further examination of many others produced
relatively few of the insects.
160. Where grasses streamed out in the swift current,
S. damnosum larvae and pupae could frequently be found,
provided that submergence had not been too recent. Dead
branches with leaves still attached sometimes became
wedged in the rocks. These offered an excellent habitat
for Simulium and many of them were found in such places.
161. It appears that S. damn6sum larvae and pupae can
live both on the vegetation and on the rocky bed of the-
river in shallow, rapid water. The insects attach them-
selves on the leaves, and sometimes on the stems, of
plants trailing near the surface of the moving water. On
the rocks, the insects are common near the surface, but do
not thrive well below a depth of a foot or so. The pupae
found on rocks in deeper water probably pupated on the
rocks before the rise in water level.
162. Countless numbeniof Simulium are able to breed in
the tremendous area of shallow water in the Kpong rapids.
I have been told by the local natives that as the river
rises, so the number of "ma-ma-taw" rises, and at flood
level the insects are at their peak abundance.
163. During my work at Kpong, collection of adult
S. damnosum were made as they alighted on the legs 'of mall
children, who tended to surround and crowd about me out of
curiosity. The curiosity and interest were encouraged
as it made the catching of the flies relatively easy.
I wanted to keep the insects alive until they could be
dissected to determine percentage of infection with
0. volvulus, and so I usually chose five boys and gave
TeTm small bottles in each of which a single fly could be
caught. Flies could be caught from early morning ..to
sunset, but they were most active during the hottest hours
of the day.
164,. In all, during my stay in the Kpong area 92 adult
female S. damnosum were dissected.
The results are summarized in the table below:-
INFECTED NOT INFECTED
Blood Number Non- Blood
Date Head Thorax Abdomen meal infected infected meal
July 9 0 2 0 1 2 8 ?
102 1 2 2 ? 2 3 ?
10 4 7 5 ?7 5?
13 0 1 1 2 2 1 0
14 0 34 1 1 4 12 0
18 0 0 0 0 0 13 3
192 0 65 0 2 6 22 6
19 0 0 0 5 I
5 21 9 ? 23 69 ?
Note: 1*Not determined.
2'All specimens collected from shaded area.
3A11 specimens collected from individuals standing in sun.
'Two with "sausage" forms in thorax.
5'Five with "sausage" forms in thorax.
The percentage of infected individuals was 25% of the total.
The percentage of infective individuals (those with larvae
in the head) was 5.4% of the total.
165. The results of the 18th July dissections are a
little difficult to explain. The flies were captured from
the legs of small children seated on a large mass of boulders
about 150 feet from shore. These insects may have been
newly emerged adults which were feeding for the first time.
Certainly the size and condition of the ovaries was not that
of more mature adults.
166. While dissecting the specimens of S. dammeosum from
Kpong at the Akuse rest house, I noted an a-ult of the
species attempting to bite my arm. The specimen was
captured and during the afternoon several other individuals
were taken as they attempted to bite me or my assistant.
Ten of these were dissected and none was found to be infected
with 0. volvulus.
167. After finding the adults of the fly biting so
viciously indoors at a distance of about two miles from
the river, I began to make collections at the rest house
using available Africans as bait. In ten minutes more
than fifty adults of S. damnosum were collected from the
legs of two men. Dur'Eg the time I was making my collections,
not a single fly attempted to bite me, although they were
attacking the two Africans with vigor. Just as at Kpong,
the flies attacked the legs below the knees almost without
exception. These specimens were collected by means of an
168. The lack of infection of the flies found in the
Akuse area, in contrast to those from Kpong, may be related
to the fact that there are relatively few Africans in the
vicinity of the rest house which makes it likely that the
insects had no chance to feed on infected humans. The
normal source of their food was probably feral animals.
169. Unfortunately, the time consumed by dissecting
large numbers of Simulium is excessive and not more than
25 30 could be done in one afternoon. It was necessary
to collect the insects in the morning and to dissect them
the same day, for if they were kept for more than a few
hours, the mortality rate was high and th.e flies tended to
dry out rather rapidly.
170. At Senchi ferry the Volta is rather deep and slow
flowing. By 31st July, the rise in the level of the river
resulted in the submergence of marginal vegetation, but the
slow movement of the current hardly disturbed the plants in
the water. A careful examination of the plants revealed no
Simulium larvae. Other insects that normally inhabit quiet
backwaters or pools and ponds were present, although not
in abundance. The banks near the ferry and up and down
the river in this region are very steep and the shore zone
is narrow with a rapid drop off to deep water. This is
clearly indicated by the narrow line of rooted aquatic plants.
171. While I was looking for larvae at the bank of the
river, I noted an adult blackfly biting the leg of a nearby
workman. The specimen was captured.and several others
were observed, but they did not seem to be particulaBly
numerous in this region. This observation was reinforced
by the fact that, although I worked on the porch of the
rest house adjacent to the river all day on 2nd August,
neither my African assistant nor I received a single bite
by the flies. Working under similar circumstances at
Akuse we were frequently attacked.
172. At the lower end of the Senohi rapids the river
had risen by 1st August so that many of the rocks were
covered. Those on to which I had walked with ease in late
June were now submerged. The water flowed rapidly and
the marginal grasses streamed out in the direction of the
current. The flow here was quite sufficient for the
larvae of S. damnosum and that they were taking advantage
of the habiTat was abundantly demonstrated by the prevalence
of adults on the banks of the river.
173. On the morning of 3rd August, I returned to Senchi
rapids and with the aid of my assistant and five African
children as bait, collected 40 female-S. damnosum adults in
about 15 minutes. The flies were very numerous, p or 4
settling on the legs of a child at one time. As with
other specimens previously caught, the insects almost
invariably bit the legs; however, an occasional specimen
attacked the trunk when the child was sitting. The site
of the bite is apparently correlated with its height above
the ground. No Simulium damnosum has been observed biting
more than 3 feet above the ground and mostly they confine
their attention to those parts of the body less than 18
inches from the ground. This is in conformity with Hughes'
(1949) observation that the flies invariably attacked the
legs and that when they had an opportunity to bite a black
or a white leg, they usually chose the black.
174. The specimens collected in the morning were taken
back to the Senchi rest house and dissected to determine
infections with 0. volvulus. The following results were
INFECTED NOT INFECTED
Blood Number Non-in- Blood
Date Head Thorax Abdomen Meal infected fected meal
Aug. 3. 0 86 1 1 8 27 4
Note: 6One specimen with both "sausage" and filariaforn
175. Of the total 35 dissections, 22.9% of the females
were infected; none was found to be infective. These
results are consistent with those obtained in the Kpong
dissections, where 25% of the adults were found to be
infected. The smaller sample at Senchi accounts for the
176. At the time of my first visit to the Ajena rapids in
June, the level of the river had already risen sufficiently
to partially submerge them. At-this time, S. damnosum
adults were rather rare, although occasionalTbites were felt
by workers at the projected dam site. In July and
August the numbers of flies were still small, but by
September, when the river was flowing at a rapid rate, the
production of black flies jumped sharply so that they
became a real pest in the area. When I revisited Ajena
on 30th October, the river level had dropped markedly and
so had the numbers of blackflies.
177. At Dodi, a small village located on the east bank
of the Volta opposite the Afram river, the fringing forest
was rather heavy. By 9th August the rapid rise of the
river had covered the lower part of some of the trees, and
grass that had been growing on the banks was now well out
in the stream. The river was wide, but apparently shallow,
as was evident by the speed of the current that ran swiftly
around the tree trunks and over the submerged grass. The
northern part of a large island at the mouth of the Afram
could be plainly seen from the shore, as well as rapids
below Dodi. Conditions here were ideal for prolific
breeding of Simulium damnosum.
178. At Dukluaja, the river is deep and rather -slow
flowing, reproducing conditions that are somewhat reminiscent
of those at Senchi; however, the banks of the river are
heavily forested and were partially flooded during my visit
on 16th August. I travelled to a small island of sand in
midstream that was beginning to be covered by the rising
river and, on submerged grass growing in shallow water by
the island, I found Simulium larvae, but they were not
numerous. I could find no other breeding in this section
of the river.
179. Farther downstream at Atikpeta, conditions were
different. Travelling by canoe, I went up the east bank
of the river where the water was flowing swiftly. The
marginal vegetation was well covered with the rising water
and Simulium was becoming established on the plants. About
a half mile upstream, we paddled out into the river, across
the swift current, to a large mass of rocks projecting out
of the water. Here the water flowed swiftly, but not as
rapidly as at Kpong. A few Simulium larvae were collected
bht they were not numerous. When I inquired of the local
inhabitants about the density of the blackflies, I received
the same answer as I had got at Kpong and Senchi that
they become abundant in late September and October.
180. In late August large numbers of S. damnosum larvae
and pupae were found in the turbulent rapids of Kete Krachi
where they are living in the swiftest parts of the river that
I was able to examine. Large clumps of a plant resembling
Myriophyllum were found growing in the swift water and
bS damnosum larvae and pupae were attached to it. The great
aount .of surface provided by a plant of this type is un-
doubtedly of considerable importance in furnishing attachment
for the masses of S. damnosum larvae. The largest numbers
of larvae were fou--, however, on leaves of grasses and
floating vegetation in the swift flowing stretches of the
river. The tips of the grass blades were so thickly covered
with larvae and pupae that the plants were invisible under the
mass of small insects.
181. I travelled downstream from the rapids by canoe for
about a mile. In the shallows, there were many bushes and
plants trailing in the water, but few Simulium immatures
182. In spite of the great abundance of larvae and pupae
of S. _damnosum in the Kete Krachi rapids, no adults were
seen. On the afternoon of 23rd August, with the aid of
three native boys to serve as bait, I attempted to catch
adult flies, but in a period of two hours not a fly came.
The attempt was repeated on the morning of 24th August
with the same results. I have no explanation for the lack
if activity of the flies.
183. I worked in the vicinity of Yeji ferry for nearly
a week. Daily visits were paid to the river in early
October at which time it was at its peak level. Not a
single A. damnosum was found biting. It was not until I
had traFelled two miles upstream by canoe, and waited for
two hours, that my canoemen, serving as bait, were finally
bitten by two S. damnosum. The natives in the vicinity of
Yeji claim tha--the flies only become numerous as the level
of the water falls. This seems reasonable, for in the
region of Yeji, breeding of the insects is probably con-
fined to vegetation in swift water. At flood stage, the
habitats are deeply submerged and only small pockets of
the flies manage to persist until conditions become more
favourable for the insects.
184. The greatest concentration of S. damnosum larvae
and pupae that I found during my entire survey was in the
Daka river at Grube. At a small rock barrier built across
the stream, the water was spilling swiftly over the rocks.
Grasses were growing profusely in the rock crevices in the
swiftest water. Strands of the plant were literally paved
with larvae and pupae and not a spot was left vacant. The
conditions here reminded me of the surfaces of some of the
rocks that I had seen in mountain streams in Mexico, where
the thickly packed Simulium larvae gave the rock the
appearance of being covered with velvet.
185. For the first time during my survey, I discovered
the ova of S. damnosum at the Daka river. They were found
attached in great numbers to the lower side of the leaves
in swift water, while the other side was covered with larvae
in all stages of development. Many of the eggs were in
the process of splitting to release the young larvae, others
were empty, and others contained developing young. They had
been laid in long strands and glued to the leaves.
186. The only other tributary of the Volta in which
S. damnosum was found was the Oti river at Atafie, but the
Insect was scarce there. In spite of a long and careful
examination of tributaries of all sizes throughout the lower
Volta basin, nolarvae or pupae could be seen. Their
absence is most probably associated with the low rainfall
during 1950, for all streams were flowing far below the
normal levels during the rainy season. It is likely that
the rate of flow in the tributaries never reached a speed
sufficiently high to satisfy the minimum requirements for
colonization by S. damnosum.
Other Species of Simulium found in the Volta Drainage.
187. During the course of- my search for S. damnosum, other
species-of the-genus were also encountered. None of these
has been incriminated in the transmission of onchocerciasis
and most of them are not known to bite man. However, to
round out the picture of Simulium breeding in the Volta
river system, I will give a brief resume of my findings.
Simulium griseicolle Becker.
188. This is a second species of the Volta river and the
only other one besides S. damnosum of which I found larvae
and pupae in the main sr-eam. In every place that damnosum
was breeding, griseicolle was also present, but usually Mot
in as great abundance. The converse was not true, however,
because griseicolle appears to be able to tolerate slower
flowing water than damnosum. This is confirmed by Lewis's
(1948) findings of griseicolle breeding on moderately hard
mud in the Nile where the water was flowing at a rate of
only 0.8 kilometers per hour, which is far less than the
3.0 kilometers Wanson considers as the minimum speed
necessary for damnosum in Central Africa.
189. Whether griseicolle bites humans in West Africa is
as yet unknown. Hughes (1949) never found the species
biting or attempting to bite in the vicinity of Senchi, where
it is plentiful. No specimens of griseicolle appeared in
my adult collections from either the Kpong or Senchi areas.
Lewis (1948) reports that the females walk over the skin and
probe, causing some annoyance, but that only a very small
number ever engorge on human blood. Wanson and Fain (1947)
also report that griseicolle sometimes feed on humans.
Lewis (p. 482) points out that in the Anglo-Egyptian Sudan
"Unlike S. damnosum, S. griseicolle mainly attacks the
upper part of the body, particularly behind the ears,
engorges slowly, is not easily disturbed, and does not
leave a drop of blood at the site of the puncture".
190. While I was trying to collect S. damnosum above Yeji
ferry, three adults of another species were found on the
shoulders and chest of one of my African canoemen. These
specimens have not yet been identified; therefore, I am
simply labelling them Simulium species. The specimens were
somewhat damaged making identification particularly
Simulium alcocki Pomeroy.
191. S. alcocki was found in the smaller tributaries
of the V'olT~a.ETe individuals never occurred in large
numbers and it was sometimes necessary to hunt for an hour
before finding a larva or pupa. In streams with pebbly
or rocky bottoms, larvae and pupae could. be found attached
to stones and vegetation in the moderately flowing water
as well as to leaves caught in drifts in the flowing part
of the stream. The species is not known to feed on man.
Simulium alcocki violaceum Pomeroy.
192. Larvae were reasonably common but pupae were scarce
in the streams-where this species was found. All specimens
were taken from plants in moderately flowing water.
Simulium adersi Pomeroy.
193. Larvae and pupae were present in small numbers in
two streams, but in the Pru river there was a great
abundance of the immatures on vegetation in water with a
Sinulium unicornutum Pomeroy.
194. Immature stages were found in six small streams,
where they were living on vegetation trailing in the current.
Specimens were also taken from dead leaves caught in
drifts in the stream.
Simulium cervicornutun Pomeroy.
195. Two specimens of this species were taken from leaves
trailing in the water in a small stream.
Simuliun schoutedeni Wanson.
196. A single pupa and a few larvae of this species were
taken from the vegetation in the Pawmpawn river, a small,
shallow stream draining the north side of the Akwapin range.
197. Specific locality records for the various species of
Simulium collected during the survey are as follows (Map 1).
Volta river at Kpong, July 8 31. (S. dannosun and
Volta river at Senchi, June 22, August 1-4. (S.damnosun
and S. griseicolle).
five miles north of Akwamu West, August 3
at Ajena dam site, June 16, August 4,
September 26-28; October 1-30 (S. damnosum)
at Dodi, August 9 (S. damnosum and
at Dukluanja, August 16 (S. damnosum) and
at Atikpeta, August 16 (S. damnosum and
at Kete Zrachi, August 21-2-4. (S.damnosum
and S. griseicolle).
two miles north of Yeji ferry, October 10
(S. damnosum and S. griseicolle).
Daka river at Grube, August 22 (S. damnosum and
Oti river at Atafie, August 22 (S. damnosum).
Tahala (small stream north of Kpandu) August 8, 10
Koloe (stream crossing Xpandu-Hoboe road), August 18
(S. alcocki violaceum, S. adersi).
Deben (small stream south of Buafori on Kete Krachi -
Chiadiri road), August 22 (S. unicomrtum,
S. cervicornutum, S. alcockirviolaceum).
Pawmpawm (stream crossing road near Huhunya),
September 5 (S. schboutedeni).
Akrum (stream crossing road just south of Begoro),
September 6 (S. unicornutum,
S. alcocki violaceum).
Animi (stream near Bepong), September 12 (S.alcocki,
S. alcocki violaceum).
Small stream just north of Mpraeso, September 12
(S. alcocki, S. unicornutum).
Asuboni at Mpraeso, September 12 (S. alcocki,
Afram river at Mankrong, September 13, 14 (S.alcocki
violaceum, S. unicornutum, S. adersiT.
Ongwam (stream between Kumawu and Kwamang), September 21,
(S. alcocki violaceum.).
.Ghirimea (small stream just north of Mampong),
September 22 (S. unicornutun).
Dida (small stream near Amoamang), September 22
Pru river at Prang, October 11 (S. adersi).
GENERAL DISCUSSION AND RECOMMENDATIONS.
198. What will be the effects of impoundment on Simulium
damnosum, a species which, during its larval stages, normally
requires swift flowing water for its development? After
completion of the dam, all appreciable flow will cease, and
all rapids will be deeply submerged in the reservoir area.
Under these -circumstances, I believe that damnosum will
disappear from the lower Volta from the dam as far upstream
as that section of the river which will not be slowed by
impoundment. With the disappearance of Simulium from the
reservoir area, transmission of 0. volvulus will cease,
as one of the essential links in~Ehe life cycle will be
199. The significance of the disappearance- of the fly
above the damn is dimmed somewhat by the fact that the
density of human populations on this part of the river is
very low. The real problem areas, the Kpong and Senchi
rapids, will--remain as problems since they are located
below the dam site. If anything, the breeding of Simulium
will actually be favoured on completion of the dam,f-r--51e
height of the river will be stabilized at such a level that
the rapids will be covered to the extent most suitable for
development of damnosum. Dr. P.A.Buxton of the London
School of Tropical Medicine, in a letter to Dr. Pridie,
Chief Medical Officer, Colonial Office, dated 17th October,
1950, pointed out that Mr. D.J.Lewis found S. damnosum.
larvae and pupae occurring in great numbers_"n a channel
running long the foot of the Sennar dam in the Anglo-
Egyptian Sudan. The channel is designed to carry away
the small amount of water that leaks through when the
sluices are closed. Similar situations will probably be
created at the Ajena dam.
200. The high population densities, the constant high
level of fly breeding, and the almost universal infections
with onchocerciasis in the Senchi-Kpong-Akuse regions
require that steps be taken to deal with the problem. The
construction of the dam and its auxiliary units reinforce
the necessity for early action against the insect vectors.
The importation of a large non-African working force will
require that they be protected, not so much because
of the mild, immediate effects of the disease, but
against the possibility of the very serious, later mani-
festations. Many of the Africans will be immigrating from
areas free of onchocerciasis; it will not take long for
them to become infected under present conditions.
201. Wanson, Curtois, and Lebied working at Leopoldville
demonstrated in 1949 that S. damnosum could not only be
controlled, but eradicated, in parts of the Congo river,.
Garnham and McMahon (1946) eradicated S. neavi, an East
African vector, in an onchocerciasis aFrea; however, I do
not believe that their methods would be applicable in a
stream as large as the Volta. On the other hand, Wanson's
technique appears to be the real answer to the Volta problem.
202. As early as 1942, the Belgian workers discovered
that larvicides will destroy larvae for short distances,
but application of insecticides in rapids was found not to
be practical because of the speed and volume of the water.
Enormous quantities of insecticide would be required to
maintain a sufficient concentration for the time necessary
to kill the larvae. Consequently, they decided to attack
the adult stage. Preliminary tests indicated that D.D.T.
was an ideal insecticide to use. After experimenting with
various methods of applying the material, it was found that
spreading it in the form of an aerosol by aeroplane was
the most effective by far. A very careful preliminary
study of the area was made, and the density of the adult
Simulium population was determined. The spraying was then
begun on the theoretical assumption that eight fumigations
in two weeks would destroy all adults which were "on the
wing" or which might emerge from the river. Since no
additional eggs would be laid, the species would be
203. Actually, it was not so simple. More frequent flights
were necessary at first. Later, it was discovered that the
area in which eradication was desired was being reinfected
from a reservoir population that had been overlooked. On 10th
October, 1948, ,using-a single airplane, flights were made every
other day over the area and the work was continued for 26
days. Their results were sensational. Within the first two
days Leopoldville was practically freed of the insects. On
the' fourth day small native villages within the 45 kilometer
perimeter were completely cleared. By the end of three weeks
of spraying, larval habitats were deserted, indicating a
cessation of oviposition. One year later, Leopoldville was
still free of Simulium. There still remained the danger of
reinfestation of the region from rapids where the. flies were
uncontrolled, By establishing barrier zones beyond the great-
est possible flight range of the adult females, they have
insured against establishment of the fly in the rapids at
204. It appears that control of S. damnosum in the Volta may
be even more easily accomplished tIn "it was in the Congo. As
there is no other basis for estimating the maximum flight
range of S. damnosum, it would be well to follow the procedure
of the Belgian workers who accepted Gibbins claim of having
found S. damnosum at a distance of 72 kilometers from its
breediE- grounds. They then proceeded to extend the barrier
zone to 100 kilometers. In the Volta, I do not believe that
there is any breeding area below the Kpong rapids, therefore,
the southern barrier zone is a natural one.
205. On the other hand, as protection of the workers in the
Ajena area must be undertaken prior to construction, a barrier
zone will have to be established with the Obosum river as its
upper limit. There are numerous minor rapids in the section of
the river from the Obosum to Ajena where breeding of the fly
takes place on a small scale. These can all be dealt with by
aerial spraying in the same way in which the extensive Kpong
and Senchi rapids are treated.
206. Before eradication is begun, there should be a very
careful study of the distribution of S. damnosum in the section
of the Volta from the Obosum southward in order to make certain
that no breeding areas will be missed in the spraying. This
work should be undertaken under the supervision of an entomolo-
gist. A carefully prepared map of all breeding areas will be
a most valuable adjunct in making quick checks to determine the
effectiveness of the control project.
207. It would be well for the entomologist chosen for this
project to visit Dr.Wanson at Leopoldville and see his method
of eradication first hand. Much time might then be saved by.
eliminating minor distracting factors that are bound to appear
in such an extensive undertaking.
208. After the initial aerial spraying has been accomplished,
constant checking and re-checking will be required to make sure
that reinfestation of the rapids of the lower river does not
occur. Once impoundment has been achieved and the rapids
above t.he dam are eliminated, the danger of the Kponv and Senchi
rapids being reinvaded by Simulium damnosan is almost non-
* It is anticipated that the bed of- the Volta will be
practically dry for the two or three years required for the
filling of the reservoir. Under these circumstances, the flow
through the Kpong and Senchi rapids will be so much diminished
that eradication of S. damnosum might be effected. Although the
desired results may be- obtained finally, it would be far better
to undertake eradication of the fly prior to inception of
209. It may be possible, after the initial spraying is
ended, to remove any larvae that become re-established in
the barrier zone with a crew of labourers. By uprooting
any vegetation-growing in swift, shallow water, cutting
branches off that trail in the water, and using stiff
brushes to scrub the rocks in the rapids, small populations
might be controlled. Wanson and his co-workers tried this
method prior to aerial spraying and found that the number
of adult flies was appreciably reduced, but eradication
was not possible. Details such as these will have to be
determined by the entomologist as contingencies arise.
210. Hughes has found that in the Senchi area the
density of the fly population is apparently at its lowest
level from January to June, from then onwards there is a
constant rise in the numbers of adults, with a noticeable
decrease becoming evident towards the end of the year.
My observations from June to the end of October bear out
Hughes' impressions. Unfortunately, no quantitative data
are yet at hand, although it is strongly advisable that
this information be collected. It would seem logical to
attack the fly at a time when population densities are
at their minimum level and the river is restricted to its
low water channel, but the final decision is not quite so
simple as that.
211. The White Volta and its tributaries, which during
the dry season stop flowing, are prolific producers of
Simulium once the rains start and the flow is strong and
constant. The same appears to be true of the smaller
streams of the lower Volta in years of normal or high
rainfall. The question of how these flies become
established in streams within a few days or weeks after
-they begin to flow is still unanswered. No one knows how
long it takes the fly to appear after suitable ecological
conditions are produced. It has been suggested that the
flies lay a drought resistant egg which can survive the
long dry season; that some of the females hibernate during
the dry season; and that a resistant pupa is produced. So
far there is no direct evidence for any of these being the
answer. A fourth possibility is rapid migration of the
insects from perennial centres of population to reinvade the
intermittent streams. Of the four possible explanations,
the drought resisting egg seems to be the most plausible,
although migration certainly cannot be discounted. It has
been claimed that species of Simulium in' other parts of the
world survive long dry seasons by laying these drought
resistant eggs in the beds of intermittent streams. Many
species of insects, and other animals. as well, utilize
shells or cases that are resistant to desiccation to tide
them over difficult times that would kill other stages.
The hibernating adult and the resistant pupa do not appear
to be tenable ideas.
212. If drought resistant eggs are present on the rocks
in the rapids and in the tributaries emptying into the
Volta, when they are resubmerged with rise in level of the
water, they will hatch, and control of adults at low water
will have proved to be a useless undertaking. The obvious
conclusion to be drawn, then, is to undertake control in
September as the river is rising. If drought resistant
eggs are produced they would probably not be laid until
the river starts receding.
213. The method by which S damnosum survives the dry
season is one of the most pressing problems of its biology
that must be solved soon, as much of the effectiveness of
control or eradication depends on this information. It is
of utmost importance that studies be undertaken at an early
date to answer this question,
214. Trypanosomiasis, a disease both of humans and
livestock and transmitted by the bite of various species of
Glossina, has long been known to be endemic in the Gold
Coast. Because of the seriousness of the disease, con-
siderable attention has been paid to its control throughout
equatorial Africa. Through the years tremendous strides
have been made in learning the biology of the tsetse and
methods of control.
215. With the recognition of the importance of
trypanosomiasis in the Gold Coast, the Medical Department
has organized two services to deal with detection and.,
control of the disease. One of these, under the direction
of a medical officer, is concerned with epidemiological
surveys and treatment of infected humans. The second has
as its function the study of tsetse flies and their control
and is headed by a medical entomologist.
216. A complete description of the control of trypanoso-
miasis by control of tsetse in the north-western part of
the Volta basin was published by Dr. K.R.S.Morris in 1946
and 1949. In these papers he reviews the past history of
the disease in the Gold Coast, its extent and significance,
and gives a thorough discussion of entomological methods
used in controlling the vectors.
217. During the latter half of 1945 and early 1946,
Dr. T.A.M.Nash, medical entomologist with the tsetse
campaign in Nigeria, at the request of the Colonial Office,
toured all of the British West African colonies on a fact-
finding mission. In his survey, he.spent the period from
21st February to 13th May, 1946, in the Gold Coast. His
findings and comments are given in his comprehensive
report '"Tsetse Flies in British West Africa", published
by the Colonial Office in 1948.
218. As both Dr. Morris' and Dr. Nash's reports
clearly describe conditions in the Gold Coast with regard
to tsetse and tsetse control, I shall do little more than
report the species I collected, their localities, attempt
to evaluate future problems, and make recommendations.
219. In July, the Tsetse Control Department at Lawra
undertook the training of three fly-boys for my use on the
Volta. The boys completed their course in mid-September
and reported to me shortly thereafter. Without delay, I
stationed them in the Ajena area with instructions to
catch not only tsetse but also any other biting insects
which attacked them. In spite of the fact that I was
able to use their services for only a little over a month,
the data which they collected was well worth the time and
trouble in putting them in the area. It is my understanding
that the service of these trained fly-boys will be retained
so that they can continue to collect specimens in the same
region. This will assist in providing a basis for estimating
seasonal fluctuations in populations of biting insects.
COLLECTIONS AND DISTRIBUTION OF TSETSE.
220. During the course_ of my work, there was no place
at which I stopped along the Volta river where tsetse were
not collected. Of those places on the river itself, the
greatest numbers were taken or observed at the Tawtibo,
a small stream 31/ miles south of Ajena, at the point where
it joins the main river., and at Yeji about one mile up-
stream of the ferry landing. All specimens except those
from Yeji, were Glossina palpalis; the-Yeji specimens were
G.. tachinoides. Specific records are given below:
Tawtibo (stream at Ajena
(Volta river at Dukludja
(Volta river at Atikpeta
(Volta river at Kete Krachi
(Volta river near Ajena
(Volta river at Yeji ferry
(Volta river 1 mile upstream
(above Yeji ferry.
July 9-14, 18-24,
221. Tsetse collected by the fly-boys operating along the
banks of the Volta from one mile below Ajena as far up-
stream as Misikrom were all G. palpalis. In 26 days of
collecting, 166 adults were collected of which 89 were
males and 77 were females. The greatest number of flies
taken in one day was 21, the least was nil.
222. While I was working the tributaries and their head-
waters, I was often attacked by tsetse. Of all the places
where they were noted, I found them most troublesome at
Mankrong on the shore of the Afram river. The only species
which was taken at these places was G. palpalis, one of the
two most important carriers of human sleeping sickness in
the country. Specific records are as follows:-
Tahala (small stream near To)
Ebuni (stream valley near Anum)
Dayi river at Huime
Ekple (stream valley near Dafo)
Asuakawkaw river at main road north
Deben (stream valley near Nkatekwan)
Oti river at Atafie
Pawmpawm (stream near Huhunya)
Afram river at Achi September 12
Afram rirer at-Mankrong September 12-14.
Afram river at Nkatepa September 14.
Worobong (stream flowing into Afram) September 13.
Asuboni (stream flowing into Afram) September 15.
Kume (stream between Krota and September 20
Ongwam (stream between Kumawu and September 21.
Afram river (between Aframso and October 5
Roadside ditch near Afrante October 5.
Pru river (between Ejura and Nkoranza)October 6.
223. According to Dr. Nash's report, G. morsitans occurs
in the Volta region north of Kete Krachl7 G. longipalpis,
the fourth species of importance, is found a-long the Volta
from the region of Akuse to Kete Krachi, except in the
Senchi-Anum section. Also according to this author,
Dr. G. Saunders of the Trypanosomiasis Centre at Kintampo
is of the opinion that G. morsitans is of secondary
importance and that G. longipalpis is not associated
with a high incidence-of human sleeping sickness. Nash
considers G, nigrofusca, a fifth Gold Coast species, to be
of little importance in the country.
224. From the standpoint of human sleeping sickness, then,
G. palpalis, and G. tachinoides are of primary importance.
!ash (1948, p.36) says, "Captain Stewart considers that
G. tachinoides is the main carrier of bovine trypanosomiasis.
sincee the cattle owners are settled, and G. morsitans is
absent from the most heavily populated areas where cattle
are most numerous, this species cannot be as dangerous as
G. tachinoides. He considers that G. tachinoides is more
important than G. palpalis because IT is far commoner in
the extreme north, where the bulk of the cattle are found.
"In the coastal belt one would expect G. longipalpis
to be a much more serious vector of bovine trypanosomia-
sis than G. palpalis because of its preference for game,
and resulTant high infection rates."
225. Nash summarises the effects of seasonal changes on
abundance of the various species of tsetse. He points out
that Pomeroy and Morris claim that in the woodland-savannah
country, as found in northern Ashanti, Togoland, and the
northern territories, G. palpalis increases steadily
in numbers. from the beginning of the rains to a peak
density in September. From this time the numbers fall
through the remainder of the rains and early dry season.
Nash concludes that these observations were based on
rivers subject to flooding.
226. At Pong Tamale, where the White Volta floods ex-
tensively in September, the density of G. tachinoides
increases very rapidly at the beginning-of the rains
and reaches the maximum just before the heavy rains.
There follows a sharp drop with the population of flies
very low during the remainder of the rainy season.
Populations then increase rapidly in the dry months.
227. Again to quote Nash (p. 39), "Morris (Rep, Med.
Dept., 1930) attributes the great decrease in G. tachinoides
in the latter half of the rains on the Volta .aT-eji
to the inundation of all the breeding grounds, and
believes that G. tachinoides is more seriously affected
than G. palpalrs because it is less ready to colonize
secondary foci away from the river."
228. For neither G. morsitans nor G_. longipalpis are good
seasonal data avail-ble.
PRESENT METHODS OF CONTROL,
229. Morris has discussed his present methods of dealing
with G, palpalis and G. tachinoides at some length (1946,
1949). His recent results in The Lawra district of the
Gold Coast speak well for the techniques he has developed
in the region, particularly since widespread deforestation
becomes unnecessary. He described (1949) his method of
eradicating the two species of tsetse being based on the
"concept that the tsetse community on each river system
forms a natural biological unit. The communities extend
widely along the water-courses during the rains but
contract to well defined and often very restricted foci
during the adverse climatic conditions of the dry season.
These foci or permanent fly-belts are confined to certain
definite vegetation associations containing a limited
number of species of trees and shrubs whose presence is
essential for the survival of the tsetse during the hot
and arid period from December through March, The removal
of only these essential species of trees throughout a
whole river system is sufficient to ensure the dis-
appearance of G. palpalis and G. tachinoides, since their
dry season fooT-are now untenatle Tor at least four months
230. Though Morris' method seems to be an ideal technique
because many valuable species of trees are spared, cost
is low, and the prime object of eradication of G. palpalis
and GQ tachinoides is accomplished, Nash expresses a good
deal of scepticism as to its value. He criticises
selective clearance because, if, in some places, a valuable
species of tree is present, it may be spared leaving
conditions that will be conducive to the continued presence
of tsetse. He also-points out that neglect by the local
authorities may mean that slashings are not repeated
annually and there is rapid regeneration of plant
associations favourable to tsetse. Nash also feels that
Morris' technique is inelastic, leading in some cases to
unnecessary ruthless clearing and in other cases to making
tsetse eradication impossible. Nash points out that the
tsetse is not affected directly by the species of plant in
a plant association, but by the climate produced mechanically
by the screening effect of vegetation. It is Nash's
opinion that ruthless barrier clearings and partial
clearings are necessary to control tsetse, but he definitely
does not agree with Morris' selective clearing. The
points of disagreement between these two experts clearly
indicate that much-more research is necessary before
techniques will be worked out that will satisfy all
interested parties as to their efficacy. On the other
hand, Nash's remarkable success in clearing the Anchau
corridor in Nigeria, where sleeping sickness was serious,
by waterside clearing speaks well for his system. The
disease was eradicated from 700 square miles so that there
are now some 50,000 healthy people with cattle living there.
They have good water supplies, improved agriculture and
markets, and the beginnings of a much richer life than
in the past (Buxton, 1948).
231. A recent communication from Dr. K.R.S.Morris points
out that in the initial clearing, by his selective technique,
of over 1,050 square miles in the Lawra District, tsetse
fly numbers were reduced from a catch of 2700 7500 flies
per year to 1 8 flies per year on cleared rivers.
Sleeping sickness was reduced by 92% between 1958 and 1944
and by 97% in 1947. The control programme affected some
232. Dr. Morris further points out that by the end of 1949,
3000 square miles of country were freed of G. palpalis and
G. tachinoides by selective clearing. The 1Titial 1100
square miles showed 99% reduction of trypanosomiasis,
virtual elimination of the disease. He found his method
of control so effective and rapid that "a severe epidemic
in south-west Wa, which showed a 100% rise in trypanosomiasis
incidence between .1940 and 1947, was got under complete
control between 1948 and 1949 with 73% reduction in number
of cases by the end of that year, i.e., immediately after
233. Continuing, Dr. Morris states that in 1949, 800 sq.
miles were freed from tsetse and 88,000 people were affected.
In that period there was only a single European officer
assisting. At present, the Department of Tsetse Control is
engaged in. an "expanding programme of trypansosmiasis
eradication because the clearings are so stable that they
can be turned over to the local Authorities for maintenance".
234. The utilisation of modern ecological technique in
the biological control of Tsetse as practised by Morris, and
the remarkable results attained by his methods of selective
clearings prove that this scheme of control is decidedly
effective. In my opinion it is the technique to be desired
where tsetse control must be undertaken in the Volta River
GENERAL DISCUSSION AND RECOMMENDATIONS.
235. In the light of present information, what effects
will impoundment have on G. palpalis and G. tachinoides?
The former species is distributed throughout the river
system where it is associated closely with the riverine
vegetation. G. tachinoides is confined to the woodland-
savannah zoneand it occurs along streams throughout the
Northern Territories and as far south as Kete Krachi, where
it likewise is associated with riverine vegetation. The
formation of the extensive Volta reservoir will flood the
present banks of the lower river and the lower sections
of the tributaries (map 2) killing the existing marginal
vegetation from the dam at Ajena as far upstream as Kete
Krachi, or possibly even a little further. The widespread
destruction of the fringing forests would, in effect,
temporarily accomplish the same results as the clearing
projects of Morris or Nash. But this destruction will
be confined to the present river banks, leaving a reservoir
of tsetse in the upper sections of the tributaries as well
as along that section of the river where extensive flooding
will not occur, that is, above Kete Krachi and below Ajena.
236. Unless steps were taken to prevent it, riverine
vegetation would become established along the banks of. the
newly formed lake within a few years. The normal difference
between high and low levels in the reservoir will be about
12 to 15 feet which is similar to the present fluctuation
in the river level from flood stage to low water. This
means that essentially the same conditions that now favour
the growth of fringing forest will be reproduced and would
favour the re-establishment of the typical vegetation.
During the drawdown, the dead trees of the old banks and
the flooded forests will be exposed, unless they are
removed prior to impoundage as recommended in the section
on malaria control; these places will provide additional
breeding grounds for tsetse.
237. I have pointed out in the discussion on malaria that
there will be marginal pools and marshes formed by the
reservoir at high water. Many of these will hold water
for long periods, perhaps they may even be perennial.
Within a short time the marginal vegetation that would
become established might easily form suitable areas for
the development of tsetse. However, if the recommendation
made in the malaria section be followed that the marginal
swamps and pools be drained these habitats will be far
less dangerous from the standpoint of tsetse production.
238. That section of the Volta below Ajena will be un-
affected by impoundment in so far as tsetse is concerned.
It is this part of the river that is most heavily populated,
but, in spite of this, the clearing for farming of the
river banks and islands covered by riverine vegetation, so
far, has been insufficient to eliminate tsetse, although
they do not appear to be present in large numbers. Human
sleeping sickness in the lower Volta is causing no concern
on the part' of the Medical Department, and, in fact, may
be absent from this section of the river. The general
consensus of opinion seems to be that the disease occupies
a position of very minor importance in this section, despite
the presence of the vectors. Although epidemiological
surveys for this area are projected in the near future, it
is at present not possible to adequately evaluate the
potential dangers from epidemics of sleeping sickness. It
may be true that although the vectors are present, there
is an insufficient reservoir of the disease for flies to
become infected, and trypanosomiasis is thus disappearing
in the same way that malaria is disappearing in the United
States in spite of the widespread presence of the vector
239. I find that it is extremely difficult to assess
the need for tsetse control in the lower Volta below Kete
Krachi. The paucity of information about the incidence
of sleeping sickness in this region needs to be remedied
by surveys of the population. Davey (1948) summarizes
the position in this regard by concluding that the exact
status of sleeping sickness in the Gold Coast is still
unknown, but he assumes that it is endemic throughout the
country, though its endemicity is negligible over large
areas. This author claims that progress in West Africa
is dependent on the control of trypanosomiasis and that
such measures as are necessary to eradicate it should be
taken as a preliminary step.
240. With the importation of a large labour pool, people
will be drawn to the Ajena region from many parts of the
Gold Coast, as well as other parts of West Africa.
Sleeping Sickness surveys of the Northern Territories
have shown an infection rate of 1 per cent; in Ashanti
1.8 per cent, of 37,000 persons were infected. In some
villages 10-50 per cent of the population was found to
be infected while in others the disease was absent or
infections negligible. That means a part of this unsorted
migrating group will very likely be infected with
trypanosomiasis and may form a nucleus for infection of
the fly, which could easily spread the disease through an
area that now appears to be free of it. Therefore, during
the construction phase of the project, where G. palpalis
is uncontrolled, a potential danger certainly exists. How
can this problem be dealt with most satisfactorily? Will
it be necessary to undertake extensive clearing of fringing
forests, or is there some simpler method of handling the
241. As the danger appears to be only a potential one,
it seems to jne to be more reasonable to undertake the
following steps in preventing the introduction of
trypanosomiasis into the Ajena area.
(i) A thorough survey of the present population in
the villages from Kpong to Misikrom, on both
banks of the Volta should be undertaken to
determine whether there are infected humans
present. If any are discovered, treatment
should be instituted to rid them of the parasite.
(ii) The medical officer in charge of the health
service of the Volta River Development Scheme
should require that all persons coming into the
area to work be examined for possible trypanosome
infection. Anyone showing evidence of the disease
should be treated before being allowed near the
(iii) Migrant followers of the labourers wives,
children, etc., should not be permitted to come
within the construction zone where infected
individuals might transmit their infection to
tsetse. The medical officer should undertake to
check as many of these migrants as possible so
that adequate measures can be taken to eradicate
(iv) Check points should be introduced above
Misikrom and below Kpong for the examination
'of migrants travelling through the protected
area via the river. Infected persons should
be given treatment.
242. Although striking results have been obtained by mass
survey and treatment of infected individuals, treatment
alone is unable completely to eradicate sleeping sickness.
Such mass surveys and treatment have ordinarily been con-
ducted in zones of epidemic or highly endemic areas. At
an infection rate of 1% or less, any sudden increase in
incidence would soon be discovered and could be dealt with
adequately (Wilcocks and Corson, 1946).
As the presently suspected rate of infection with
Trypanosoma gambiense is presumed to be considerably less
than 1% in the Ajena area, the measures indicated in
points 1 4 above should be sufficient to control the
243. Even though trypanosomiasis may be controlled, but
possibly not eradicated by the recommended measures, the
problem of the tsetse still remains. It is somewhat of a
nuisance in the Ajena area during the rainy season; however,
complaints from both European and African workers during
1950 were not excessive.
244. Morris (1949) points out that in his experience and
that of other workers, eradication of tsetse is necessary
to eliminate the disease in true epidemic centres. "The
more lightly infected areas should be dealt with when this
first stage in the attack is completed. This is a matter of
economy as well as of expediency. Complete control of the
serious outbreaks can be expected to have a marked effect on
the incidence of the disease in the lightly infected zones,
and from our studies of epidemiology, it is evident that the
full scale measures that are necessary to reduce the serious
epidemic will not be required in areas of low infection.
Indeed, the factors may be so delicately balanced that a
quite minor interference, judiciously applied, might cause
the disappearance of the disease. It is in these areas, and
in those like the forest where tsetse eradication is at
present impossible, that one can profitably consider the
application, alone or in combination, of other measures
such as mass treatment of prophylaxis, the continued
destruction of flies at important points of contact by traps
or insecticides, or ever changes in the habits of the
people brought about for example, by the provision of
fly-free water supplies or the consolidation of scattered
hamlets into compact villages so as to reduce the periphery
of exposure to tsetse".
PART III : MALARIA, ONCHOCERCIASIS, TRYPANOSOMIASIS
AND THE ACCRA PLAINS IRRIGATION SCHEME.
ACCRA PLAINS IRRIGATION SCHEME.
245. Goincident with the development of the Volta dam,
a scheme has been put forward to use water from the
reservoir to irrigate the relatively dry Accra plains.
Some half-million acres of land would be supplied with
this water and land that now lies idle could thus be
brought into active production.
It is planned to extend the main canal from the
reservoir area along the 250 feet contour and to drain into
laterals from this section. The "Preliminary Report on
Development of the River Volta Basin" by Sir William
Halorow and Partners (25th July, 1950) includes a sketch
map of the proposed Accra Plains Irrigation Scheme showing
the extent of the .anal system and its relation to the
IRRIGATION AND MALARIA.
246. Soon after my arrival in West Africa, I learned
from Dr. Leonard Chwatt, Malariologist for Nigeria, that
he had recently prepared a "Memorandum on Malaria
Incidental to Irrigation Projects in West Africa." This
report is so comprehensive that I am taking the liberty
of quoting large sections of it verbatim. Although I was
unable to get Dr. Chwatt's permission to borrow his work
because of his absence from West Africa, I am sure that
he would heartily approve of my employing his report in
"The Report of the British West African Rice Mission
deals with a most impressive and ambitious project to
increase the paddy area in British West Africa immediately
by some 170,000 acres and ultimately by 2,000,000 acres.
With the exception of one larger area in Sierra Leone
and a small one in Nigeria in the Niger Delta, the Report
of the British West African Rice Mission does not favour
the cultivation of rice in mangrove swamps. Main emphasis
is laid on the enormous possibilities of growing rice in
reclaimed lagoons (Gold Coast), vast island depressions
and flats irrigated either by storage of flood water from
dammed rivers or by impounding water from large catchment
areas depending on rainfall,
"System of canals will distribute the water from
flooded rivers or better from large reservoirs. Drainage
by gravity or where necessary by pumping.
"The association of irrigation projects in tropical
and sub-tropical areas with epidemics of malaria is too
well known to be recounted in detail. The classical
examples of this association need only be quoted to
realise their frequency.
Tennessee River (Hales Bar Dam, U.S.A.) 1912.
Tjihea Scheme (Java, N.E.I.) 1917.
Gantt Reservoir (Alabama, U.S.A.) 1923.
Sennar Dam (Sudan) 1924.
Irwin Canal Project (Mysore) 1932.
Lower Chenab Canal (Punjab) 1927.
Sukkur Barrage Project (Sind) 1934.
Tanjore Irrigation Scheme (Madras) 1938.
"These are the most prominent examples of epidemics of
malaria following the impounding of vast quantities of water
distributed through a system of channels.
"Since most of these irrigation projects were carried
out with the view to increase the amount of wet crops,
chiefly rice, it is not surprising that they have raised the
problem 'Rice fields and Malaria'. It seems pertinent to
recollect what is known of this particular public health
aspect of agriculture in the tropics.
"Outbreaks of malaria due to rice cultivation have been
repeatedly described in Spain, Portugal, Italy, Bulgaria
and Northern Egypt. These reports need not be summarized
here as they did not refer to tropical areas.
"Information provided by numerous Indian reports can
be summarized as follows:-
"There is not one problem of malaria caused by -wet (rice)
cultivation but as many distinct problems as areas.
Differences are often very great and are due to climatic
conditions, methods and amount of cultivation, schedule of
farming activities, vector species, the amount of endemic
malaria existing prior to rice cultivation, composition of
soil, type of irrigation, standard of maintenance of channels.
"Three main malaria vectors are involved. A. annularis,
A. culicifacies (found throughout India and notorious as the
causative agent of severe and widespread regional epidemics
in North West India and Ceylon) and A. fluviatilis.
"When the vector is A. annularis all rice fields are
dangerous, when the vectoFris A. culicifacies the rice fields
are dangerous until the rice pTa-t is about one foot high
(fallow fields are very dangerous), when the vector is
A. fluviatilis all non-seepage rice fields are harmless,
Tu-t seeping foothills are very dangerous.
"Breeding of dangerous Anopheles following rice
cultivation, takes place either in -addy fields themselves
or in irrigation channels or in both, These two types of
anopheline breeding must be considered separately.
"The general consensus of opinion is that with the
exception of the North of Bombay Presidency dangerous
anopheline breeding in growing paddy fields is not unduly
frequent or not widespread and found either only in shallow
fields or foothill fields (Bengal) or in very small numbers
(Mysore) or only at the grassy edges of paddy fields
Central Provinces, Burma) or until the rice plant is about
one foot high or when the water is in a continuous motion
through the field.
"However, Senior White's opinion that 'ricefield
breeding is, with a few exceptions, a matter of definite
moment in controlling malaria in 4ast Central India'
must be taken into account.
"On the other hand there is ample evidence that very
prolific breeding of dangerous vectors occurs in fallow
fields after the harvest or in water collections directly
associated with rice cultivation e.g. in swamps created by
the run off from the rice fields, (Sind).
"In the second case, however, most important breeding
takes place in irrigation channels and seepages or leakages
Anopheles in Ricefields.
"It is difficult, if not impossible to arrive at a
general statement of the effect of irrigation on anopheline
breeding without a careful local investigation. One
country differs widely from another in ecological conditions
favouring the breeding of dangerous vectors. The preferences
of one species of mosquito are very different from those
of another species. Some Anopheles prefer clear, still,
shaded pools, some fast sunlit streams, other brackish,
tidal marshes, other silted muddy pools and puddles.
"The only positive limited statement is that an
increase of the water surface (or the water edge) provides
an increased opportunity for anopheline breeding.
"The main malaria vector in West Africa is A. gambiae
the breeding places of which are extremely varie, though
it can be stated that it is essentially a small pool breeder
with its breeding places entirely or partially exposed to
sunlight. Its breeding places are puddles, shallow ponds,
borrow pits, irrigation ditches, furrows, seepages, foot-
print and cattle hoof prints, edges of lakes and swamps,
rock holes, clay pits, dammed streams, etc.,etc,
"In equatorial Africa rice cultivation is of recent
introduction and published reports on Anopheles breeding
and malaria associated with paddy fields are very few.
"Observations made by M. Thomson in Sierra Leone (1946)
are therefore of great value. This author investigated
two areas where swamp water rice is grown, Maswari and the
"It was found that in Maswari, where rice fields are
cut up into small plots in which water is left undisturbed
from July till the end of November, either in a continuous
sheet or in separate pools, numerous larvae of A. gambiae
are present during all this time. Other larvaeT(A. coustani
and A. squamosus, both non-vectors) are present in small
"In the Great Scarcies, where rice is grown on river
banks and along the creeks the fields are not cut up into
polders. During and immediately after the rains when there
are wide expanses of fresh or slightly brackish water dense
breeding of A. gambiae is limited to the extreme swamp edge
and follows Te'e changing level of the swamp edge.
"At the beginning of the dry season when the continuous
sheet of water changes into numerous small pools scattered
over a wide area the breeding of A. gambiae increases very
considerably and lasts until the w-Tole area dries up.
A. funestus the other malaria vector was found
occasionally in November. Other Anophelines found in paddy
fields (A. coustani, A. obscurus, A. ziemanni, A. moucheti,
A_ squamo us) are non-vectors.
"According to personal unpublished observations made
in the Gambia (Cape St. Mary's and Bakau) in October and
November 1944 larvae of A. gambiae in empoldered fallow
rice fields were very numerous (average 52 Larvas per 10
dips of the scoop).
"They were rare (average 3 larvae per 10 dips) in a
field with a well known unharvested rice standing in a
sheet of water.
I"In overgrown ditches close to the paddy field larvae
of A. gambiae were fairly numerous (average 8 larvae per
10 dTps) and found together with A. pharoensis and
"From Kenya (Grainger 1947) comes the report on
anopheline breeding in a rice field (230 acres) near
Kisumu, irrigated by a channel from a river 3 miles away.
A. gambiae was a persistent breeder throughout two seasons
'th an average production varying between 0.3 and 3.5
mosquitoes per 12 square feet of paddy field per day.
A. funestus and A. constani were present during the rainy
season in small numbers. A. pharoensis (a vector of
secondary importance) was '-und consistently producing
between 0.01 and 0.2 mosquitoes per 12 square feet of
paddy field per day.
"The great Mamanve swamp near Kampala was originally
free from larvae of this species and only when the
reclamation began the conditions became suitable for
Sambiae, (Han3ock 1934).
"In French West Africa the Segoa Irrigation Scheme
(Office du Niger) and the large scale production of wet
crops was followed by a high increase of the density of
A, gambiae (Col. Bernet, personal communication).
"The same applies to Bamfora area in the French Ivory
Coast where peasant cultivation of rice has become very
popular and where rice fields consist of small bunded plots
irrigated by the rainfall and during the dry season by the
"shadoni" system of bucket irrigation. (Col. Bernet,
Effects of Defective Irrigation.
"Two types of irrigation are planned on a vast scale
by the British West African Rice Mission: the inundation
irrigation and the perennial irrigation. The first type
of irrigation is planned either permanently or temporarily
to precede the more technically involved perennial
"The inundation irrigation which is actually similar
to the tidal irrigation depends on the level reached and
maintained in the river during the rainy season. The
perennial irrigation depends on permanent impounding of
water in big rivers to a determined height by means of dams.
The impounded water is delivered from a higher level system
"While the inundation irrigation may be conducive to
mismanagement of water and to a seasonal increase of malaria
it is the perennial irrigation which causes the greatest
amount of malaria incidence of all irrigation systems
practised in India. (Covell 1946).
"According to Covell the principal causes of malaria
associated with perennial irrigation are:-
"Increased height of subsoil water and the greater
tendency of the waterlogged soil to form long-standing
pools and puddles during the rains. These pools and
puddles are very extensive in fallow fields, close to the
"Prolific breeding in canals where they are temporarily
shut off either for repairs. or for periodical distribution
of water. These canals when re-opened carry anopheline
larvae into the subsidiary channels.
"Along the high level canals there are string o.f water-
"Breeches in canals due to natural causes or illicit
removal of water produce numerous breeding places.
"Bad maintenance and subsequent leakages from small
channels under the "control" of local population.
"Obstruction of old canals or streams by roads or new
canals which form land-locked depressions.
"Insufficient number of bridge .crossings and subsequent
damage to the banks by traffic.
"The dramatic history of the Irwin Canal Project
reported by Sir Gordon Covell gives a perfect illustration
of widespread effects of a perennial irrigation scheme
where these incidental but serious defects were common.
"Similar sequence of events followed the construction
of the Lower Chenab Canal (Punjab), (Lal and Shah 1933) the
Sukkur Barrage Project (Sind) (Covell and Bailly, 1936) and
the Irrigation Scheme at Pattukkotai in Tanjore District
(Madras Pres) (Russell 1938).
"P. Russell (1938) studied in great detail the
epidemiology of malaria caused by the Pattukkotai Irrigation
Scheme and drew the attention to specific factors
contributing to the increased breeding of malaria vectors
due to defective and untidy perennial irrigation.
"Those factors are classified under two sub-heads :-
(a) General effects, and (b) Defects in construction,
operating and maintenance of the irrigation.
(a) General Effects.
Dry faluk (fadama, in Nigeria) becomes a wet one and
wet crops are introduced where dry crops were grown;
Fallow fields remain wet longer than before;
Rivers carry more water during the season;
Subsoil water level rises with the result that old
deep wells hold water at high level. This permits breeding
of Anopheles in wells previously unsuited to it. It causes
seepages and springs to appear where none were seen before.
It makes it easy to dig shallow wells during the dry season
which provide numerous breeding places;
Unplanned streams became numerous and dry ditches filled
(b) Defects in operation and
maintenance of the irrigation system.
Excessive supply of water which overflows and gives
rise to pools along the canal;
Seeping canal banks with water oozing through the
poorly consolidated sides;
Insufficient number or a total absence of drainage
canals to lead the irrigation water out into natural river
Absence of planned and controlled system of field
Cutting off sections of old canals which become blind
ends or rather elongated ponds;
Breaching of canal-banks by villagers deprived of
bridges and by cattle. Poor maintenance of canal banks
conducive to scouring, or growth of marginal vegetation or
silting. The old bug-bearl! borrow-pits dug to get spoil
for the banks;
Defective sluice gates and distributing chambers,
"Not less interesting (on account of the geographical
position) is the history of the Sennar Dam in the Sudan.
The Sennar Scheme is based on perennial irrigation from a
Nile reservoir irrigating 300,000 acres. The work began
in 1919. In 1923 the irrigation was carried to the
subsidiary and terminal water canals.
"In 1924- at Wad Medani the control of water became a
matter of great difficulty, water being pumped into canals
that were already full. There was considerable waterlogging
and it became impossible for sanitary gangs to cope with
the widespread leakage of water followed by a high mosquito
infestation. A sharp epidemic of malaria occurred, which
threatened to bring the work to a standstill. The situation
was eventually improved in 1926 after a large expenditure of
money and labour.
Irrigation and Anophelism in West Africa.
"It-would be over-optimistic to believe that all or
most of these defects will be absent in irrigation projects
in West Africa.
"On the contrary it is thought that in West Africa
the increase of the breeding activity of the main wide-
spread vector (A gambiae), the other, more localized
vector (A. funet-us) and some secondary vectors
(A hargreavesi, nili) will be very great.
"Even the best irrigation scheme, an example and a
model of major engineering, depends on the management of
terminal parts of the scheme, parts that are outside the
scope of the technical staff and are maintained by the
agricultural community, which directly benefits from it.
"In contra-distinction to the Far Eastern and North
African peoples the West African is not traditionally
irrigation-conscious and will have much to learn before
he realizes the importance of an adequate maintenance of
the small subsidiary channels that brings the irrigation
water to his field.
(A lot of the trouble, the French (Office du Niger)
are having with the Niger Irrigation Scheme near Segou
is due to the carelessness of the native farmers).
"As shown by Russell, it is this unspectacular and
common place peripheral part of the irrigation scheme that
is usually neglected and produces innumerable seepages,
leakages, pools, puddles, in which breeding of many
Anopheles species is abundant.
"The argument put forward by Bentley in Bengal (1925)
that large areas of water are less dangerous than small
pools holds true since the amount of Anopheline breeding
in West Africa is proportional not to the area of water
surface but to the length of water edge.
"It happens that the most dangerous malarial vector
throughout Equatorial Africa breeds by preference either
along the trimmed, overgrown edges of collections of
standing or slowly flowing water or better still in
muddy, sunlit, small collections of standing water.
"It adapts itself easily to many waters from fresh
to brackish, from clear to muddy, from open to lightly'
shaded, from clean to slightly polluted.
"A. gambiae will have the whole gamut of these
breeding places readily available in connection with any
irrigation scheme in West Africa.
"As everywhere so in West Africa the effect of
irrigation schemes on the increased breeding of malaria
vectors will depend on the type of irrigation, on the
features of the irrigated areas and on the method of
wet crop cultivation.
"The primitive irrigation of small empoldered fields
as seen here and there in the Gambia and Sierra Leone is
certainly dangerous since it creates many breeding places
of anophelines which have a preference for standing in
muddy, sunlit water.
"The effects of tidal irrigation on breeding of
A. gambiae was described in the preceding chapter and there
can be little doubt that the inundation type of irrigation
contributes substantially to the output of the main vector
(A. gambiae) to a limited extent during the growing period
or rice but far more profusely after the harvest and in
"The breeding in the actual growing rice fields is
still insufficiently investigated but it is already obvious
that the breeding of A. gambiae in these fields is strictly
seasonal and not dense. however it must be remembered
that a small A. gambiae output per unit field over a large
area may become a problem. The great capacity of A.gambiae
breeding in fallow fields has already been mentioned and
the seriousness of this cannot be over-emphasized.
"Another still open problem is the possibility of
A. funestus (the other important malaria vector) breeding
Tn growing rice fields. Findings of A. funestus in growing
rice fields were reported from Kenya by Aders, Other
secondary vectors (A. ni, pharoensis, hancocki, hargreavesi,
moucheti) may be temporarily dismissed.
Technique of Control.
"Control of anopheline breeding in rice fields is a
perplexing problem to the malariologist. The wide expanse
of shallow warm water with vertical vegetation, the necessity
to keep the field flooded, the susceptibility of the growing
rice to injury by larvicides present difficulties that are
not easy to overcome.
"A detailed entomological survey of tropical rice-field
breeding, carried out particularly in Malaya and in India
has revealed that the amount of dangerous anopheline breed-
ing in paddy fields is relatively small and. that most of
the breeding occurs in adjacent low areas covered with
seepage water and in badly designed or maintained irrigation
"Similarly the problem of control in Africa will
immediately be divided into (a) Rice field control and (b)
"Rice field control; If the main amount of breeding
occurs along the edge of the field it can be reduced by
trimming the edges and keeping them free from vegetation.
"The application of larvicidal oil has been condemned
by most investigators since it damages the crops.
"However. Grainger in Kenya (1947) found that in the
control of A. gambiae the oil drip technique was the only
procedure that was successful. There was no decrease of
the yield of the oil treated crops, on the contrary some of
these plots gave the highest yield, probably because of the
toxic effect on the "blast" a fungus that affects the wet
"The application of Paris green was extensively
discussed by Covell (1935) who concluded that this
larvicide applied in the usual one per cent. dilution
exerts no harmful effect on the rice crops. Rao and
Sweet (1937) have proved in Mysore that Paris green
(in one per cent dilution in road dust or wood ash) has
not decreased the rice and straw yield or increased the
amount of chaff and that it can be employed even during
the period when the paddy is in flower.
"There can be little doubt that D.D.T. either as
dust or as an emulsion will give the same or even better
effects without injuring the paddy, which is mainly wind
"An interesting method was developed by the University
of Arkansas Rice Experimental Station (Knowles and Fisk,
1945). The method consists of introducing a water miscible
D.D.T. compound into the irrigation waters at the pumping
station. It was shown that a complete control of anopheles
(quadrimaculatus) was obtained with D.D.T. concentration
of one part per million. In comparison with a controlled
plot the treated plots contained 50% less larvae, but it
should be noted that the production of mosquito larvae by
this method was not entirely eliminated.
"Airplane distribution of Paris green or D.D.T. oil
solutions or water suspensions is an ideal method for large
"According to Bagster Wilson (1947) air spraying in
Dar es Salaam and Kisumu from between 200 to 400 feet with
a D.D.T. oil emulsion at a dosage of 32 mgm D.D.T. per
square meter gave an immediate 98 per cent reduction of the
anopheline larval density even though the vegetation cover
"The cost of the air spraying technique was not cal-
culated in this experimental work. According to American
sources the cost of airplane dusting with one Stearman
biplane amounted to about 6 shillings per acre. (Murray
and Knutson, 1944).
"Of all naturalistic methods of control of anopheline
breeding in rice fields the intermittent irrigation is the
most important. It was used with great success in Portugal,
Bulgaria, French Indo-China and particularly in Java (Smalt
1937) where it was enforced by law. Russell, Menou and Rao
(1942) studied this method successfully in the Tanjore
district of the Madras Presidency and obtained a complete
control of A. culicifacies by interposing 2 to 4 dry days
between each 5 wet days. (When the rains occur during the
growing period of the rice the intermittent irrigation is
not very effective).
"It was found that this cycle does not adversely affect
the yield of grain or straw.
"The technique of intermittent irrigation required good
sluice gates and channels and above all a very careful super-
vision. Types of soil and climate are important factors in
delimiting the optional dry wet cycle for each locality.
"However, the 3 dry to 7 wet days intermittent
irrigation used in Kenya (Grainger) was not very successful
and the numbers of larvae A, gambiae found bore little
relations to the periodicity of the irrigation.
"The use of Larvivorous fish (Gambusia and others)
(Murray and Knutson 1944) cannot be successful in vast and
shallow rice fields. Grainger in Kenya found that the
introduction of Gambusia had little effect on Anopheles
production although fish did well.
"Modern developments of the technique of residual
spraying suggest that it may be possible to control effect-
ively not only the adult mosquito density but also the
larval breeding in rice fields simply by treating all the
inhabited houses in an area by residual D.D.T. or B.HC.
spraying. This striking result was achieved in Mexico in
the Cuernacava area where a 5 per cent D.D.T. water emulsion
applied to all native houses at the dose of 200 mgm D.D.T.
per square foot has reduced the density of adult mosquitoes
(A. pseudopunctipennis) by 98 per cent and the numbers of
larvae by 90 per cent (Gahan and Payne, 1947; Stage 1946).
"Control of anopheline breeding due to
"The sudden conversion of a stream or river into a
broad lake results in major hydrological changes since it
tends permanently to flood the land below the level of the
impounded water. The biological changes follow and the
result is that the vegetation in the submerged area is killed
but the marginal growth of the vegetation creates favourable
conditions for breeding of anopheles,
"This was the guiding principle of malaria control of
impounded water so brilliantly carried out by the malario-
logists of the Tennessee Valley Authority. The technique
used was the treatment of the reservoirs before impounding,
correct timing of the impoundage, periodical fluctuation of
water level, marginal clearing, etc.
"All these control measures adopted by the T.V.A. will
be of very limited value in West Africa. Breeding of
A. gambiae is not necessarily associated with the presence
of vegetation. This particular mosquito shows a prefer-
ence for shallow, muddy pools without any or with little
"There can be no doubt that a control of breeding of
A, gambiae due to an irrigation scheme will depend to a far
greater extent on adequate dealing with all multiple,
deceptively small, widely scattered, difficult to supervise
factors that make an irrigation scheme untidy and were so
pointedly enumerated by Russell (quoted above).
"A good control of these factors will require: (a)
Very careful preliminary planning of the distribution of
water and an efficient supervision of it when the irrigation
has started, (b) Wide legal powers to be able to enforce
a proper maintenance of irrigation channels by farmers.
(c) An adequate malaria control organisation working hand
in hand with the engineering services.
247, The statements quoted above adequately demonstrate
the very great dangers from malaria inherent in any irriga-
tion scheme, particularly where a mosquito as adaptable as
A. gambiae is concerned. It is therefore of the greatest
Importance that the Accra Plains Irrigation Scheme be care-
fully co-ordinated with the malaria control project of the
Volta river development. The malaria control team, provided
that it is supplied with all necessary equipment and supplies
and sufficient supervisory and junior staff, should be able
adequately to cope with the problem. In summary, I would
like to reiterate four major recommendations:
(i) There should be a very careful preliminary planning
of the distribution of water and an efficient
supervision of it when irrigation has started.
(ii) Wide legal powers should be granted to the
authorities to permit them to enforce a proper
maintenance of the irrigation channels by the
(iii) An adequate malaria control organisation should
work hand in hand with the engineering services.
(iv) The engineering services, the supervisory
authorities, and the malaria control organisation
must work constantly to avoid the malpractices
pointed out by Dr. Chwatt.
248. The dry Accra Plains are entirely free of Simulium
at present, but with the instituting of the contemplated
irrigation scheme, there is some danger of S. damnosum
spreading into the area. I have been informed y Sir
William Halcrow and Partners that the maximum rate of flow
of the water in the irrigation ditches will be 3 feet per
second. It has been pointed out in a previous section of
this report that in the Congo, S. damnosum larvae thrive
in a current of 3.5 to 4.5 kilometers per hour (3.2 to 4.1
feet per second), and will survive and mature at 3 kilometers
(2.7 feet per second). Since ecological conditions maybe
produced which are suitable for the development of damnosum,
preliminary planning should be undertaken to deal with this
eventuality if it should arise.
249. Assuming that aerial spraying of the Volta rapids
is successful in eradicating Simulium from the river, I
believe that it would be most unlikely to find the insect
in the irrigation drains provided the eradication program
starts before the irrigation channels are in operation.
The only available source of adult S. damnosum, from which
the species might readily reach the-drains, is located in
the Volta. Invasion from outside the river is such a
remote possibility that it can be dismissed.
250. If eradication in the Volta is not undertaken or
not successful, damnosum may become established in the 1ccra
plains. In the paved irrigation ditches it should be
relatively easy to eliminate any breeding by using Garnham's
and McMachon's (1946) method of dripping D.D.T. emulsion on
to the water surface at intervals along its course. These
workers have shown that S. neavi could be eradicated from
small streams in this manner. The nature of the irrigation
system will prevent control by intermittent flow in the
channels. Routine brushing of the drains to remove
larvae will be more expensive and less effective than the
use of the larvicide.
The possibility of spreading S. damnosum into the
Accra plains simply reinforces the argument for aerial
fogging of the rapids'of the Volta to eradicate the species,
and so stop the transmission of 0. volvulus,
251. In 1929 and 1930 A. W. J. Pomeroy conducted surveys
of the tsetse flies of the Accra plains. He pointed out
that a great part of the plains are waterless in the dry
season, and under those conditions tsetse are restricted to
the few localities where water is permanently present and
shade conditions are suitable. His collections revealed
that considerable tsetse infestations follow the course of
every small stream during the rains, and, where large
surface pools form and sufficiently dense thickets occur,
tsetse migrate to them. Pomeroy's collection produced
Glossina palpalis from the following places:
Palupa river at
Malagwe and Keli
December 3, 1929.
December 3, 1929.
March 4, 1930 (Pomeroy
postulates movement of tsetse
from the Volta along the Hwapa;
thence into the Sagreme, a
stream a mile from the Hwapa)
February 26, 1930 (a stream
valley crossing the road
between Dodowah and Prampram).
March 7, 1930 (stream valleys
north-east of Old Nugo).
March 7, 1930 (located near
the river Volta south-east of
March 10, 1930.
Pomeroy expressed the opinion that the section of
the Accra plains between Tamatuku and Ada is free of tsetse.
252. Nash's investigation (1948) revealed that the Accra
coastal plains now support approximately 53,000 cattle. He
indicates that the plains are south of the range of
G. longipalpalis, since the species is unable to enter the
grass lands. He also reports that G. palpalis is present
in the few stream beds. Zebu cattle were crossed with the
dwarf cattle in this area, and the resultant cross-breed
herds are in much better condition than similar herds in
the north. Map 1 of Nash's report shows that the Accra
plains are free of human sleeping sickness, as well as
being one -of the two areas in the'Gold Coast where Zebu
cattle can be kept.
253,. In early July, and also later in the month, I
travelled over the Accra-Ada road and on both occasions,
every stream valley which the road crossed, was absolutely
dry although the early rains had just ended. The road from
Accra to Akuse follows the upper edge of the Accra plains.
Here, too, the country was extremely dry from mid-June to
late October, and I have never seen water standing for more
than a few days in the majority of stream valleys which the
254. With irrigation, and consequent soaking of the
ground, it is likely that the stream beds of the Accra plains,
which are now dry for most of the year, will change in
character and will carry water in the dry, as well as in the
rainy season. Under conditions such as these, tsetse will
be favored since they will no longer be restricted to small
foci as they are at present,
255. Irrigation and the concomitant farming may serve to
negate the influence of irrigation alone on improving con-
ditions for tsetse. If the banks of the present stream
valleys, as well as those of water holding depressions, are
farmed, the necessary removal of the trees will result in a
reduction, or possible eradication, of G. palpalis.
256. Where there is fringing vegetation suitable for the
development of tsetse or for harbouring adults, and the land
is not farmed, clearing should be undertaken, provided that
the area is used by humans or cattle. The amount of clear-
ing required will be relatively small and inexpensive, and
should be conducted before irrigation begins so that the
flies presently established in the area can be eliminated
at a time when their numbers are at the lowest level.
257. Pomeroy (1930) was of the' opinion that much of the
infestation of the Accra plains with tsetse came with the
spread of the fly shortly after the onset of the rains. If
this is true, and it certainly seems reasonable, clearing
for control of tsetse should be conducted in such a way as
to widen the gaps between the permanent population along the
banks of the Volta and the small streams draining across
the Accra plains. If this clearing could be conducted
during the dry season, repopulation of the Accra plains
would be difficult during the rains, when it normally occurs,
at the same time when the gaps between the headwaters of
the Volta tributaries and the Accra plains streams are being
widened, the small residual populations can be destroyed
easily by a minimum amount of clearing.
258. Since the present limitation on the number of cattle
kept on the Accra Plains seems to be due to lack of food and
water, and not trypanosomiasis, it is likely that the herds
will be greatly benefited by the irrigation scheme, especial-
ly if all danger of future sleeping sickness infections is
PART IV : OTHER INSECT BORNE DISEASES AND THE VOLTA
259. This is a disease produced by the round worm,
Wuchereria bencrofti, and transmitted to' humans by various
species of mosquitoes. Filariasis has long been known in
West Africa as well as in other parts of the world between
30 N. and 32 S. latitude. It became particularly well
known during the last war when the armies in the South
Pacific Theatre devoted considerable attention to it, and
emphasized the external manifestations that are sometimes
produced, in order to induce the soldiers to take precautions
against mosquito bites. It has been proved that the enlarged
body parts are associated with filariasis in only a small
fraction of infections (usually less than 5%.o)
260. So far as I can ascertain very little is known about
the incidence of filariasis in the lower Volta basin. The
general opinion seems to be that the disease is rare or
absent. Certainly, the more overt effects of the disease,
elephantiasis, are almost never seen in visits to villages
along the river. In fact, the only persons that I encounter-
ed during my survey who showed such symptoms, were three men
in Ada. Whether their elephantiasis was due to filariasis
or not remains to be demonstrated. The "Annual Report of
the Medical Department" of the Gold Coast repeatedly, year
after year, states that filariasis is reported from most
stations, but in no instance is undue local importance attash-
ed to it; however, no specific information is given about
its location or incidence.
261. Complete development of the Larvae of W. bancrofti
has been demonstrated in 30 species of mosquitos distributed
in various parts of the world (Manson-Bahr, 1945). It is
rather fortunate that, of the large group, Manson-Bahr in-
criminates only two Gold Coast species, On the other hand
Evans (1938) includes another species, and it is most likely
that four others can serve as vectors in West Africa, al-
though proof has not been established. The species are as
Known vectors: Potential vectors (from U.S.
Army Bulletin, 1944):
Anopheles Rambiae Culex fatigans
Anopheles funestus Aedes aegypti
Anopheles nili Mansonia africanus
262. Evans reports that Gordon, in 1932, dissected 22
female A. nili taken near Freetown and found 4.5% of them
with inTec 7ions in their thorax; Barber found an incidence
of filaria of 2.7% in 254 females, Evans points out that
where A. funestus occurs, it must be an important vector and
cites Six references to workers who have found the species
infected. A. gambiae is probably the most important carrier
of W. bancroti. Infections of this species have been
reported by many workers.
2635. None of the species of mosquitoes listed as
potential vectors has ever been incriminated in West Africa,
although Manson-Bahr lists both M, uniforms and M. africanus
as species in which development Thas been observedin- Central
264. My collections of mosquitoes from African houses
has produced all of the species which I have listed as
proven or potential filaria vectors. Of the lot,
A. gambiae and A. funestus are undoubtedly the most to
Be feared. The possible effects of impoundment of the
Volta on these species has been discussed in the section on
malaria control and steps to be taken against the mosquitoes
have been detailed. Measures taken to control malaria
mosquitoes will, at the same time, control the proven vectors
of filariasis. The threat of this disease is yet another
argument reinforcing the recommendations for mosquito
265. loa loa is widely distributed in West Africa and is
especially common in the Cameroons, Little information
about the incidence of the disease in the Gold Coast is
available and whether it is endemic in the country is still
to be determined. Nevertheless, I believe that the disease
should be considered in the overall picture of the insect-
borne diseases of the Volta river basin.
266. The most recent study of the deer-fly, Chrysops,
the known vector of loasis, is that of Gordon, Chwatt, and
Jones (1948) conducted as a preliminary survey at Kumba,
British Cameroons, where the disease is common. After
Dr. Chwatt discovered the habitat of the immature stages
in 1947, interest was aroused in the insect and the authors
travelled to Kumba in 1948 to undertake a brief study of its
biology. During the investigation several important dis-
coveries about C. dimidiata and C. silacea were made, the
most important Seing the location of the larval habitat,
which previously was unknown. They found that the breeding
places of the fly were very restricted, and appear to be
confined to certain habitats in densely shaded streams where
slowly moving water passes over a layer of mud covered by
decaying vegetation. The larvae were found in the mud
usually about three inches from the surface.
267. Observations on time of feeding by the deer-flies
showed that they began to bite after 8.30 A.M. and ceased
usually by 4.30 P.M. The period varies considerably,
depending on the weather. Of 500 flies dissected at Kumba
(480 C. silacea and 20 C, dimidiata), 36 or 7.2% were found
to be-infected with filaria which were presumed to be
268. In the Kumba survey, examination of the Africans
showed a higher incidence of the disease in adults than in
children. Gordon et al point out that Davidson, working
in the same town in 1944, examined 870 persons and found
3.7% of the children and 23% of the adults with microfilaria
in their blood. Chwatt, in 1947, found 6%o of 110 children
and 6.5% of 84 adults with the filaria.
269. A careful search at all streams where collections
were made in the Volta basin did not produce a single
specimen of Chrysops. The tsetse fly-boys, working along
- the Volta from the Ajena dam site to Misikrom from late
September to early November, did not collect Chrysops, al-
though other tabanids including Tabanus, fippocentrum,
Haematopota, and Pangonia were taken by them. These results
certainly do not mean that the flies are absent from the
Volta and its tributaries, but only that their density was
very low during the time of year collections were made, or
Chrysops does not bite humans in this part of the Gold Coast.
270. It would be extremely useful to know what the pre-
valence of loasis is in the Volta basin. As the microfilaria
show a'diurnal periodicity, a survey for this disease might
easily be combined with malaria surveys. The search for
Chrysops should be continued and, if it is discovered, steps
should be taken to protect the inhabitants against its bites,
provided the presence of Loa loa is also established.
Fortunately, the distribution--oT the fly appears to be rather
local and widespread and expensive control would be
271. At present, no effective means are known for the
control of Chrysops, but Gordon, Chwatt, and Jones suggest
that clearing the bush over the streams in which the larvae
live might be effective. They also suggest spraying the
water with D.D.T. in oil. Clearing the bush from around
houses also deters them from entering to bite.
272. This is another filarial worm which apparently is a
harmless human parasite. It is widespread through much of
West Africa and is known to occur in the Gold Coast, although
no studies of its incidence appear to have been made.
The worm has been shown to be transmitted by
Culicoides austeni and C. graham, midges that are common
in many parts of the GoT Coast. Both species were
collected by me at several of the smaller tributaries of the
Volta. At one stream, on a cloudy day, they were biting
viciously in the early morning under dense shade.
273. The insects breed commonly in rot-holes of trees;
therefore, control of the midges would be extremely difficult.
Since they bite mostly at night, screened quarters, in which
the screens have been treated with D.D.T. or gammexane,
would offer adequate protection against the pests. They
are small enough to go through 18 mesh screening, but in
doing so will contact the insecticide. Normally, the
insects do not fly far, and clearing bush around dwellings
will do much to discourage their presence.
274. Dissections of Anopheles gambiae in Liberia by
American entomologists during the war, showed a small
percentage of the mosquitoes also carried the infective
larvae of A. perstans. Steps taken in mosquito control
will, therefore, eliminate the possibility of transmission
from this source.
275. The coastal area of the Gold Coast has long been
known for its epidemicaof yellow fever, the last occurring
in 1937. The Medical Department has kept a close watch to
detect any possible resurgence of the disease, and although
an occasional case is seen, there have been no further
276. Aedes aegypti, the well-known yellow fever mosquito,
is particularly susceptible to control as was shown in
Brazil during a vigorous eradication campaign. The Gold
Coast has met the problem by appointing sanitary inspectors,
who have as one of their jobs the locating and destruction
of any breeding places of the species. It is particularly
fortunate that the mosquito utilizes man-made objects so
extensively for its immature stages, as these can easily be
dealt with by a house to house inspection. Hopkins (1936)
reports that aegypti breeds commonly in tree-holes, but
that it is more common in artificial breeding places such
as barrels, discarded cans, bottles, automobile tyres, vases,
gutters, or other water-holding articles of that sort.
Sometimes it is found in the axils of banana trees. In the
Accra region, I have also found the larvae on rare occasions
breeding in ground pools.
277. Because of the nature of its larval habitat, aegypti
will be only indirectly influenced by impoundment of he
Volta rivoc. As the mosquito will not breed in the reservoir
itself, this huge lake will offer no problem. The problem
will be found in the mass migration of people into the Ajena
area and to the shores of the reservoir, bringing countless
containers with them that may hold water sufficiently long to
breed A. aegypti. Through the long years of effort by the
Medical Department in educating the Gold Coast people to
the need for eliminating the mosquito, most of the new
settlers will be co-operative in emptying water from the
containers. The Gold Coast government reinforces this co-
operation by imposing a small fine on the person living on
the premises where aegypti is found breeding.
278. Soper et al (1943, p,5), in the description of their
fight against aegypti on Brazil, point out that a "permanent
campaign falls naturally into three quite distinct phases:
(i) The initial clean-up campaign for the elimination
of the easily accessible aegypti foci. This
phase is similar in all respects to the early
emergency yellow fever campaigns which rid
localities of yellow fever but not of the
(ii) The discovery and elimination of the final
hidden, inaccessible breeding places responsible
for maintaining the species in the face of
extensive anti-aegypti measures.
(iii) The permanent maintenance of a sentinel service
to discover and eliminate any reinfestation which
may occur. (Reinfestation in the early months of
the sentinel service may be either 'internal',
due to returning to active use of dry containers
with viable eggs, or 'external', due to importation
of the vectors from infested areas, as already
described. Later reinfestations come only from
279. Control of A. aegypti should be one of the first
health problems attacked n the Ajena region. As the
larger towns have the-ir regular sanitary inspectors who
search for Aedes, they need not be considered in the
overall problem. I would like to recommend, however,
that regular inspection for aegypti breeding in the
region of the dam site and all nearby dwellings be
instituted without delay.
280. The recently discovered relationship existing
between jungle yellow fever and human infections has led
to extensive research on the problem in parts of South
America and Africa. Taylor and Theiler (1948, p.508)
explain that "besidesA. aegypi severalsother species
of Aedes may be involvetrin the transmission of the
disease from man to man. Aedes simpsoni has been in-
criminated and Aedes vittaTus, taylori, furcifur suspected.
These mosquitoes have diverse breeding habits, and they
are less amenable to antilarval control. A. simpsoni,
for example, is a plant axil breeder and is found not only
in plantations about human dwellings but also along the
edges of forests. In Africa even A. aegypti is not so
uniquely domestic in its habits as-Tt is in the Western
Hemisphere, and may be encountered in forests far from
human habitations. Urban outbreaks of yellow fever still
occur in Africa, and in 1946 there was a rather extensive
epidemic involving a number of large towns and cities in
Nigeria. Whether or not this and similar outbreaks in
the past represent extensions from unrecognized endemic
urban foci or result from the introduction of the virus
from forests is uncertain. There is good evidence for
believing, however, that in parts of East Africa village
epidemics are periodically initiated by A. simpsoni, which
after becoming infected from marauding monkeys, serves as
vector in transmitting the disease from man to man.'"
281. The jungle cycle of the disease does not involve man
directly, but it is important in serving as a source of
infection for persons entering forests where the virus is
present. It is also important as it represents an un-
controlled, and perhaps uncontrollable, reservoir of the
virus that may give rise to epidemics. A. africanus,
one of the supposed chief vectors of jungle yellow Tfever,
feeds at night, and as man seldom enters the forest after
dark, yellow fever only infrequently gets contracted by
282. In West Africa, where A. aegypti is the principal
urban vector, epidemics are usually associated with
infection of primates in the adjacent forest (Taylor and
283. I have introduced the subject of jungle yellow fever,
not because the construction of a dam on the Volta will
have any effect on the spread or decline of the disease,
but solely to indicate that a potential source of the
disease does exist in the vicinity of the dam site. It
is impractical to even consider the control of mosquitoes
with such breeding habits as those of the incriminated
species of Aedes, which live in the surrounding forests
where monkeys are also present in large numbers; however,
it is highly desirable to immunize all workers in the
Ajena area. A scheme of country-wide immunization has
been discussed for the Gold Coast, but no action has yet
been taken; therefore, local protection must be depended
upon to protect personnel.
284. Dengue is another of the virus diseases transmitted
by mosquitoes. Aedes aegypti is the chief vector involved
in the transmissTonof he disease in West Africa. "The
Annual Report of the Medical Department'" of the Gold Coast
from time to time mentions small outbreaks, but the
incidence of dengue has never been high as sometimes occurs
in epidemics in other parts of.the world,
285. The same measures directed against Aedes aegpti
in the control of yellow fever are also applTcabe o the
control of dengue.
PART V : SCHISTOSOMIASIS AND GUINEA WORM.
286. This disease is another important one which appears
to be widespread throughout much of the Gold Coast but with
isolated centres of particularly high incidence. The
intermediate hosts are not insects, but as some preliminary
statement was desired about the disease, I shall include a
short discussion here. In September, Dr. Elmer Berry,
malacologist of the United States Public Health Service,
came to the Gold Coast to arrange for a special study of
the disease and its intermediate hosts that he will under-
take in 1951. During his brief stay, I was fortunate in
being able to discuss the problem with him and to get some
of his general impressions about possible effects of
impoundment and irrigation on the snail hosts,
287. A check of the "Annual Report of the Medical
Department" of the Gold Coast for the years 1925 through
1948, shows that the number of persons treated for
schistosomiasis has gradually increased through the years.
In 1948, the latest year for which figures were published,
5,274 cases were reported and 12 deaths were attributed
to it. The 1945 report claims that the disease is much
commoner than the figures of treated individuals reveal,
and that the disease is the cause of much chronic ill
health. The same conclusion was expressed in 1938 but
the-report for that year specifically stated that ida,
at the mouth of the river Volta, and Akuse, a town about
a mile from the river, suffered from a high incidence of
the disease. Doubtlessly, comprehensive surveys would
reveal a high proportion of infections in many of the
villages and towns of the Volta basin.
288. Although both Schistosoma haematobium and S. mansoni
both occur in the Gold Coast, the former species appears
to be more prevalent in the lower Volta region. The
distribution of these two parasites is closely linked
with that of the non-operculate snails of the genera
Physopsis and Bulinus. The life cycle begins with the
passing of ova from the definitive host.The eggs hatoh on
reaching water and liberate the free-swimming miraoidia.
This stage then penetrates an appropriate species of snail
in which it undergoes a metamorphosis to form primary and
secondary sporocysts and cercariae. The cercariae leave
the snail and, if successful, penetrate the skin of man
beginning a new infection.
289. Wartime research has indicated that schistosomes are
primarily parasites of children. Children commonly play
in and around water that may be contaminated by human
excreta, and in which the host snails live. Adults,
likewise, may become infected in these waters, but in
endemic areas, almost all persons have become infected
by the time they reach maturity. Natives have observed
the nature of the disease in children, and, when asked
about it, almost always are of the opinion that only children
290. Even in endemic areas, distribution of the disease
is spotty, because ecological conditions may not be satis-
factory for the snail or for the parasite. Large bodies
of water and swift flowing streams are not so dangerous as
are the slow flowing or stagnant streams, ditches, drains,
291. In spite of the complicated life cycle, the wide
prevalence of schistosomes makes it clear that the parasite
is eminently successful in propagating itself. There are
three points of attack in breaking the man-parasite cycle.
The first, and most logical, is to prevent the access of
human excreta to water. Sanitary disposal might solve
part of the problem, but not all. Promiscuous urination
and defecation, particularly by children, is widespread and
it would be especially difficult to make an illiterate
population conscious of the needs for even primitive sanitary
measures. Nevertheless, a well-planned campaign might go
far toward discouraging fouling of the drinking water and
292. The second method of attacking the problem is to
destroy the snails which serve as intermediate hosts. Huge
sums of money and great effort have gone into projects
designed to do just this in Egypt, but success has been
limited and the disease still takes a great toll. Recently,
new molluscicides have been developed which Dr. Berry
believes hold great promise in controlling or eradicating
snails in limited areas. For wise use of this method,
comprehensive surveys would be required to locate endemic
centers of the disease, then molluscicides could be applied
to all water in the vicinity in which the species of snails
involved in schistosomiasis transmission were living. This,
at least, would contain the disease until educative methods
could begin to influence the people.
2935. A third method of attack, and the one with the
greatest promise of success, is to supply all villages and
towns with sanitary drinking and washing .water. If excreta
of all vertibrate hosts of the schistosomes are prevented
from having access to the water supply, the cycle of the
parasite would be quickly broken. That the disease is found
only sporadically in the larger cities with pipe-borne water
supplies is adequate evidence of the efficiency of this method
in eradicating schistosomiasis.
294. When Dr. Berry examined the Volta river in the
vicinity of the Ajena dam site, he expressed the opinion
that conditions near the banks, where vegetation was present
and the water relatively quiet, might be conducive to the
establishment of schistosomes in the region. As no snail
collections were made at that time, the presence of Physopsis,
Bulinus, etc. was not determined, although Dr. Berry felt
that the animals might be living there.
295. Following impoundment, marginal conditions in the
new lake will very likely favour the development of the
appropriate snail hosts and schistosomiasis transmission
could easily become more widespread than it is even at
present. Many of the labourers and villagers, now infected
with the disease, will undoubtedly urinate at the banks of
the lake or into the water itself, even though adequate
latrines may be provided. It appears that the spreading
of propaganda about the disease and its method of trans-
mission may be one of the most effective and cheapest means
of dealing with schistosomiasis.
296. The problem that will be created in the Accra
plains irrigation scheme will be great indeed. Irrigation
and spread of schistosomiasis are well known in Egypt, where
infections are now widely disseminated because of the great-
ly increased amount of water available for snails propogation.
Dr. Berry had the general impression that it was this part
of the Volta development scheme that would pose the greatest
297. It is essnetial that before irrigation begins, a
thorough survey be made of the villages of the Accra plains
to determine the incidence of schistosomiasis. As com-
pletion of the project is still several years in the future,
it might be most appropriate to begin the attempt to educate
the rural population to the simple means of breaking the
life cycle of the parasite. Motion pictures, accompanied
by elementary explanations, could probably do more good than
any other educative technique.
298. Dracunculus medinensis is not an insect-borne
parasite; however, I have been asked to include a statement
about the disease in the Gold Coast and its relationship to
the Volta river scheme.
299. Guinea worm is prevalent throughout the rural areas
of the Gold Coast wherever there is an inadequate or un-
sanitary supply of drinking water. It is most important to
a working force, as its widespread incidence and crippling
effects can incapacitate large sections of the community.
With modern techniques of drilling wells and supplying
satisfactory drinking water, there is scarcely any excuse
for the continued existence of a disease, the cycle of which
can be so easily broken.
300. To-day, in the Gold Coast, there are many small
villages which depend on open pools and drying mud holes
as their sole source of water. To obtain the water it is a
common practice for men, women, and children to wade into the
pool and dip up the water to fill containers. If these
persons are infected with mature guinea worms, the larvae are
released into the water where they are ingested by Cyclops,
a minute Crustacean, which is almost universally present.
Manson-Bahr (1945) points out that there is a regular rhythm
of the infection in the Gold Coast during the dry months
of January and February. At this time it seems that in-
fected Cyclops are not active, but lie at the bottom of the
wells. Since water is low in the wells and pools at this
time of the year, many more infected Cyclops are brought up
than when the wells are full. If the water is used for
drinking, the Cyclops are ingested with it, the acid in the
stomach of the host activates the larvae, and they then
escape from the Crustacean and invade the tissues of the
301. The obvious place to break the cycle, man--cyclops--
man, is at the weakest point, the infection of Cyclops.
This can easily be accomplished by supplying all villages
with wells with cemented sides and raised curbing to pre-
vent the people from washing or bathing in them and thus
contaminating the water. Wells should also be sealed
and provided with pumps whenever possible. Administrative
Officers in the Gold Coast have pointed out to me that this
has been tried, but the people soon break the pumps; there-
fore, they have discontinued their use. This breakage is
almost inevitable at first as the poeple who are using these
machines are entirely unaccustomed to such modern devices;
however, I believe that as the people learn the techniques of
using the pumps, the amount of breakage will soon lessen.
The initial costs of installing these simple devices will be
more than offset by the tremendous improvement in the over-
all health and well-being of the native population.
PART VI RECOMMENDATIONS ON THE VOLTPA RIVER PROJECT
videe Part II)
302. It will be necessary to provide protection against
disease-bearing insects for the personnel working on the
construction of the dam as well as for the future well-being
and protection of the inhabitants of the Volta river basin.
That the recommendations here submitted can orly form a
basis for a beginning of the improvement of the overall
health of the natives of the Gold Coast is abundantly
apparent; however, a start should be made and the tremend-
ous development which is to be undertaken in the Volta
river section can serve as a model for future work of this
sort in West Africa.
303. Specifically, I should like to recommend that the
following measures be taken to control or eradicate insect-
borne diseases in the Volta river region:
(1) In my opinion, a medical service should be set up
without delay to deal with the diseases of the
Volta river project. This health service should
contain a division dealing exclusively with insect-
borne diseases which, as is readily apparent from
this report, are major problems in the Gold Coast.
My first recommendation, therefore, would be to
set up a health service headed by a specialist in
tropical medicine. The staff should include a
malariologist and an entomologist experienced in
working with insects of medical importance. An
engineer, trained in malaria control of impounded
waters, is one of the very necessary persons for
the staff of the health service. Adequate
laboratories, equipment, and assistants must be
provided for this highly trained personnel who
can do much in the control and eradication of
many of the tropical diseases.
(2) Immediate protection of all non-African (non-
immune) workers by:
(a) Personal measures malaria suppressives or
prophylactics; repellants; protective
clothing after dark.
(b) Screened and properly mosquito-proofed
(c) Spraying of quarters with insecticides
having residual effects.
(5) Immediate recruitment of an experienced malaria
control staff. This staff should include:-
(a) A malariologist;
(b) A medical entomologist;
(c) An engineer experienced in malaria control
of impounded waters;
(d) Junior staff and assistants as recommended
by the malariologist.
(4) All phases in the planning of the reservoir area
and other construction should be closely co-
,prdinated with the planning of the malaria control
(5) The malaria control unit should be given all
necessary equipment, laboratory space, storage
space, docking facilities, etc. Equipment should
include necessary aircraft, boats, and land
(6) In co-operation with the malaria control team,
adequate measures should be taken to prepare the
reservoir basin so that a clean water surface and
a clean shore line will be present throughout as
much of the reservoir area as is feasible.
Marginal marshes, ponds, swamps, and depressions
should be drained. All steps should be taken to
prepare the reservoir in accordance with established
and proved practices.
(7) The malaria control team, headed by the malariologist,
should make a careful epidemiological study of the
malaria and other mosquito problems prior to in-
ception of the malaria control measures, and con-
tinue to make checks as the work progresses.
(8) The malariologist should establish liaison with
Dr. Leonard Chwatt, malariologist, Lagos, Nigeria,
who is determining whether residual spraying of
native villages will permit the partial or complete
omission of larviciding in malaria control programs
in West Africa.
(9) A botanist should be employed to make a survey of
the vegetation of the Volta River system to determine
those species of plants which may constitute a
problem in the reservoir. He will then be able to
recommend measures to be taken in eradicating the
(10) All borrow pits must be dug with their lowest point
above the 260 foot contour, and all of them should
be prepared so that water will rapidly drain away.
(11) All native villages which must be moved from the
future reservoir should be rebuilt in a safe area;
well-planned, well-built housing should be provided
for the inhabitants; and adequate, completely
sanitary, drinking water should be supplied. No
open pools, dug by the Africans for water, should
be permitted. The edges of all wells should have
concrete platforms, and be so built that spilled
water will drain away quickly. By no means should
villages be allowed to spring up in undesirable
areas. All housing in the villages should be built
according to specifications drawn up by a planning
council, and the houses should be of concrete
wherever possible to prevent the Africans from
digging pits to obtain building materials,
(12) The Volta river below the Ajena dam should be con-
sidered as part of the malaria control project and
should receive attention equivalent to that given
to the reservoir area.
(13) There should be an immediate assignment of an
entomologist to survey the breeding areas of
Simulium damnosum in the Volta river in detail;
to institute and.supervise eradication methods;
and to conduct studies of the biology of S. damnosum,
with particular emphasis on determining the methods
by which the species survives during periods of
(14) Arrangements should be made for the entomologist
to study Wanson's methods of eradication of Simulium
(15) All necessary supplies, including aircraft and
boats, which can also be used for mosquito control,
laboratories, and adequate assistance, as requested
by the entomologist, should be furnished.
(16) The epidemiological survey now being undertaken by
the Medical Research Institute, Accra, should be
encouraged and given every assistance so that a
very necessary part of the overall program will not
be side-tracked. After eradication of Simuliumm
continuing surveys should be conducted to determine
the effects of the cessation of transmission of
onchocerciasis on the population.
(17) The program of eradication should be started at the
earliest feasible date. Delays will only result
in more widespread infections.
(18) The newer insect repellents will deter the flies
from biting. Long trousers also make it imposs-
ible for the flies to bite the legs, their favourite
feeding areas. Admittedly, it will be difficult
to enforce its use, but as a preliminary measure it
is suggested that all Europeans in the area be .
issued repellent to be used against Simulium during
the day. The advantages of wearing long trousers
should be pointed out to all of them as well.
(19) A thorough survey of the present population in the
villages from Kpong to Misikrom on both banks of
the Volta should be undertaken to determine whether
humans infected with trypanosomiasis are present.
If any are discovered, treatment should be
instituted to rid them of the parasites.
(20) The medical officer in charge of the health service
of the Volta river project should require that all
persons coming into the area to work be examined
for possible trypanosomiasis infection. Anyone
showing evidence of the disease should be treated
before being allowed near the river.
(21) Migrant followers of the labourers wives,
children, etc., should not be permitted to come
within the construction zone where infected in-
dividuals might transmit their infection to
tsetse. The medical officer should undertake
to check as many of these migrants as possible
so that adequate measures can be taken to
(22) Check points should be introduced above Misikrom
and below Kpong for the examination of migrants
travelling through the protected area via the
river. Infected persons should be given
PART VII : RECOMMENDATIONS ON THE ACCRA PLAINS
(Vide Part III)
304. (1) Malaria control in the Accra plains must be
carefully co-ordinated with that of the Volta
(2) There should be a very careful planning of the
distribution of water and efficient supervision
of it when irrigation has started.
(3) Wide legal powers should be granted to the
authorities to permit them to enforce a proper
maintenance of the irrigation channels by the
(4) The engineering services, the supervisory
authorities, and malaria control organisation
must work constantly to avoid the malpractices
pointed out by Dr. Chwatt, and described in
this report in the section on malaria control
(5) If Simulium damnosum appears in the irrigation
drains the method used by Garnham and MoMahon
in Kenya of dripping D.D.T. emulsion on to the
water surface might readily be employed in the
control of damnosum in these confined waterways.
(6) After irrigation is started, farming the stream
valleys should be encouraged so that fringing
vegetation will be removed by the natives, thus
removing breeding areas for tsetse. Where there
is fringing vegetation suitable for the develop-
ment of tsetse or for harbouring adults, and the
land is not farmed, clearing should be under-
taken provided the land is utilized by humans
(7) Clearings for control of tsetse should be con-
ducted in such a way as to widen the gaps
between the riverine vegetation of the Volta,
where permanent populations of the fly persist,
and that of the small streams draining across
the Accra plains.
PART VIII : RECOMMENDATIONS ON CONTROL OF OTHER
*- INSECT-BORNE DISEASES.
(Vide Part IV)
305. (1) Measures taken to control malarial mosquitoes
will, at the same time, control proved vectors
(2) A survey to determine the amount of this disease
in the Volta river basin should be undertaken
along with surveys to determine the prevalence
of other parasitic diseases.
(3) At present there seems to be little reason to
undertake control of this disease; however, if
later research proves that loa loa (loasis) is
present and is more debilitating than now
believed, steps should be taken to control the
vector species of Chrysops by;
(a) spraying water in.which the larvae live
with D.D.T. in oil;
(b) clearing bush over streams in which
(c) clearing bush from around houses.
(4) There is believed to be little need to under-
take control of this parasite; however, the
incidence of infections with the worm is un-
known. Also the effects of the parasite on
the human host are not known with certainty.
If, at a later date, it is proved that this
worm, which is transmitted by Culicoides, is
affecting the human population, steps should
be taken in the control of the midges.
(5) Regular inspection of villages should be start-
ed in the Ajena region and in the smaller'towns
along the Volta to eliminate Aedes aegypti
breeding places in the villages themselves.
(6) As soon as feasible, countrywide immunization
of the population against yellow fever should be
undertaken by the Medical Department.
(7) Measures directed against Aedes aegypti in the
control of yellow fever are the same as those
used in the control of dengue.
PART IX : RECOMMENDATIONS ON SCHISTOSOMIASIS AND
(Vide Part V)
306. (1) I shall not give recommendations for the
control of Schistosomiasis as Dr. Elmer Berry
will soon undertake a comprehensive study of
the disease and its vectors in the Gold Coast.
His report will include appropriate
(2) Control of this disease, which is so important
in the Gold Coast, can be undertaken by
supplying all villages with sanitary wells
having cemented sides and raised curbing.
Wells should have simple pumps provided
whenever possible, and should be sealed.
(3) Educating the natives in the method of spread
of this disease would go far in reducing its
incidence. Simple motion pictures shown in
the villages will probably be the easiest way
of approaching the problem,
1r-vJ $-"~"*0~ ~ '
/c 4 / 3,7).- /r / /-
Annual Reports of the Medical Department, Gold Coast.
Atlas of the Gold Coast. 1949. Gold Coast Survey Depart-
Belding, D. L. 1942. Textbook of Clinical Parasitology.
A. Appleton-Century Co., New York.
Bird,C.St.John.1949. The Development of the Volta River.
Buxton, P. A. 1948. The problem of tsetse flies (Glossina.)
Proc. Fourth Inter. Cong. Trop. Med. Malaria.
Vol. 2; 1630-1637.
Carter, H.F., A. Ingram, and J. W. S. MacFie. 1921.
Observations on the Ceratopogonine midges
of the Gold Coast with descriptions of new
species. Ann. Trop. Med. Parasit.
14:187-274. 27 figs., p1s. IV-VI.
Chwatt, L. J. 1949. Memorandum on malaria incidental to
irrigation projects in West Africa, being a
comment on the British West African Rice
Mission's report 1948. Mimeo.
Chwatt, L. J. 1950. Malaria in Nigeria. Summary of basic
Davey, T. H. 1948. Trypanosomiasis in British West Africa.
Colonial Office, London.
Demeillon, B. 1947. The Anophelini of the Ethiopian
Geographic Region. Publ. S. African In-
stitute for Med. Res. Johannesburg. No.XLIX.
Edwards, F.W. 1941. Mosquitoes of the Ethiopian Region.
Part III. Br. Mus. (Nat. Hist.). London.
Evans, A. M. 1938. Mosquitoes of the Ethiopian Region.
Part II. Br. Mus. (Nat. Hist.). London.
Garnham,P.C.G. and J. P. McMahon, 1946. The eradication
of Simulium neavi, Roubaud from an
onchocerciasis area in Kenya colony.
Bull.-ent. Res. 37: 619-627, 2 figs., 6 tbls.
Gordon, R,M., L.J. Chwatt, and C. M. Jones. 1948. The
results of a preliminary entomological survey
oof loasis at Kumba, British Cameroons,
-together with a description of the breeding-
places of the vector and suggestions for
future research and possible methods of con-
trol. Ann. Trop. Med. Parasit. 42(4):
364-376, pl. XVI.
Halcrow, Sir William, and Partners. 1950. Preliminary
report on development of the River Volta
Herms, W. B., and H. F. Gray. 1944. Mosquito Control.
2nd ed. The Commonwealth Fund, New York.
Hocking, K. S. and D. G. Maclnnes. 1948. Notes on the
bionomics of Anopheles gambiae in East Africa.
Bull. Ent. Res. 39: 453-465, 5 figs.
Hopkins, G,H.E. 1936. Mosquitoes of the Ethiopian Region.
Part I. Br. Mus. (Nat. Hist.)., London.
Hughes, M. H.
Junner, N. R.
Kitson, A. E.
Lewis, D. J.
1949. African Onchocerciasis. Ms. (in library
of Medical Research Institute, Accra.)
and T. Hirst. 1946. The Geology and Hydrology
of the Voltaian Basin. Geological survey,
Gold Coast Colony, Memoir No. 8.
1925. Outlines of the Mineral and Water-Power
Resources of the Gold Coast, British West
Africa. with Hints on Prospecting. Gold Coast
Geological Survey, Bull, No.1.
1948. The mosquitoes of the Jebel Auliya
reservoir on the White Nile, Bull. Ent. Res.
39; 135-157, 3 figs., pl. XII.
1948. The Simuliidae of the Anglo-Egyptian
Sudan. Trans. Royal Ent. Soc. Lond. 99;
475-496, 1 p1., 14 maps.
Manson-Bahr,P.H. 1945. Manson's Tropical Diseases. Cassell &
Co. Ltd., London.
1946. The control of trypanosomiasis by
entomological means. Bull. Ent. Res.
37: 201-250, 5 figs., 6 tbis., maps 1-3,
1949. Planning the control of sleeping sick-
ness. Trans. Roy. Soc. Trop. Med. Hyg. 43 (2):
165-198, 3 maps, pis. I-II.
_and M.G. Morris. 1949. The use of traps
against tsetse in West Africa. Bull. Ent,Res.
40; 491-528, 5 figs., 3 tbls, p1s. XXIII-XXIV.
Muirhead Thompson, R.C. 1947. Recent knowledge about malaria
vectors in West Africa and their control,
Trans. Roy. Soc. Trop. Med. Hyg. 40 (4):
1948. Tsetse Flies in British West Africa.
Colonial Office, London.
Pomeroy,A.W.J. 1929. Preliminary report on tsetse fly in-
cidence in the livestock zones of the Eastern
Province. Med. Res. Inst., Accra. Report No.
1930. Tsetse survey, Accra plains. Med. Res.
Inst. Accra. Report No. 40/1929-30.
1949., Note preliminaire sur 1'epidemiologie
et le traitment de l'onchocercose humaine a
0. volvulus en pays Mossi. Bull. Soc. Path.
Exot. 42(11.12): 558-561. (Abst. Trop. Dis.
Bull. 47 35); 484, 1950).
L.S. West, and R.D. Manwell, 1946. Practical
Malariology. W.B. Saunders Co., Philadelphia.
1948. A Handbook for the Identification of
Insects of Medical Importance, 2nd ed. Br. Mus.
(Nat. Hist.,) London.
D.B. Wilson, S. Lima, and W.S. Atunes. 1943.
The Org4nization of Permanent Nation Wide Anti-
Aedes Aegypti Measures in Brazil. Rockefeller
Foundation, New York.
Swynnerton, C.F.M. 1936. The Tsetse Flies of East Africa.
Trans. Roy. Ent. Soc. Lond. Vol. 84.
and Max Theiler. 1948. The epidemiology of
yellow fever. Proc. Fourth Inter. Cong. Trop.
Med, and'Malaria. Vol. 1. 506-519.
Tennessee Valley Authority, 1947. Malaria Control On Impounded
Water, Superintendent of Documents, Washington.
Waddy, B. B.
Army, 1944. Filariasis, epidemiology and
Control. Malaria and Epidemic Disease Control
Training Manual No,6. Hdq., South Pacific Area.
V.D. and J. McMahon. 1950. Phoretic association
between Afronurus and Simulium species, and the
discovery of the early stages of Simulium neavi
on freshwater crabs. Nature. 166 (4217):
350-351, 1 fig.
1949. An investigation of blindness and
onchocerciasis in the Northern Territories of
the Gold Coast. Mimeo. (files of Gold Coast
1948. Nota sobre biologia y control de los
Simulidos. Proc. Fourth Inter. Cong. Trop.
Med. and Malaria. Vol. II: 1619-1629.
L. Courtois, and B. Lebied. 1949. L'eradication
du Simulium damnosum (Theobald) a Leopoldville.
Ann. Soc. Med. Trop. 29 (3): 373-403, 2 tbls.,
and C. Henrad. 1945. Habitat et comportement
larvaire du Simulium damnosum Theobald. Rec.
Trav. Sci. Med. Congo Belge. No. 4:113-121.
and J.F. Corson, 1946. A survey of recent work
on Trypanosomiasis and tsetse flies. Bur. Hyg.
Trop. Dis. Rev. Mono. No.1, pp. 1-81.
and E.B. Cram. 1948. Wartime research in human
schistosomiasis. Proc. Fourth Inter. Cong. Trop.
Med. and Malaria. Vol. II: 997-1012.
of the Volta river as it passes through the Ajena gorge.
2. A small part of the rock outcrop at the Kpong rapids. The
photograph was taken during the rising stage of the river
level in early July.
3. A closer photograph showing some of the rocks at the Kpong
rapids. Larvae and pupae of Simulium damnosum were found
living in the swiftest parts of the rapids.
4. The Volta river at Dodi in August. Marginal vegetation was
partially submerged as the river level rose.
5. The Volta at Yeji. The river was at peak flood when the
photograph was taken in early October. Note the quiet water
near shore. Conditions such as these are not conducive to
6. Cattle hoofprints at the edge of the Volta. The depressions,
filled with water, make excellent breeding places for
7. Children catching adult Simulium damnosum on their legs.
8. The Afram river at Mankrong. Note the heavy fringing forest
along the banks of the small stream. Tsetse (Glossina palpalis)
were very common here. The picture was taken in mid-September.
9. The Afram river at Achi in September.
10. The Asuboni at Mpraeso. Larvae and pupae of Simuliuim aleooi
and S. unicornutum were numerous on the submerged vegetation
at the banks.
11. Woodland savannah near Ejura.
12. The vehicle used for much of my travel in the Gold Coast.
I-~- -----a~--I r
" ACCRA ANTI-MALARIA DRAINAGE SCHEME
Sj 'LAYOUT AND SIZES OF WATERWAYS.
(;01.1) C ANrT x lRVIcY
, :..; r i.D: -:ION HOSPITAL LABADI
ritn:,M. 'iA ' MATEEMHED WATERSHED
AC I -IIA ANA) EA-NVIRO 'Ns
4CATEH EVENT AREA
a ULI--. S
KPESHI E LAGOON
LAI. IL AILRSLHE[
t --- I-
i ACCRA AND ENVIRONS
GOLD C A.0C*7 'R'Lf
ALL LEVELS ARE BASED ON THE KAWLI LAGOON SURVEY DATUM
99-15 FEET BELOW A.T.S. DATUM
*85 FEET ABOVE KLOTEY LAGOON SURVEY DATUM
LINAGE SCHEME KLOTEY LAGOON. 4 .R KAL. MDIA
MAJOR R MAJOR AMD/
LAYOUT PLAN. MALARIA ENG NEI AREA MALARLOGIST za