Proc. 3rd int. Conf. Equine Infectious Diseases,...
Proc. 3rd int. Conf. Equine Infectious Diseases, Paris 1972, pp. 12-30
(Karger, Base1 1973)
Ecology of African Horsesickness
P. BOURDIN
1. E. M. V. T. Laboratoire de Recherches Vétérinaires, Dakar
I. Introduction
Ecologie study of animal disease is defined as the study of interactions
between environment and the respective populations in a medical sense.
The so-defined ecology is related by SCHWABE [53] to epidemiology, but
he says ‘an epidemiology seen by an ecologist or a biologist who forgets his
inborn anthropocentrism to sacrifice himself to holistic discipline’. The
ecologist, be he physician or veterinarian, has to take an interest not only
in the mode of disease transmission or restricted epidemiology, but also in
comparative pathology, immunology, vertebrate and invertebrate zoology,
climatology, agronomy, geography and even history.
SCHWABE [53 ] has developed a 3-stage working method for ecologists :
First stage. The research worker registers observations, descriptive
stage.
Second stage. He compares and analyzes his observations, analytic
stage.
Third stage. He verifies hypotheses developed during the analytic stage
by experimental work, experimental stage.
The ecology of African horsesickness (AHS) is related to arbovirus
ecology, which has led to the classification of AHS virus within this
group. OLLERMANN et al. [35] classify AHS virus together with Bluetongue
virus in the subgroup of diplornaviruses. AHS ecology resembles arbovirus
ecology and suggests the presence of a vector belonging to blood-sucking
arthropods and the existence of a host. TO date, experimental proof of
vectorial transmission according to the World Health Organization criteria
[5 ] has not been established and the existence of a natural host or reservoir
BOURDIN
13
is hypothetical. But if the identification of one or more reservoirs is of no
consequence for endemically infected countries [27], it is of primary impor-
tance for endangered countries to prevent the introduction and distribution of
the virus. We Will enumerate in this article the most important points known
and we Will then deal with three as yet unanswered questions: appearance of
an epizootic, existence of a vector and of a reservoir. These three questions
are intimately linked, the ecology being the whole. In a future research
program it Will be indispensable to clarify first the problem of the reservoir;
next that of the vector(s). Definition of the epizootiology of AHS depends
directly on solution of these problems.
II. Analysis of Facts and Experimental Work Concerning Ecology
In this section discussion Will be restricted to facts and the most useful
experimental work relating to ecologic study. The reviews of CURASSON
[12] and HENNING [20], as well as the monographs of RAFYI [48], HOWELL
[21], MORNET and GILBERT [30], STELLMANN et ul. [56, 571, give a much
more detailed account.
A. Study of Facts
1. Geographic Distribution in the Past and Present
The cradle of the disease lies in Africa, more precisely in the countries
on each side of the equator in the zones of dry tropical climate, Le. Sahelo-
Sudanese climate between the isohyetes from 250 to 1,000 mm. These zones
may be extended by altitude [6] as is evident from figure 1. South of the
equator the zones include Mozambique, Rhodesia, Zambia, Tanzania,
Kenya, Angola, Transvaal, Natal and part of the Orange Free State as far as
South Africa is concerned. North of the equator there is no marked boundary
and the respective countries lie between the 10th and 12th parallel of latitude
with a ventriflexion of the band from west to east and a dorsiflexion near the
Ethiopian mountains. This band includes Senegal, the south of Mali and of
Niger, Upper Volta, the south of Chad, Sudan up to Khartoum, part of
Ethiopia and of Eritrea. This zone has for centuries primarily been agro-
pastoral or sylvo-pastoral and the use of horses as draught or prestige ani-
mals is limited in the north by the desert and in the south by the tsetse fly
[251.
14
BOURDIN
Fig. 1. African countries with Sahelo-Sudanese climate and its variants.
From time to time AHS breaks out of its original cradle and extends
into neighboring countries. This extension reaches the Cape region in South
Africa, where the disease has been known since the 17th Century and recurs
about every 20 years. North of the equator the spread of AHS is much less
regular. In 1930-1931 AHS reached Egypt. In 1933-1934 it was again
observed in Egypt, Palestine and the Lebanon. In 1959, Iran was reached
and the disease spread east from there, extending into Afghanistan, Pakistan
and India (1960); to the west it was recognized in Iraq, Jordania, Syria,
Turkey, the Lebanon and Cyprus.
In 1965, AHS conquered the Saharian barrier and invaded the south
of Morocco and Algeria. After being dormant during winter it wandered
north into Tunisia and reached Spain in October 1966.
Since then the disease has apparently receded into its originating African
zone to prepare for a new eruption.
Ecology of African Horsesickness
15
2. Spread of AHS
AHS occurs endemically and periodically in the form of epidemics.
The endemic form is habitually seen in the countries of western and central
Africa with Sahelian climate favoring the breeding of horses. In this zone
some kind of ecologic equilibrium seems to exist between vector, reservoir
and the 1.5 million horses [4].
The home was introduced into Africa south of the Sahara about 2,000
B.C.: ‘The inventory of Saharian rock dwellings shows that between this
time and the start of our era the Sahara was crossed by numerous routes on
two axes, one going from Lybia to Gao, the other from South Morocco to
Mauritania and Tombuctu’ [25]. This stage was followed by trading with
horses between the Salt-producing countries of the Maghreb and the king-
doms of western Africa rich in gold mines. Commerce in horses was very
active up to the Middle Ages.
It is probable that these animals have acquired progressively a natural
resistance against AHS, a feature of the local breeds of today. That there
must have been contact with the virus has been proved by serological sur-
veillance studies carried out in nonvaccinated animais by MAURICE and
PROVOST in Central Africa [26] and BOURDIN et al. in Senegal and Mali
181. The ecologic equilibrium in these countries is upset by the importation
of susceptible horses. Between 1882 and 1925, animais imported from France
or Morocco by the military authorities were regularly decimated until a
decision was taken to use only native horses in the mounted units. Since then
AHS has become rare; it was observed occasionally when horses were im-
ported for riding clubs or studs. MORNET [29] has seen it in Senegal, DOUTRE
and LECLERC identified it in Chad [13] and BOURDIN [9] again in Senegal.
GILBERT [19] has observed the disease in animals sired by imported stallions
out of native mares.
The epidemic form seen within the last decade in the Middle East, Asia
and the Maghreb has been known in South Africa since the 17th Century in
the Cape region, when horses were first introduced there. In the more sep-
tentrional (northern) provinces of the South African Union (Natal, Trans-
vaal) the horses, the majority being vaccinated, are very probably in close
contact with the wild virus [17]. One cari presume that there has been estab-
lished in these regions some sort of artificial equilibrium.
The spread of AHS virus in new countries seems to be linked, according
to HOWELL [21, 221, to the introduction of equidae or contaminated insect
vectors, which find a zone with favorable ecologic conditions. He applies
BOURDIN
16
this explanation to the appearance of AHS during the summer of 1959 in the
coastal regions of Iran. After a winter pause the disease spread in 1960 along
the rivers, the traditional nomadic routes and the major road axes.
For PILO-MORON et al. [45], AHS arrived in the south of Morocco and
Algeria following the habitua1 routes (fig. 2). It isestimatedthat40,000-50,000
nomads still live in south Algeria and that these people migrate between
southern Morocco and Algeria and the Sahelian regions of Mali and Niger.
These people migrate with herds consisting of dromedaries, donkeys, sheep,
goats and a few dogs. The hypothesis of spread along these migratory routes
is based on the presence of neutralizing and complement-fixing antibodies
in donkeys living in the oases through which the routes lead. These animals
could perhaps serve as markers or relays.
3. Climate and Topography
Al1 authors stress the favorable influence of high humidity always
following abundant rains and heat. Cold and drought are, on the contrary,
unfavorable factors.
i
I
N igw
’ Chad
Sudîn
Y’
r
‘.!
‘7
ATLANTIC O C E A N
. ,
-;--,--J
L.
;
;.-a , ,
.i
L
Fig. 2. Traditional migratory routes used by nomadic populations.
Ecology of African Horsesickness
17
In the south of Africa, where the rainy season starts in December, the
most dangerous months are February, March and sometimes April. In
countries with a Sahelo-Sudanian climate the rain season lasts for 2-4
months with a maximum in August; it is followed by a rigorous dry season
lasting 6-8 months. The vapor tension may attain 18-22 mm between the
middle of June and the middle of November [6]. The months of October
and November, in particular, favor the eruption of AHS.
Countries with a continental or Mediterranean climate show a distinct
climatic incidence. RAFYI [48] noted in Iran that the disease spread in
summer 1959, stopped after the first frosts and reappeared in the following
year. In Algeria [54] the disease is widespread in October, regresses with the
cold and is reactivated the following season. In Morocco, AHS reached
alarming proportions at the end of the winter of 1965 to become extinguish-
ed in the summer of 1966 [24]. This writer mentioned thatspringandsummer,
1966, were relatively dry but were preceded by a very rainy autumn in 1965,
which covered the valleys with an abundant flora.
Topographically, the zones favoring the spread of AHS are the low,
swampy regions, the humid valleys with abundant flora [20], water courses,
Wells, puddles, mudholes and bore-holes [24, 48, 49, 541. In mountainous
regions, AHS is restricted to the lower zones. THEILER [59] observed AHS
in South Africa usually below 500 m; under exceptional climatic conditions
it was seen up to 1,200 m. In Ethiopia and Eritrea the disease does not occur
at altitudes higher than 1,200-1,500 m [47]. In Kenya, closer to the equator,
PIRANI [46] found a focus at 2,400 m altitude.
4. Susceptibility of Animais
The horse is extremely susceptible, with a morbidity of more than 95%
in epizootics and a mortality of 85% [48]. The mule is less susceptible with a
50-percent morbidity during epizootics.
The South African donkey (Equus asinus somalicus) is resistant [61].
Fatal cases have been seen in Egypt and Algeria in donkeys of the subspecies
Equus asinus africanus [2, 541. ORUE [37] observed a very high mortality
among the donkeys of the Cape Vert region. The zebra (Equus burchelli)
may show a clinically evident form of disease in South Africa [22].
The dog is infected by ingestion [62]. PIERCY [43] has described enzoo-
tics in animais of packs nourished with contaminated horse-meat.
The Angora goat, not susceptible under natural conditions, produces a
febrile response to virus inoculation. The blood of such animals is virulent
[62, 641.
BOURDIN
1 8
5. Mode of Transmission
Direct transmission is exceptional and, if it occurs, it is by accident only.
It is possible to keep healthy animals together with sick ones in an insect-
free stable protected from blood-sucking arthropods. The massive use of
insecticides during epizootics stopped the spread of virus in Spain in 1966.
These two facts and the influence of climate support the theory of insect-
borne transmission.
B. Analysis of Experimental Work
1. Resistance of AHS Virus to External Factors
The virus is stable outside the body. Putrefaction, dessication and tem-
perature variation exert little effect. Only pH values below 6 are detrimen-
tal. The ideal pH for conservation varies from 6.5 to 8 [l, 381.
Wild viscerotropic strains are stable at 4°C in OCG mixture [63];
neurotropic strains are stable in the presence of serum for 90 days at this
temperature [l, 281. At 37°C in the presence of calf serum the ce11 culture
adapted neurotropic virus loses 6 logs within 40 days and only 2 logs at
-20 to -25°C [OOO].
2. Persistence of Virus in Susceptible Animais
The horse experiences a short viremia; the virus cari be found in the
blood at the beginning of the febrile reaction and disappears fairly rapidly.
After the ninth day, isolation from blood is only rarely achieved, except for
the case described by THEILER [63] in which the disease was transmitted by
massive inoculation of a healthy horse with blood from a horse that was sick
90 days previously. Viremia is difficult to detect in donkeys.
MCINTOSH [27] could not isolate the virus from vaccinated horses
responding to challenge. After intracardiac inoculation of ferrets, however,
isolation was achieved from the blood of the ferret. The virus was pathogenic
for suckling mice. The author thinks that the virus masked by antibodies
in the horse was liberated in the ferret.
3. Study of Vectors: Virus in Blood-Sucking Insects and Transmission
Experiments
In many articles on AHS a number of arthropods are cited as possible
vectors. JOUBERT [23] compiled the following list :
Ecology of African Horsesickness
19
Culicidae:
Culex pipiens, Ochlerotatus, Aëdes caballi, A. aegypti,
Anopheles, S’tegomya
Simulidae :
Simulium
Tabanidae :
Tabanus pluto
Muscidae :
Stomoxys calcitrarw, Hyperosia minuta
Ceratopogonidae:
Culicoides pallidipennis
Ixodides
But, according to this worker, many of these insects are with certainty
‘simply winged needles’ and, in fact, successful experimental transmissions
cari be counted on the fingers of one hand. Stomoxys calcitrans cari mechani-
cally transmit AHS virus [52]. Mosquitoes may exceptionally transmit the
virus [34]. These authors found a predominance of A. caballi with some
Culex and Anopheles among the mosquitoes living at Onderstepoort. They
tried to transmit the disease to susceptible horses, either by inoculation with
crushed mosquitoes that had fed on viremic horses 12 h to 55 days before,
or by letting infected mosquitoes feed on susceptible horses. After 66 ex-
periments with Aëdes they obtained 4 positive results, a11 of them by inocula-
tion. In one case the infective meal took place 12 h previously, in the three
other cases 7 days before. Anopheles stephensi and C. pipiens infected horses
when they were fed virus in vitro and put on horses 15-22 days after the meal
[40, 411. The horses died late, 25-38 days after having been stung. OZAWA
et al. [42] continued their experiments using A. aegypti, fed in vitro with a
suspension of wild type 9 virus adapted to ce11 cultures (titer lOs.aID,,/O.l
ml). The presence of virus in the mosquitoes was checked from 1 to 36 days;
on the 36th day virus titers in certain lots of mosquitoes were still as high as
102-s IDse/O.l ml. The authors explain the titer as being due to virus replica-
tion in the mosquito. AHS was also transmitted to a horse by A. aegypti
which had been infected in vitro 19 days earlier. The horse died 18 days after
contact with the mosquitoes.
Du TOIT [14] captured Culicoides sp. in an AHS region and tried with-
out success to transmit the disease to a horse by its inoculation with these
crushed arthropods. In 1945, Culicoides were fed on a viremic home and 12
days later were put on a susceptible horse, which died of AHS [15].
WETZEL et al. [67] fed A. aegypti and C. pipiens fatigans from the
Onderstepoort collection and Culicoides sp. captured in light-traps on a
virus suspension and on a viremic horse. Al1 attempts to isolate virus from
crushed arthropods in suckling mice failed l-40 days after the infective
meal; the same Culicoides were not able to infect susceptible horses.
BOURDIN
20
4. Search for and Identification of the Reservoir
For the determination of a reservoir two complementary techniques cari
be used: intracerebral inoculation of suckling mice with material obtained
from different vertebrates and a systemic serologic investigation of many
species.
a) Serologic Investigations
Species suspected of harboring virus are selected and a large number of
samples are tested. The complement fixation test indicates a contact within
the last 6 months [22] and the reaction is group-specific. Serum neutraliza-
tion detects more temporarily distant contacts and is type-specific. TO our
knowledge, no systematic large-scale investigations have been carried out.
PILO-MORON et al. [44] found neutralizing antibodies in 69% of 84
donkeys living in south Algerian oases.
In dogs living in an AHS focus, 1 in 13 samples contained antibodies
1271.SHAH [55] was able to detect antibodies in samples from various species.
10 of 20 canine samples were positive, 8 of 42 donkeys and 1 of 42 cattle
had antibodies in their serum. No antibodies were found in 34 human, 17
caprine, 21 avian and 2 ovine serum samples.
b) Inoculation of Suckling Mice or other Susceptible Species
Experimental inoculation with material from species other than equidae
has only been done on a small scale. The results have always proven negative.
THEILER [60] and BEVAN [7] were able to isolate virus from experi-
mentally infected dogs. The same was possible in Angora goats [64] and in
sheep [65], but THEILER’S attempts [61] to isolate a virus from wild animals,
birds and amphibians were unsuccessful.
The difficulties to isolate AHS virus in nature from reservoirs or vectors
are comparable to those encountered by WHO for west African arboviruses.
Two centers of WHO, the Pasteur Institute in Dakar and the University of
Ibadan, are systematically looking for arboviruses in man, domestic animals,
wild animals and arthropods by intracerebral inoculation of 1 day-old
suckling mice. In case of a positive result after 2-3 passages the sensitivity of
the virus to ether and chloroform is checked and identification of the agent
is pursued.
In Senegal, the material originates mainly from two centers, one 80 km
from Dakar on the road to M’Bour in the forest of Bandia, the other near
the Gambian border in Saboya. There are only a few horses near these centers
Ecology of African Horsesickness
21
because of the tsetse fly. The reports by ROBIN and BRES [50], ROBIN and
LE GONIDEC [51] for the virological study and by :~AUFFLIEB et al. [58] for
the entomological investigations cari be summarized in two tables. In these,
the samples are classified in large groups according to their origin: man,
wild fauna, mosquitoes, ticks, other arthropods. The group ‘wild animals’
includes mammalians (insectivores, chiroptera, primates, rodents, carni-
vores, artiodactyles) and ranges from the excessively numerous birds to
reptiles. Liver, spleen and brain are routinely used to prepare inoculums.
Mosquitoes and ticks are identified, classified and used for inoculation in
lots. The group ‘other arthropods’ includes 162 lots of culicoides, 11 lots of
tabanids and 24 lots of phlebotomes.
The total inoculations in 8 years amounted to 9,788 with 224 viruses
isolated, only two of which were ether-resistant both being identified as
Coxsackie virus (tables 1, II).
In Nigeria, the WHO arbovirologists work at Ibadan, a town situated
in a humid tropical climate with very few horses. Investigations are also
Table Z. Number of inoculations with material from different species (extracted from
WHO reports)
Origin
1964 1965 1966 1967 1968 1969 1970 Total
M a n
113
552
284
1 9
54
332
162
1,516
Wild animals
1 7 1
297
529
291
448
241
1,230
3,207
Mosquitoesl
5
113
145
488
327
483
1,171
2,732,
Ticks
1
80
74
256
439
106
1,016
1,974
Other arthropods2
-
-
-
-
-
154
205
359
1 Number of lots.
2 The group consists of the following lots: Culicoides, 297, tabanids, 14, phlebotomes,
48.
Table ZZ. Identified viruses (extracted from WHO reports)
Origin
1968
1969
1970
Total
M a n
39
1 6
2
57
Wild animals
46
-
10
56
Mosquitoes
8 1
10
9 1
Ticks
1 2
-
16
28
Other arthropods
2
0
0
2
BOURDIN
22
carried out in north Nigeria, a zone of Sahelo-Sudanian climate. Other
investigations take place in Chad, Dahomey, Togo and the north Cameroons.
The yearly reports (1964-1969) published by the University of Ibadan in the
.
framework of the Arbovirus Research Project cari be summarized in two
tables. Table III concerns the inoculations per year and the zoological groups.
Table IV gives the number of viruses isolated from and identified in these
groups. In this table each column contains two series of numbers: those to
the left for arboviruses, those to the right for other viruses.
The samples from domestic animals are blood samples from animals
from the central and northern provinces and slaughtered at Ibadan (bovine,
Table III. Yearly inoculations performed at Ibadan
Origin
1964
1965
1966
1967
1968
1969
Total
M a n
1,815
662
2,212
1,793
2,560
2,115
11,157
Domestic animal+
220
1,568
1,295
619
330
1,048
5,140
Indicator animals’
207
263
273
709
172
1,624
Wild faunaa
438
1,279
3,429
2,794
2,723
10,663
Ticks
1,209
1,328
579
1,436
3,093
180
7,825
Mosquitoes
96
140
193
280
714
1,423
Culicoides
13
-
399
830
325
1,567
Other arthropods
13 83
63
68
10
297
1 Animals coming from central and northern Nigeria and slaughtered at Ibadan.
2 Indicator animals consist of suckling mice, chicken, rhesus monkeys and calves.
3 Mammalians, birds and reptiles captured in Nigeria, Chad and Dahomey.
Table IV. Virus isolated by near and species
Origin
1964
1965
1966
1967
1968
1969
Total
A NA A NA A NA A NA A NA A NA A NA
M a n
8 79
1 5
20 31
21 12
41 23 104 6 195 156
Domesticanimals
14 -
160 2
45 13
13 14
26 19
20 1 278 49
Indicator animals -
-
4 -
2 -
5 -
8 10
-
-
21 10
Wild fauna
-
-
1 -
7 3
16 1
25 10
1 2
50 16
Ticks
179 -
152 -
66 5
182 -
191 7
25 - 795 12
Mosquitoes
-
-
-
-
- -
1 -
1 -
19 -
21 -
Culicoides sp.
-
-
-
-
- -
44 -
3 7
1 2
48 9
Other arthropods -
-
-
-
- -
2 -
1 -
- -
3 -
A = arbovirus, NA = nonarbovirus.
Ecology of African Horsesickness
23
ovine, caprine and porcine samples). The ‘sentine1 animals’ or indicator
animals (suckling mice, rhesus monkeys, chickens and calves) are exposed in
the park of the University of Ibadan or on a nearby farm. The number of
samples taken for inoculation amounts to 39,696 with 1,613 viruses isolated
and identified in 6 years, 1,361 of these were arboviruses and 252 other viruses
(herpes, myxovirus, picornavirus, poxvirus, rhabdovirus, reovirus). Among
the ‘other viruses’, 9 bluetongue viruses were isolated, 8 from Culicoides sp.
and one from Crocidura SP., both captured near Ibadan.
III. Discussion
AHS is an endemic, essentially African disease occurring on both sides
i of the equator in dry, tropical climates, such as the mode1 Sahelo-Sudanian
’
climate with its variants. The Sahara constitutes the northern limit of the
endemic zone
The disease appears in seasonal rhythm and is favored by heat aud
humidity. The horse seems to be an accidental host. In western Africa the
horse has acquired a very solid natural immunity due to its very ancient
cohabitation with the virus. Periodically, the disease breaks out of its habit-
ual geographic area and reaches countries with a more temperate climate.
This occurs generally during the hot season in those countries. It Will dis-
appear in winter and cari sometimes recur during the following summer to
become extinguished after the first frosts.
Arboviruses are transmitted in different cycles [lO]. The same diversity
of transmission must be expected to apply to AHS. Accordingly, we cari
distinguish a basic sylvatic cycle in the absence of the horse: infected verte-
brate -f arthropod -+ susceptible vertebrate.
In the proximity of such a cycle an equine population may exist and,
depending on the frequency of contacts, the result Will be a zone of weaker or
stronger endemicity. A zone of strong endemicity is characterized by frequent
contacts with susceptible populations, horses in our case, the contacts being
old in adults, recent in Young animals. The arbovirologists designate the
form of evolution as ‘rural’, following, in the case of AHS, this scheme:
infected vertebrate -f arthropod -f horse. Whether the horse is a dead-end
is a question that remains unanswered [lO].
The infected horse may be introduced into a zone with susceptible
populations at a time when blood-sucking arthropods abound. The arthro-
pods may be true biological vectors or ‘simple winged needles’; in any case,
BOURDIN
24
ANS Will become epidemic and the new cycle is as follows : infected horse -+
arthropod -+ susceptible horse.
In this zone the ecologic conditions may differ from the zone where the
basic cycle takes place. If there is no reservoir, the greater variations of clima-
tic conditions Will interrupt the reproduction of vectors. In certain vectors,
however, the phenomenon of the diapause may give rise to a reappearance
of the virus as soon as the temperature becomes milder. Disappearance of the
disease Will then take place during the following winter.
This explanation seems to apply to the development of AHS in Iran and
its spread to the east and the west. HOWELL [21] feels that the first focus was
due to the introduction of infected horses from East Africa. The introduc-
tion of the virus into North Africa is more difficult to explain. In Algeria
the first foci were observed in locations on the northern border of the Sahara
(Bechar, Tadjmout, Laghouat) at the end of the routes used by nomads [45].
A serologic investigation of Saharian donkeys living close to the caravan
routes reveals the presence of neutralizing and complement fixing antibodies
at high titers. The trans-Saharian routes are used by nomadic populations
amounting to about 50,000 people [66]. These nomads migrate, following
the watering places of the northern Sahara toward the nearest Sahelian
regions of Mali and Niger, where they find pastures for their herds consisting
mainly of dromedaries, sheep, goats and donkeys. According to PILO-
M ORON et al. [44], the donkeys (Equus asinus africanus) undergo an in-
apparent infection and serve as relays between Sahelian and Algerian horses.
This hypothesis remains to be confirmed by virus isolation.
Ecologie circumstances point to the transmission of AHS by blood-
sucking arthropods, but experimental evidence is contradictory. Two groups
of vectors, mosquitoes and Culicoides, are mainly suspect. Some workers
have obtained positive results with mosquitoes, negative with Culicoides;
other have found the exact contrary. It is important to realize that the isola-
tion of a virus from arthropods does not a priori imply a biological trans-
mission. Two minimal conditions must be fulfilled, namely replication in the
arthropod and transmission by the sting. W ETZEL et al. [67] apply these
requirements to AHS virus and plan to inoculate arthropods intrathoraci-
cally to verify if replication takes place. There is no profit in discussing the
advantages and inconveniences of this artificial experiment; we shall con-
sider only the ecological reasons and uncover the difficulties in identifying
the vector.
The first difficulty is of a statistical nature. In 1944, Du TOIT isolated
AHS virus once from Culicoides sp. and achieved transmission in 1945 [14,
Ecology of African Horsesickness
2s
15,671. In 1944, the same author also isolated bluetongue virus three times
from Culicoides sp. and achieved transmission by C. pallidipennis. In the
USA, FOSTER et al. [18] transmitted bluetongue with Culicoides variipennis.
In 1971, NEVILL regularly isolated bluetongue from Culicoides sp. in South
Africa during the warm season, provided large inoculums were given. Why
this difference in spite of the close resemblance of the two viruses ? Simply
because the reservoir of bluetongue is represented by 12.8 million cattle and
the host by 39 million sheep, whereas AI-KS concerns only 380,000 horses,
the only vertebrate permitting isolation in South. Africa, as long as the
reservoir is not identified.
The second difficulty is due to the Culicoides themselves; their identifi-
cation is difficult and their biology incompletely defined. Some of them live
in close contact with cattle, C. pallidipennis in particular, which lays eggs and
reproduces in fresh cattle feces [32]. This Culicoides represents 97% of a11
those identified in Africa [l 1, 31, 321. Moreover, Culicoides do not depend
on a blood meal for their first oviposition. Within a lot of Culicoides captured
in light-traps one cari distinguish the insect having completed a gonotrophic
cycle from the immature, but one cannot enumerate the cycles. The devia-
tion of life and the frequency of contacts with the host are equally poorly
defined.
The biology of mosquitoes is better known. In the case of AHS, un-
fortunately, the virus has never been isolated in the field, even though experi-
mental transmission has been successful with certain species. Mosquitoes
are, therefore, suspect but unconfirmed as vectors. The presence of virus
in Iran and Algeria during two successive years may be explained by the
persistence of the virus in certain species during hibernation. Proof remains
to be supplied.
In conclusion, the criteria applied by arbovirologists have not been
fulfilled for many viruses transmitted by arthropods, among them AHS
virus. The easiest of these criteria is the isolation of wild virus in arthropods.
The study of transmission is more dificult and the investigation of ecologic
factors regulating the role of the vector cari only be carried out with long-
lasting and costly research work.
The last point, the reservoir, is hypothetical and, to use an expression of
MORNET and GILBERT [30], it is a ‘fact of logic’. TO prove its existence, how-
ever, is of utmost importance for the countries free of AHS [3].
There is no point in citing the possible reservoir species. We prefer to
propose a research plan following the pattern of epidemiologic investiga-
tions of arboviruses. The starting point of this plan is a serological investiga-
BOURDIN
26
tion in horses to determine the zone of strong endemicity with frequent
contacts between horse and virus (rural form of the disease). These contacts
imply that vector and reservoir are present. The serological investigation
must, therefore, allow localization of the ecologic nidus where the virus is
maintained in its basic cycle.
In Senegal, where AHS occurs habitually at the end of the hot season
in October and November, BOURDIN et al. [8] have examined 1,500 serum
samples from different regions in December and January. The samples cari
be classified in two groups according to the zone of origin. The first group
consists of samples of horses from agricultural regions, the second cornes
from agro-pastoral zones. The complement fixation test reveals the interest-
ing feature that 20 ‘A of the ‘agricultural’ horses have antibodies compared
with 66 ‘A in the ‘pastoral’ horses.
Before we comment on these results, it is necessary to summarize the
mode of life of the horses from both regions. In the agricultural zone the
draught horses live together with sheep in the yard of the house. They are
the abject of much attention and are fed and watered in the yard. In the
bigger agricultural villages 50-100 horses, 300-400 sheep and a few donkeys
may be found. Additionally there are cattle, usually kept in migratory herds,
which corne from sylvo-pastoral zones in the dry season and which live at
the outskirts of the village.
In the sylvo-pastoral zone the horses live differently. This zone is situ-
ated in the northwestern part of Senegal and is essentially pastoral, breeding
being extensive. Water supply has long been the limiting factor for breeding,
as the only water was found in deep Wells. Fortunately, the installation of
drilled Wells with large outputs 20 years ago has solved this problem. Around
these Wells 5,000-20,000 cattle and as many sheep and goats may live within
a radius of 20-30 kilometers. The Wells have initiated the founding of Peul’s
breeder villages where some horses, used for pack work or prestige, and
numerous donkeys, used for water transportation, are kept. The approach
to such a well is characterized by a sandy zone barren of pasture but covered
with excrement. Multitudes of birds breed close to the water and wild animals
are more abundant than in the agricultural region. Bynighthyenasandjackals
are frequent visitors.
Drilled or traditional Wells and their surroundings cari constitute an
ecological niche, in which the virus would be transmitted within its basic
cycle. In the immediate surroundings of the well two elements favoring the
multiplication of potential vectors are present: stagnating water, cisterns,
puddles and fresh cattle feces. More distantly there lives, apart from dogs
Ecology of African Horsesickness
2 7
and other domestic animais, a rather abundant wild fauna attracted by the
water and here a reservoir may be found.
This first tentative investigation must be confirmed by other surveys on
horses living close to other drilled Wells and natural Wells in the sylvo-
pastoral zone. Later, it Will become necessary to extend a serological investi-
gation by neutralizing and complement fixing tests to other domestic and
wild animals and to complete this study by capturing blood-sucking arthro-
pods and wild animals, particularly at the end of the hot season to try to
isolate the virus in suckling mice.
Acknowledgements
We would like to thank Dr. Y. ROBIN, Director of the Pasteur Institute of Dakar;
Dr. M. CORNET, entomologist of ORSTOM, and Mm. G. AIME, whose precious advice
has helped me to finish this work in time. We also thank Dr. DIALLO, Director of the
Service de l’Elevage du Sénégal and his coworkers and our own direct coworkers: Mm.
A. LAURENT, Mr. G. BERNARD and Mr. A. M’BAYE for their efficient and competent help.
References
1
ALEXANDER, R. A. : Studies on the neurotropic virus of horsesickness. Some physical
and chemical properties. Onderstepoort J. vet. Res. 4: 323-348 (1935).
2
ALEXANDER, R. A.: The 1944 epizootic of horsesickness in the Middle East. Onder-
stepoort J. vet. Sci. 23: 77-92 (1948).
3
Anonymous: Arbovirus Research Project. Animal report. (University of Ibadan,
Ibadan 1964-1969).
4
Anonymous: Animal health yearbook (FAO, WHO, Rome 1970).
5
Anonymous: Les arbovirus et leur rôle dans la pathologie humaine. WHO Tech.
Rep. No. 369 (WHO, Geneva 1971).
6
AUBREVILLE, A.: Climats, forêts et desertification de l’Afrique tropicale (Société
d’Editions géographiques maritimes et coloniales, Paris 1949).
7
BEVAN, L. E. W.: The transmission of African horsesickness to the dog by feeding.
Vet. J. 67: 402-408 (1911).
8
BO-IN, P.; BERNARI), G.; LAURENT, A. et M’BAYE, .A.: Enquête sérologique sur
la peste équine au Sénégal (in press).
9
Boum~~, P.: Unpublished information (1966).
10
Bans, P. et ROBIN, Y.: Les arbovirus. Généralités. Dengue. Fièvre a phlebotomes.
Encycl. Med. chu. Mal. Inf. 8065 (EIO): 1-12 (1970).
11
CORNET, M. : Persona1 communication (1972).
12
CURASSON,
G.: Traité de pathologie exotique vétérinaire et comparée, 2nd ed.,
pp. 170-207 (Vigot, Paris 1942).
BOURDIN
28
1 3
DOUTRE, M. P. and LECLERC, A.: Existence of African horsesickness virus type 9
in Tchad. Rev. Elev. Med. vet. Pays trop. 15: 241-245 (1962).
1 4
Du TOIT, R. M.: Transmission of bluetongue and horsesickness by culicoides.
Onderstepoort J. vet. Sci. 19: 7-16 (1944).
1 5
Du TOIT, R. M.: Cited in WETZEL et al. [67].
1 6
ELS, H. J. and VERWOERD, D. W.: Morphology of bluetongue virus. Virology 38:
213-219 (1969).
1 7
ERASMUS, B. J.: Persona1 communication (1971).
1 8
FOSTER, N. M.; JONES, R. H., and MCCRORY, B. R.: Preliminary investigation on
insect transmission of bluetongue virus in sheep. Amer. J. vet. Res. 24: 1195-1200
(1963).
1 9
GILBERT, Y.: Rapports Annuels du Laboratoire National (Curasson, Dakar 1954,
1957).
20
HENNING, M. W.: Animal diseases in South Africa, 3rd ed., pp. 785-808 (Central
News Agency Ltd., Onderstepoort 1956).
2 1
HOWELL, P. C. : The isolation and identification of further antigenic types of African
horsesickness virus. Onderstepoort J. vet. Res. 29: 139-149 (1962).
22
HOWELL, P. G.: La peste équine africaine. Maladies nouvelles des animaux, pp.
75-l 16 (FAO, Rome 1964).
23
JOUBERT, L. : Alerte a la peste équine. L’épizootie au Maroc et la protection en France.
Rec. Méd. vét. 118: 49-57 (1967).
24
LAABERKI, A.: Evolution d’une épizootie de P.E.A. au Maroc. Bull. 0. 1. E. 71:
921-936 (1969).
25
Tableau géographique de l’ouest africain au Moyen âge; thése doctorat és lettres,
Paris (1961).
26
MAURICE, Y. et PROVOST, A. : La peste équine à type 9 en Afrique centrale. Enquête
sérologique. Rev. Elev. Méd. vét. Pays trop. 20: 21-25 (1967).
27
MCINTOSH,
B. M.: Immunological types of horsesickness virus and their significance
in immunization. Onderstepoort J. vet. Res. 274:6 5-538 (1958).
28
MIRCHAMSY, H. and TASLIMI, H.: Thermal stability of African horsesickness virus.
Arch. ges. Virusforsch. 20: 275-282 (1967).
29
MORNET, P.: Sur une évolution atypique de la peste équine particulière à l’A.0.F.
Rev. Elev. Méd. vét. Pays trop. 3: 101-104 (1941).
30
MORNET, P. et GILBERT, Y.: La peste équine. Les maladies animales a virus. Collec-
tion de monographies (Expansion scientifique française, Paris 1968).
3 1
NEVILL, E. M.: Cattle and culicoides biting bridges as possible overwintering hosts of
bluetongue virus. Onderstepoort J. vet. Res. 38: 65-72 (1971).
32
NEVILL, E. M.: A significant new breeding site of Culicoides paIlid@nnis Carter,
Ingram and Macfie (Diptera: Caratopogonidae). J. sth afr. vet. med. Ass. 39: 61
(1968).
33
NIESCHULZ, 0.; BERFORD, C. A. H., and TOIT, R. Du: Investigation into the trans-
mission of horsesickness at Onderstepoort during the season 1931-32. Onderstepoort
J. vet. Sci. 3: 275-334 (1939).
34
NIESCHULZ,
0. and TOIT, R. Du: Investigation into the transmission of horsesickness
at Onderstepoort during the season 1932-33. Onderstepoort J. vet. Sci. 8: 213-270
(1937).
Ecology of African Horsesickness
29
35
OLLERMANN, R. A.; ELS, H. J., and ERASMUS, B. J.: Characterization of African
horsesickness virus. Arch. ges. Virusforsch. 29: 163-174 (1970).
36
WHO: Les arbovirus et leur rôle dans la pathologie humaine. Tech. Rep. Ser. N,-J.
369 (WHO, Geneva 1967).
37
ORUE, J.: Persona1 communication (1972).
38
OZAWA, Y. and HAZRATI, A.: Growth of African horsesickness virus in mo&ey
kidney ce11 cultures. Amer. J. vet. Res. 25: 505-511 (1964).
39
OZAWA, Y. and NAKATA, G. : Experimental transmission of African horsesickness by
means of mosquitoes. Amer. J. vet. Res. 26: 744-748 (1965).
40
OZAWA, Y.: NAKATA, G., and SHAD DEL, F. : Experimental transmission of African
horsesickness by means of mosquitoes. Arch. Inst. Razi 18: 119-125 (1966).
4 1
OZAWA, Y.; NAKATA, G.; SHAD DEL, F., and NAVAI, S.: Transmission of African
horsesickness by a species of mosquito, Aëdes aegypti Linnaeus. Arch. Inst. Razi 18:
147-151 (1966).
42
OZAWA, Y.; SHAD DEL, F.; NAKATA, G., and NAVAI, S.: Transmission of African
horsesickness by means of mosquito bites and replication of the virus in Aëdes
aegypti. Arch. Inst. Razi 22: 113-I 22 (1970).
43
P~ERCY, S. E.: Some observations on African horsesickness including an account
of an outbreak amongst dogs. East afr. agric. J. 17: 6244 (1951).
44
PILO-MORON, E.; VINCENT, J.; AIT MESBAH, 0. et F~RTHOMME,
G.: Origine de la
peste équine en Afrique du Nord: résultats d’une enquête sur les ânes du Sahara
algérien. Arch. Inst. Pasteur Algérie 47: 105-118 (1969).
45
PILO-MORON,
E. ; VINCENT, J. et SUREAU, P. : Présence de virus de la peste équine type
9 en République Algérienne. Identification des souches de virus isolées en 1965-1966.
Arch. Inst. Pasteur Algérie 44: 78-101 (1966).
46
PIRANI, A.: Solla peste equine in Erytrea. Nuova Vet. 5: 69-82 (1930).
47
POSTIGLIONE,
E. : Il servizio veterinario e le pui gravi malltra diffusibili del bestiame
nelle nostre colonie dell’Africa orientale. Clin. Vet. 58: 614-690 (1935).
48
RAFYI, A.: La peste équine. Bull. 0.1. E. 56: 170-215 (1961).
49
REID, N. R.: African horsesickness. The 1959-1969 epizootic of African horsesickness
in the Near East region and India and a note on technical assistance provided by the
F.A.O. of the United Nations. Brit. vet. J. 117: 137-142 (1961).
50
ROBIN, Y. et BRES, P.: Recherches effectuées sur l’écologie des arbovirus au Sénégal.
WHO Annual Report. (WHO, Geneva 1965-1969).
5 1
ROBIN, Y. et GONIDEC, C. LE: Recherches effectuées sur l’écologie des arbovirus au
Sénégal. WHO Annual Report (WHO Geneva 1970).
52
SCHUMBERG, A. und KUHN, P.: Uber die Ubertragung von Krankheiten durch
einheimische stechende Insekten. Arb. Gesundh. Amt 40: 209-234 (1912).
53
SCHWABE, C. E.: Veterinary medicine and human health, 2nd ed., pp. 160-228
(Williams & Wilkins, Baltimore 1969).
54
SERS, J. L.: La peste équine en Algérie; Etude sur l’épizootie d’automne 1965; thèse
doct. vét., Alfort (1965).
5 5
SHAH, K. V. : Investigation of African horsesickness in India. 1. Study of the natural
disease and the virus. Ind. J. vet. Sci. 34: 1-14 (1964).
56
STELLMANN, C.; MIRCHAMSY, H.; GILBERT H. et SANTUCCI, J.: La peste équine
africaine. Arch. Inst. Razi 22: 57-101 (1970).
B O U R D I N
30
51
STELLMANN, C.; M IRCHAMSY, H.; GILBERT, H. et SANTUCCI, J. : La peste équine
africaine. Bull. 0.1. E. 67: 887-947 (1967).
58
TAUFFLIEB, R.; CORNET, M. et CAMICAS, J. L.: Recherches effectuées sur l’écologie
des arbovirus au Sénégal. WHO Annual Reports 0. (WHO, Geneva 1965-1970).
59
THEILER, A.: Uber sud-afrikanische Zoonosen. Die Pferdeseuchen. 1. Dunparden-
ziekte. Schweiz. Arch. Tierheilk. 35: 145-162, 202-219 (1893).
60
THEILER, A.: Transmission of horsesickness into dogs. Rep. Govt. Vet. Bacterio-
logist, Transvaal, 1905-1906, pp. 192-213 (1907).
6 1
THEILER, A.: African horsesickness. Sth Africa Dept. Agric. Bull. 99: (1921).
62
THEILER, A.: The susceptibility of the dog to African horsesickness. J. camp. Path.
23: 315-325 (1910).
63
THEILER, A. : African horsesickness. In: A system of bacteriology in relation to medi-
cine, pp. 362-373 (HMSO, London 1930).
64
THEILER, A. and STOCKMANN, S.: On the correlation of virus diseases in stock in
South Africa. J. camp. Path. 18: 155-160 (1905).
6 5
VAN SACEGHEM: La peste du cheval ou horsesickness au Congo Belge: Bull. Soc.
Path. exp. II: 423-432 (1918).
66
VINCENT, J. : Persona1 communication (1972).
67
W ETZEL, H.; NEVILL, E. M., and ERASMUS, B. J.: Studies on the transmission of
African horsesickness. Onderstepoort J. vet. Res. 37: 165-168 (1970).
Author’s address: Dr. P. BOURDIN, I.E.M.V.T. Laboratoire de recherches vétérinaires,
B.P. 2057, Dakar (Senegal)
(Karger, Base1 1973)
Ecology of African Horsesickness
P. BOURDIN
1. E. M. V. T. Laboratoire de Recherches Vétérinaires, Dakar
I. Introduction
Ecologie study of animal disease is defined as the study of interactions
between environment and the respective populations in a medical sense.
The so-defined ecology is related by SCHWABE [53] to epidemiology, but
he says ‘an epidemiology seen by an ecologist or a biologist who forgets his
inborn anthropocentrism to sacrifice himself to holistic discipline’. The
ecologist, be he physician or veterinarian, has to take an interest not only
in the mode of disease transmission or restricted epidemiology, but also in
comparative pathology, immunology, vertebrate and invertebrate zoology,
climatology, agronomy, geography and even history.
SCHWABE [53 ] has developed a 3-stage working method for ecologists :
First stage. The research worker registers observations, descriptive
stage.
Second stage. He compares and analyzes his observations, analytic
stage.
Third stage. He verifies hypotheses developed during the analytic stage
by experimental work, experimental stage.
The ecology of African horsesickness (AHS) is related to arbovirus
ecology, which has led to the classification of AHS virus within this
group. OLLERMANN et al. [35] classify AHS virus together with Bluetongue
virus in the subgroup of diplornaviruses. AHS ecology resembles arbovirus
ecology and suggests the presence of a vector belonging to blood-sucking
arthropods and the existence of a host. TO date, experimental proof of
vectorial transmission according to the World Health Organization criteria
[5 ] has not been established and the existence of a natural host or reservoir
BOURDIN
13
is hypothetical. But if the identification of one or more reservoirs is of no
consequence for endemically infected countries [27], it is of primary impor-
tance for endangered countries to prevent the introduction and distribution of
the virus. We Will enumerate in this article the most important points known
and we Will then deal with three as yet unanswered questions: appearance of
an epizootic, existence of a vector and of a reservoir. These three questions
are intimately linked, the ecology being the whole. In a future research
program it Will be indispensable to clarify first the problem of the reservoir;
next that of the vector(s). Definition of the epizootiology of AHS depends
directly on solution of these problems.
II. Analysis of Facts and Experimental Work Concerning Ecology
In this section discussion Will be restricted to facts and the most useful
experimental work relating to ecologic study. The reviews of CURASSON
[12] and HENNING [20], as well as the monographs of RAFYI [48], HOWELL
[21], MORNET and GILBERT [30], STELLMANN et ul. [56, 571, give a much
more detailed account.
A. Study of Facts
1. Geographic Distribution in the Past and Present
The cradle of the disease lies in Africa, more precisely in the countries
on each side of the equator in the zones of dry tropical climate, Le. Sahelo-
Sudanese climate between the isohyetes from 250 to 1,000 mm. These zones
may be extended by altitude [6] as is evident from figure 1. South of the
equator the zones include Mozambique, Rhodesia, Zambia, Tanzania,
Kenya, Angola, Transvaal, Natal and part of the Orange Free State as far as
South Africa is concerned. North of the equator there is no marked boundary
and the respective countries lie between the 10th and 12th parallel of latitude
with a ventriflexion of the band from west to east and a dorsiflexion near the
Ethiopian mountains. This band includes Senegal, the south of Mali and of
Niger, Upper Volta, the south of Chad, Sudan up to Khartoum, part of
Ethiopia and of Eritrea. This zone has for centuries primarily been agro-
pastoral or sylvo-pastoral and the use of horses as draught or prestige ani-
mals is limited in the north by the desert and in the south by the tsetse fly
[251.
14
BOURDIN
Fig. 1. African countries with Sahelo-Sudanese climate and its variants.
From time to time AHS breaks out of its original cradle and extends
into neighboring countries. This extension reaches the Cape region in South
Africa, where the disease has been known since the 17th Century and recurs
about every 20 years. North of the equator the spread of AHS is much less
regular. In 1930-1931 AHS reached Egypt. In 1933-1934 it was again
observed in Egypt, Palestine and the Lebanon. In 1959, Iran was reached
and the disease spread east from there, extending into Afghanistan, Pakistan
and India (1960); to the west it was recognized in Iraq, Jordania, Syria,
Turkey, the Lebanon and Cyprus.
In 1965, AHS conquered the Saharian barrier and invaded the south
of Morocco and Algeria. After being dormant during winter it wandered
north into Tunisia and reached Spain in October 1966.
Since then the disease has apparently receded into its originating African
zone to prepare for a new eruption.
Ecology of African Horsesickness
15
2. Spread of AHS
AHS occurs endemically and periodically in the form of epidemics.
The endemic form is habitually seen in the countries of western and central
Africa with Sahelian climate favoring the breeding of horses. In this zone
some kind of ecologic equilibrium seems to exist between vector, reservoir
and the 1.5 million horses [4].
The home was introduced into Africa south of the Sahara about 2,000
B.C.: ‘The inventory of Saharian rock dwellings shows that between this
time and the start of our era the Sahara was crossed by numerous routes on
two axes, one going from Lybia to Gao, the other from South Morocco to
Mauritania and Tombuctu’ [25]. This stage was followed by trading with
horses between the Salt-producing countries of the Maghreb and the king-
doms of western Africa rich in gold mines. Commerce in horses was very
active up to the Middle Ages.
It is probable that these animals have acquired progressively a natural
resistance against AHS, a feature of the local breeds of today. That there
must have been contact with the virus has been proved by serological sur-
veillance studies carried out in nonvaccinated animais by MAURICE and
PROVOST in Central Africa [26] and BOURDIN et al. in Senegal and Mali
181. The ecologic equilibrium in these countries is upset by the importation
of susceptible horses. Between 1882 and 1925, animais imported from France
or Morocco by the military authorities were regularly decimated until a
decision was taken to use only native horses in the mounted units. Since then
AHS has become rare; it was observed occasionally when horses were im-
ported for riding clubs or studs. MORNET [29] has seen it in Senegal, DOUTRE
and LECLERC identified it in Chad [13] and BOURDIN [9] again in Senegal.
GILBERT [19] has observed the disease in animals sired by imported stallions
out of native mares.
The epidemic form seen within the last decade in the Middle East, Asia
and the Maghreb has been known in South Africa since the 17th Century in
the Cape region, when horses were first introduced there. In the more sep-
tentrional (northern) provinces of the South African Union (Natal, Trans-
vaal) the horses, the majority being vaccinated, are very probably in close
contact with the wild virus [17]. One cari presume that there has been estab-
lished in these regions some sort of artificial equilibrium.
The spread of AHS virus in new countries seems to be linked, according
to HOWELL [21, 221, to the introduction of equidae or contaminated insect
vectors, which find a zone with favorable ecologic conditions. He applies
BOURDIN
16
this explanation to the appearance of AHS during the summer of 1959 in the
coastal regions of Iran. After a winter pause the disease spread in 1960 along
the rivers, the traditional nomadic routes and the major road axes.
For PILO-MORON et al. [45], AHS arrived in the south of Morocco and
Algeria following the habitua1 routes (fig. 2). It isestimatedthat40,000-50,000
nomads still live in south Algeria and that these people migrate between
southern Morocco and Algeria and the Sahelian regions of Mali and Niger.
These people migrate with herds consisting of dromedaries, donkeys, sheep,
goats and a few dogs. The hypothesis of spread along these migratory routes
is based on the presence of neutralizing and complement-fixing antibodies
in donkeys living in the oases through which the routes lead. These animals
could perhaps serve as markers or relays.
3. Climate and Topography
Al1 authors stress the favorable influence of high humidity always
following abundant rains and heat. Cold and drought are, on the contrary,
unfavorable factors.
i
I
N igw
’ Chad
Sudîn
Y’
r
‘.!
‘7
ATLANTIC O C E A N
. ,
-;--,--J
L.
;
;.-a , ,
.i
L
Fig. 2. Traditional migratory routes used by nomadic populations.
Ecology of African Horsesickness
17
In the south of Africa, where the rainy season starts in December, the
most dangerous months are February, March and sometimes April. In
countries with a Sahelo-Sudanian climate the rain season lasts for 2-4
months with a maximum in August; it is followed by a rigorous dry season
lasting 6-8 months. The vapor tension may attain 18-22 mm between the
middle of June and the middle of November [6]. The months of October
and November, in particular, favor the eruption of AHS.
Countries with a continental or Mediterranean climate show a distinct
climatic incidence. RAFYI [48] noted in Iran that the disease spread in
summer 1959, stopped after the first frosts and reappeared in the following
year. In Algeria [54] the disease is widespread in October, regresses with the
cold and is reactivated the following season. In Morocco, AHS reached
alarming proportions at the end of the winter of 1965 to become extinguish-
ed in the summer of 1966 [24]. This writer mentioned thatspringandsummer,
1966, were relatively dry but were preceded by a very rainy autumn in 1965,
which covered the valleys with an abundant flora.
Topographically, the zones favoring the spread of AHS are the low,
swampy regions, the humid valleys with abundant flora [20], water courses,
Wells, puddles, mudholes and bore-holes [24, 48, 49, 541. In mountainous
regions, AHS is restricted to the lower zones. THEILER [59] observed AHS
in South Africa usually below 500 m; under exceptional climatic conditions
it was seen up to 1,200 m. In Ethiopia and Eritrea the disease does not occur
at altitudes higher than 1,200-1,500 m [47]. In Kenya, closer to the equator,
PIRANI [46] found a focus at 2,400 m altitude.
4. Susceptibility of Animais
The horse is extremely susceptible, with a morbidity of more than 95%
in epizootics and a mortality of 85% [48]. The mule is less susceptible with a
50-percent morbidity during epizootics.
The South African donkey (Equus asinus somalicus) is resistant [61].
Fatal cases have been seen in Egypt and Algeria in donkeys of the subspecies
Equus asinus africanus [2, 541. ORUE [37] observed a very high mortality
among the donkeys of the Cape Vert region. The zebra (Equus burchelli)
may show a clinically evident form of disease in South Africa [22].
The dog is infected by ingestion [62]. PIERCY [43] has described enzoo-
tics in animais of packs nourished with contaminated horse-meat.
The Angora goat, not susceptible under natural conditions, produces a
febrile response to virus inoculation. The blood of such animals is virulent
[62, 641.
BOURDIN
1 8
5. Mode of Transmission
Direct transmission is exceptional and, if it occurs, it is by accident only.
It is possible to keep healthy animals together with sick ones in an insect-
free stable protected from blood-sucking arthropods. The massive use of
insecticides during epizootics stopped the spread of virus in Spain in 1966.
These two facts and the influence of climate support the theory of insect-
borne transmission.
B. Analysis of Experimental Work
1. Resistance of AHS Virus to External Factors
The virus is stable outside the body. Putrefaction, dessication and tem-
perature variation exert little effect. Only pH values below 6 are detrimen-
tal. The ideal pH for conservation varies from 6.5 to 8 [l, 381.
Wild viscerotropic strains are stable at 4°C in OCG mixture [63];
neurotropic strains are stable in the presence of serum for 90 days at this
temperature [l, 281. At 37°C in the presence of calf serum the ce11 culture
adapted neurotropic virus loses 6 logs within 40 days and only 2 logs at
-20 to -25°C [OOO].
2. Persistence of Virus in Susceptible Animais
The horse experiences a short viremia; the virus cari be found in the
blood at the beginning of the febrile reaction and disappears fairly rapidly.
After the ninth day, isolation from blood is only rarely achieved, except for
the case described by THEILER [63] in which the disease was transmitted by
massive inoculation of a healthy horse with blood from a horse that was sick
90 days previously. Viremia is difficult to detect in donkeys.
MCINTOSH [27] could not isolate the virus from vaccinated horses
responding to challenge. After intracardiac inoculation of ferrets, however,
isolation was achieved from the blood of the ferret. The virus was pathogenic
for suckling mice. The author thinks that the virus masked by antibodies
in the horse was liberated in the ferret.
3. Study of Vectors: Virus in Blood-Sucking Insects and Transmission
Experiments
In many articles on AHS a number of arthropods are cited as possible
vectors. JOUBERT [23] compiled the following list :
Ecology of African Horsesickness
19
Culicidae:
Culex pipiens, Ochlerotatus, Aëdes caballi, A. aegypti,
Anopheles, S’tegomya
Simulidae :
Simulium
Tabanidae :
Tabanus pluto
Muscidae :
Stomoxys calcitrarw, Hyperosia minuta
Ceratopogonidae:
Culicoides pallidipennis
Ixodides
But, according to this worker, many of these insects are with certainty
‘simply winged needles’ and, in fact, successful experimental transmissions
cari be counted on the fingers of one hand. Stomoxys calcitrans cari mechani-
cally transmit AHS virus [52]. Mosquitoes may exceptionally transmit the
virus [34]. These authors found a predominance of A. caballi with some
Culex and Anopheles among the mosquitoes living at Onderstepoort. They
tried to transmit the disease to susceptible horses, either by inoculation with
crushed mosquitoes that had fed on viremic horses 12 h to 55 days before,
or by letting infected mosquitoes feed on susceptible horses. After 66 ex-
periments with Aëdes they obtained 4 positive results, a11 of them by inocula-
tion. In one case the infective meal took place 12 h previously, in the three
other cases 7 days before. Anopheles stephensi and C. pipiens infected horses
when they were fed virus in vitro and put on horses 15-22 days after the meal
[40, 411. The horses died late, 25-38 days after having been stung. OZAWA
et al. [42] continued their experiments using A. aegypti, fed in vitro with a
suspension of wild type 9 virus adapted to ce11 cultures (titer lOs.aID,,/O.l
ml). The presence of virus in the mosquitoes was checked from 1 to 36 days;
on the 36th day virus titers in certain lots of mosquitoes were still as high as
102-s IDse/O.l ml. The authors explain the titer as being due to virus replica-
tion in the mosquito. AHS was also transmitted to a horse by A. aegypti
which had been infected in vitro 19 days earlier. The horse died 18 days after
contact with the mosquitoes.
Du TOIT [14] captured Culicoides sp. in an AHS region and tried with-
out success to transmit the disease to a horse by its inoculation with these
crushed arthropods. In 1945, Culicoides were fed on a viremic home and 12
days later were put on a susceptible horse, which died of AHS [15].
WETZEL et al. [67] fed A. aegypti and C. pipiens fatigans from the
Onderstepoort collection and Culicoides sp. captured in light-traps on a
virus suspension and on a viremic horse. Al1 attempts to isolate virus from
crushed arthropods in suckling mice failed l-40 days after the infective
meal; the same Culicoides were not able to infect susceptible horses.
BOURDIN
20
4. Search for and Identification of the Reservoir
For the determination of a reservoir two complementary techniques cari
be used: intracerebral inoculation of suckling mice with material obtained
from different vertebrates and a systemic serologic investigation of many
species.
a) Serologic Investigations
Species suspected of harboring virus are selected and a large number of
samples are tested. The complement fixation test indicates a contact within
the last 6 months [22] and the reaction is group-specific. Serum neutraliza-
tion detects more temporarily distant contacts and is type-specific. TO our
knowledge, no systematic large-scale investigations have been carried out.
PILO-MORON et al. [44] found neutralizing antibodies in 69% of 84
donkeys living in south Algerian oases.
In dogs living in an AHS focus, 1 in 13 samples contained antibodies
1271.SHAH [55] was able to detect antibodies in samples from various species.
10 of 20 canine samples were positive, 8 of 42 donkeys and 1 of 42 cattle
had antibodies in their serum. No antibodies were found in 34 human, 17
caprine, 21 avian and 2 ovine serum samples.
b) Inoculation of Suckling Mice or other Susceptible Species
Experimental inoculation with material from species other than equidae
has only been done on a small scale. The results have always proven negative.
THEILER [60] and BEVAN [7] were able to isolate virus from experi-
mentally infected dogs. The same was possible in Angora goats [64] and in
sheep [65], but THEILER’S attempts [61] to isolate a virus from wild animals,
birds and amphibians were unsuccessful.
The difficulties to isolate AHS virus in nature from reservoirs or vectors
are comparable to those encountered by WHO for west African arboviruses.
Two centers of WHO, the Pasteur Institute in Dakar and the University of
Ibadan, are systematically looking for arboviruses in man, domestic animals,
wild animals and arthropods by intracerebral inoculation of 1 day-old
suckling mice. In case of a positive result after 2-3 passages the sensitivity of
the virus to ether and chloroform is checked and identification of the agent
is pursued.
In Senegal, the material originates mainly from two centers, one 80 km
from Dakar on the road to M’Bour in the forest of Bandia, the other near
the Gambian border in Saboya. There are only a few horses near these centers
Ecology of African Horsesickness
21
because of the tsetse fly. The reports by ROBIN and BRES [50], ROBIN and
LE GONIDEC [51] for the virological study and by :~AUFFLIEB et al. [58] for
the entomological investigations cari be summarized in two tables. In these,
the samples are classified in large groups according to their origin: man,
wild fauna, mosquitoes, ticks, other arthropods. The group ‘wild animals’
includes mammalians (insectivores, chiroptera, primates, rodents, carni-
vores, artiodactyles) and ranges from the excessively numerous birds to
reptiles. Liver, spleen and brain are routinely used to prepare inoculums.
Mosquitoes and ticks are identified, classified and used for inoculation in
lots. The group ‘other arthropods’ includes 162 lots of culicoides, 11 lots of
tabanids and 24 lots of phlebotomes.
The total inoculations in 8 years amounted to 9,788 with 224 viruses
isolated, only two of which were ether-resistant both being identified as
Coxsackie virus (tables 1, II).
In Nigeria, the WHO arbovirologists work at Ibadan, a town situated
in a humid tropical climate with very few horses. Investigations are also
Table Z. Number of inoculations with material from different species (extracted from
WHO reports)
Origin
1964 1965 1966 1967 1968 1969 1970 Total
M a n
113
552
284
1 9
54
332
162
1,516
Wild animals
1 7 1
297
529
291
448
241
1,230
3,207
Mosquitoesl
5
113
145
488
327
483
1,171
2,732,
Ticks
1
80
74
256
439
106
1,016
1,974
Other arthropods2
-
-
-
-
-
154
205
359
1 Number of lots.
2 The group consists of the following lots: Culicoides, 297, tabanids, 14, phlebotomes,
48.
Table ZZ. Identified viruses (extracted from WHO reports)
Origin
1968
1969
1970
Total
M a n
39
1 6
2
57
Wild animals
46
-
10
56
Mosquitoes
8 1
10
9 1
Ticks
1 2
-
16
28
Other arthropods
2
0
0
2
BOURDIN
22
carried out in north Nigeria, a zone of Sahelo-Sudanian climate. Other
investigations take place in Chad, Dahomey, Togo and the north Cameroons.
The yearly reports (1964-1969) published by the University of Ibadan in the
.
framework of the Arbovirus Research Project cari be summarized in two
tables. Table III concerns the inoculations per year and the zoological groups.
Table IV gives the number of viruses isolated from and identified in these
groups. In this table each column contains two series of numbers: those to
the left for arboviruses, those to the right for other viruses.
The samples from domestic animals are blood samples from animals
from the central and northern provinces and slaughtered at Ibadan (bovine,
Table III. Yearly inoculations performed at Ibadan
Origin
1964
1965
1966
1967
1968
1969
Total
M a n
1,815
662
2,212
1,793
2,560
2,115
11,157
Domestic animal+
220
1,568
1,295
619
330
1,048
5,140
Indicator animals’
207
263
273
709
172
1,624
Wild faunaa
438
1,279
3,429
2,794
2,723
10,663
Ticks
1,209
1,328
579
1,436
3,093
180
7,825
Mosquitoes
96
140
193
280
714
1,423
Culicoides
13
-
399
830
325
1,567
Other arthropods
13 83
63
68
10
297
1 Animals coming from central and northern Nigeria and slaughtered at Ibadan.
2 Indicator animals consist of suckling mice, chicken, rhesus monkeys and calves.
3 Mammalians, birds and reptiles captured in Nigeria, Chad and Dahomey.
Table IV. Virus isolated by near and species
Origin
1964
1965
1966
1967
1968
1969
Total
A NA A NA A NA A NA A NA A NA A NA
M a n
8 79
1 5
20 31
21 12
41 23 104 6 195 156
Domesticanimals
14 -
160 2
45 13
13 14
26 19
20 1 278 49
Indicator animals -
-
4 -
2 -
5 -
8 10
-
-
21 10
Wild fauna
-
-
1 -
7 3
16 1
25 10
1 2
50 16
Ticks
179 -
152 -
66 5
182 -
191 7
25 - 795 12
Mosquitoes
-
-
-
-
- -
1 -
1 -
19 -
21 -
Culicoides sp.
-
-
-
-
- -
44 -
3 7
1 2
48 9
Other arthropods -
-
-
-
- -
2 -
1 -
- -
3 -
A = arbovirus, NA = nonarbovirus.
Ecology of African Horsesickness
23
ovine, caprine and porcine samples). The ‘sentine1 animals’ or indicator
animals (suckling mice, rhesus monkeys, chickens and calves) are exposed in
the park of the University of Ibadan or on a nearby farm. The number of
samples taken for inoculation amounts to 39,696 with 1,613 viruses isolated
and identified in 6 years, 1,361 of these were arboviruses and 252 other viruses
(herpes, myxovirus, picornavirus, poxvirus, rhabdovirus, reovirus). Among
the ‘other viruses’, 9 bluetongue viruses were isolated, 8 from Culicoides sp.
and one from Crocidura SP., both captured near Ibadan.
III. Discussion
AHS is an endemic, essentially African disease occurring on both sides
i of the equator in dry, tropical climates, such as the mode1 Sahelo-Sudanian
’
climate with its variants. The Sahara constitutes the northern limit of the
endemic zone
The disease appears in seasonal rhythm and is favored by heat aud
humidity. The horse seems to be an accidental host. In western Africa the
horse has acquired a very solid natural immunity due to its very ancient
cohabitation with the virus. Periodically, the disease breaks out of its habit-
ual geographic area and reaches countries with a more temperate climate.
This occurs generally during the hot season in those countries. It Will dis-
appear in winter and cari sometimes recur during the following summer to
become extinguished after the first frosts.
Arboviruses are transmitted in different cycles [lO]. The same diversity
of transmission must be expected to apply to AHS. Accordingly, we cari
distinguish a basic sylvatic cycle in the absence of the horse: infected verte-
brate -f arthropod -+ susceptible vertebrate.
In the proximity of such a cycle an equine population may exist and,
depending on the frequency of contacts, the result Will be a zone of weaker or
stronger endemicity. A zone of strong endemicity is characterized by frequent
contacts with susceptible populations, horses in our case, the contacts being
old in adults, recent in Young animals. The arbovirologists designate the
form of evolution as ‘rural’, following, in the case of AHS, this scheme:
infected vertebrate -f arthropod -f horse. Whether the horse is a dead-end
is a question that remains unanswered [lO].
The infected horse may be introduced into a zone with susceptible
populations at a time when blood-sucking arthropods abound. The arthro-
pods may be true biological vectors or ‘simple winged needles’; in any case,
BOURDIN
24
ANS Will become epidemic and the new cycle is as follows : infected horse -+
arthropod -+ susceptible horse.
In this zone the ecologic conditions may differ from the zone where the
basic cycle takes place. If there is no reservoir, the greater variations of clima-
tic conditions Will interrupt the reproduction of vectors. In certain vectors,
however, the phenomenon of the diapause may give rise to a reappearance
of the virus as soon as the temperature becomes milder. Disappearance of the
disease Will then take place during the following winter.
This explanation seems to apply to the development of AHS in Iran and
its spread to the east and the west. HOWELL [21] feels that the first focus was
due to the introduction of infected horses from East Africa. The introduc-
tion of the virus into North Africa is more difficult to explain. In Algeria
the first foci were observed in locations on the northern border of the Sahara
(Bechar, Tadjmout, Laghouat) at the end of the routes used by nomads [45].
A serologic investigation of Saharian donkeys living close to the caravan
routes reveals the presence of neutralizing and complement fixing antibodies
at high titers. The trans-Saharian routes are used by nomadic populations
amounting to about 50,000 people [66]. These nomads migrate, following
the watering places of the northern Sahara toward the nearest Sahelian
regions of Mali and Niger, where they find pastures for their herds consisting
mainly of dromedaries, sheep, goats and donkeys. According to PILO-
M ORON et al. [44], the donkeys (Equus asinus africanus) undergo an in-
apparent infection and serve as relays between Sahelian and Algerian horses.
This hypothesis remains to be confirmed by virus isolation.
Ecologie circumstances point to the transmission of AHS by blood-
sucking arthropods, but experimental evidence is contradictory. Two groups
of vectors, mosquitoes and Culicoides, are mainly suspect. Some workers
have obtained positive results with mosquitoes, negative with Culicoides;
other have found the exact contrary. It is important to realize that the isola-
tion of a virus from arthropods does not a priori imply a biological trans-
mission. Two minimal conditions must be fulfilled, namely replication in the
arthropod and transmission by the sting. W ETZEL et al. [67] apply these
requirements to AHS virus and plan to inoculate arthropods intrathoraci-
cally to verify if replication takes place. There is no profit in discussing the
advantages and inconveniences of this artificial experiment; we shall con-
sider only the ecological reasons and uncover the difficulties in identifying
the vector.
The first difficulty is of a statistical nature. In 1944, Du TOIT isolated
AHS virus once from Culicoides sp. and achieved transmission in 1945 [14,
Ecology of African Horsesickness
2s
15,671. In 1944, the same author also isolated bluetongue virus three times
from Culicoides sp. and achieved transmission by C. pallidipennis. In the
USA, FOSTER et al. [18] transmitted bluetongue with Culicoides variipennis.
In 1971, NEVILL regularly isolated bluetongue from Culicoides sp. in South
Africa during the warm season, provided large inoculums were given. Why
this difference in spite of the close resemblance of the two viruses ? Simply
because the reservoir of bluetongue is represented by 12.8 million cattle and
the host by 39 million sheep, whereas AI-KS concerns only 380,000 horses,
the only vertebrate permitting isolation in South. Africa, as long as the
reservoir is not identified.
The second difficulty is due to the Culicoides themselves; their identifi-
cation is difficult and their biology incompletely defined. Some of them live
in close contact with cattle, C. pallidipennis in particular, which lays eggs and
reproduces in fresh cattle feces [32]. This Culicoides represents 97% of a11
those identified in Africa [l 1, 31, 321. Moreover, Culicoides do not depend
on a blood meal for their first oviposition. Within a lot of Culicoides captured
in light-traps one cari distinguish the insect having completed a gonotrophic
cycle from the immature, but one cannot enumerate the cycles. The devia-
tion of life and the frequency of contacts with the host are equally poorly
defined.
The biology of mosquitoes is better known. In the case of AHS, un-
fortunately, the virus has never been isolated in the field, even though experi-
mental transmission has been successful with certain species. Mosquitoes
are, therefore, suspect but unconfirmed as vectors. The presence of virus
in Iran and Algeria during two successive years may be explained by the
persistence of the virus in certain species during hibernation. Proof remains
to be supplied.
In conclusion, the criteria applied by arbovirologists have not been
fulfilled for many viruses transmitted by arthropods, among them AHS
virus. The easiest of these criteria is the isolation of wild virus in arthropods.
The study of transmission is more dificult and the investigation of ecologic
factors regulating the role of the vector cari only be carried out with long-
lasting and costly research work.
The last point, the reservoir, is hypothetical and, to use an expression of
MORNET and GILBERT [30], it is a ‘fact of logic’. TO prove its existence, how-
ever, is of utmost importance for the countries free of AHS [3].
There is no point in citing the possible reservoir species. We prefer to
propose a research plan following the pattern of epidemiologic investiga-
tions of arboviruses. The starting point of this plan is a serological investiga-
BOURDIN
26
tion in horses to determine the zone of strong endemicity with frequent
contacts between horse and virus (rural form of the disease). These contacts
imply that vector and reservoir are present. The serological investigation
must, therefore, allow localization of the ecologic nidus where the virus is
maintained in its basic cycle.
In Senegal, where AHS occurs habitually at the end of the hot season
in October and November, BOURDIN et al. [8] have examined 1,500 serum
samples from different regions in December and January. The samples cari
be classified in two groups according to the zone of origin. The first group
consists of samples of horses from agricultural regions, the second cornes
from agro-pastoral zones. The complement fixation test reveals the interest-
ing feature that 20 ‘A of the ‘agricultural’ horses have antibodies compared
with 66 ‘A in the ‘pastoral’ horses.
Before we comment on these results, it is necessary to summarize the
mode of life of the horses from both regions. In the agricultural zone the
draught horses live together with sheep in the yard of the house. They are
the abject of much attention and are fed and watered in the yard. In the
bigger agricultural villages 50-100 horses, 300-400 sheep and a few donkeys
may be found. Additionally there are cattle, usually kept in migratory herds,
which corne from sylvo-pastoral zones in the dry season and which live at
the outskirts of the village.
In the sylvo-pastoral zone the horses live differently. This zone is situ-
ated in the northwestern part of Senegal and is essentially pastoral, breeding
being extensive. Water supply has long been the limiting factor for breeding,
as the only water was found in deep Wells. Fortunately, the installation of
drilled Wells with large outputs 20 years ago has solved this problem. Around
these Wells 5,000-20,000 cattle and as many sheep and goats may live within
a radius of 20-30 kilometers. The Wells have initiated the founding of Peul’s
breeder villages where some horses, used for pack work or prestige, and
numerous donkeys, used for water transportation, are kept. The approach
to such a well is characterized by a sandy zone barren of pasture but covered
with excrement. Multitudes of birds breed close to the water and wild animals
are more abundant than in the agricultural region. Bynighthyenasandjackals
are frequent visitors.
Drilled or traditional Wells and their surroundings cari constitute an
ecological niche, in which the virus would be transmitted within its basic
cycle. In the immediate surroundings of the well two elements favoring the
multiplication of potential vectors are present: stagnating water, cisterns,
puddles and fresh cattle feces. More distantly there lives, apart from dogs
Ecology of African Horsesickness
2 7
and other domestic animais, a rather abundant wild fauna attracted by the
water and here a reservoir may be found.
This first tentative investigation must be confirmed by other surveys on
horses living close to other drilled Wells and natural Wells in the sylvo-
pastoral zone. Later, it Will become necessary to extend a serological investi-
gation by neutralizing and complement fixing tests to other domestic and
wild animals and to complete this study by capturing blood-sucking arthro-
pods and wild animals, particularly at the end of the hot season to try to
isolate the virus in suckling mice.
Acknowledgements
We would like to thank Dr. Y. ROBIN, Director of the Pasteur Institute of Dakar;
Dr. M. CORNET, entomologist of ORSTOM, and Mm. G. AIME, whose precious advice
has helped me to finish this work in time. We also thank Dr. DIALLO, Director of the
Service de l’Elevage du Sénégal and his coworkers and our own direct coworkers: Mm.
A. LAURENT, Mr. G. BERNARD and Mr. A. M’BAYE for their efficient and competent help.
References
1
ALEXANDER, R. A. : Studies on the neurotropic virus of horsesickness. Some physical
and chemical properties. Onderstepoort J. vet. Res. 4: 323-348 (1935).
2
ALEXANDER, R. A.: The 1944 epizootic of horsesickness in the Middle East. Onder-
stepoort J. vet. Sci. 23: 77-92 (1948).
3
Anonymous: Arbovirus Research Project. Animal report. (University of Ibadan,
Ibadan 1964-1969).
4
Anonymous: Animal health yearbook (FAO, WHO, Rome 1970).
5
Anonymous: Les arbovirus et leur rôle dans la pathologie humaine. WHO Tech.
Rep. No. 369 (WHO, Geneva 1971).
6
AUBREVILLE, A.: Climats, forêts et desertification de l’Afrique tropicale (Société
d’Editions géographiques maritimes et coloniales, Paris 1949).
7
BEVAN, L. E. W.: The transmission of African horsesickness to the dog by feeding.
Vet. J. 67: 402-408 (1911).
8
BO-IN, P.; BERNARI), G.; LAURENT, A. et M’BAYE, .A.: Enquête sérologique sur
la peste équine au Sénégal (in press).
9
Boum~~, P.: Unpublished information (1966).
10
Bans, P. et ROBIN, Y.: Les arbovirus. Généralités. Dengue. Fièvre a phlebotomes.
Encycl. Med. chu. Mal. Inf. 8065 (EIO): 1-12 (1970).
11
CORNET, M. : Persona1 communication (1972).
12
CURASSON,
G.: Traité de pathologie exotique vétérinaire et comparée, 2nd ed.,
pp. 170-207 (Vigot, Paris 1942).
BOURDIN
28
1 3
DOUTRE, M. P. and LECLERC, A.: Existence of African horsesickness virus type 9
in Tchad. Rev. Elev. Med. vet. Pays trop. 15: 241-245 (1962).
1 4
Du TOIT, R. M.: Transmission of bluetongue and horsesickness by culicoides.
Onderstepoort J. vet. Sci. 19: 7-16 (1944).
1 5
Du TOIT, R. M.: Cited in WETZEL et al. [67].
1 6
ELS, H. J. and VERWOERD, D. W.: Morphology of bluetongue virus. Virology 38:
213-219 (1969).
1 7
ERASMUS, B. J.: Persona1 communication (1971).
1 8
FOSTER, N. M.; JONES, R. H., and MCCRORY, B. R.: Preliminary investigation on
insect transmission of bluetongue virus in sheep. Amer. J. vet. Res. 24: 1195-1200
(1963).
1 9
GILBERT, Y.: Rapports Annuels du Laboratoire National (Curasson, Dakar 1954,
1957).
20
HENNING, M. W.: Animal diseases in South Africa, 3rd ed., pp. 785-808 (Central
News Agency Ltd., Onderstepoort 1956).
2 1
HOWELL, P. C. : The isolation and identification of further antigenic types of African
horsesickness virus. Onderstepoort J. vet. Res. 29: 139-149 (1962).
22
HOWELL, P. G.: La peste équine africaine. Maladies nouvelles des animaux, pp.
75-l 16 (FAO, Rome 1964).
23
JOUBERT, L. : Alerte a la peste équine. L’épizootie au Maroc et la protection en France.
Rec. Méd. vét. 118: 49-57 (1967).
24
LAABERKI, A.: Evolution d’une épizootie de P.E.A. au Maroc. Bull. 0. 1. E. 71:
921-936 (1969).
25
Tableau géographique de l’ouest africain au Moyen âge; thése doctorat és lettres,
Paris (1961).
26
MAURICE, Y. et PROVOST, A. : La peste équine à type 9 en Afrique centrale. Enquête
sérologique. Rev. Elev. Méd. vét. Pays trop. 20: 21-25 (1967).
27
MCINTOSH,
B. M.: Immunological types of horsesickness virus and their significance
in immunization. Onderstepoort J. vet. Res. 274:6 5-538 (1958).
28
MIRCHAMSY, H. and TASLIMI, H.: Thermal stability of African horsesickness virus.
Arch. ges. Virusforsch. 20: 275-282 (1967).
29
MORNET, P.: Sur une évolution atypique de la peste équine particulière à l’A.0.F.
Rev. Elev. Méd. vét. Pays trop. 3: 101-104 (1941).
30
MORNET, P. et GILBERT, Y.: La peste équine. Les maladies animales a virus. Collec-
tion de monographies (Expansion scientifique française, Paris 1968).
3 1
NEVILL, E. M.: Cattle and culicoides biting bridges as possible overwintering hosts of
bluetongue virus. Onderstepoort J. vet. Res. 38: 65-72 (1971).
32
NEVILL, E. M.: A significant new breeding site of Culicoides paIlid@nnis Carter,
Ingram and Macfie (Diptera: Caratopogonidae). J. sth afr. vet. med. Ass. 39: 61
(1968).
33
NIESCHULZ, 0.; BERFORD, C. A. H., and TOIT, R. Du: Investigation into the trans-
mission of horsesickness at Onderstepoort during the season 1931-32. Onderstepoort
J. vet. Sci. 3: 275-334 (1939).
34
NIESCHULZ,
0. and TOIT, R. Du: Investigation into the transmission of horsesickness
at Onderstepoort during the season 1932-33. Onderstepoort J. vet. Sci. 8: 213-270
(1937).
Ecology of African Horsesickness
29
35
OLLERMANN, R. A.; ELS, H. J., and ERASMUS, B. J.: Characterization of African
horsesickness virus. Arch. ges. Virusforsch. 29: 163-174 (1970).
36
WHO: Les arbovirus et leur rôle dans la pathologie humaine. Tech. Rep. Ser. N,-J.
369 (WHO, Geneva 1967).
37
ORUE, J.: Persona1 communication (1972).
38
OZAWA, Y. and HAZRATI, A.: Growth of African horsesickness virus in mo&ey
kidney ce11 cultures. Amer. J. vet. Res. 25: 505-511 (1964).
39
OZAWA, Y. and NAKATA, G. : Experimental transmission of African horsesickness by
means of mosquitoes. Amer. J. vet. Res. 26: 744-748 (1965).
40
OZAWA, Y.: NAKATA, G., and SHAD DEL, F. : Experimental transmission of African
horsesickness by means of mosquitoes. Arch. Inst. Razi 18: 119-125 (1966).
4 1
OZAWA, Y.; NAKATA, G.; SHAD DEL, F., and NAVAI, S.: Transmission of African
horsesickness by a species of mosquito, Aëdes aegypti Linnaeus. Arch. Inst. Razi 18:
147-151 (1966).
42
OZAWA, Y.; SHAD DEL, F.; NAKATA, G., and NAVAI, S.: Transmission of African
horsesickness by means of mosquito bites and replication of the virus in Aëdes
aegypti. Arch. Inst. Razi 22: 113-I 22 (1970).
43
P~ERCY, S. E.: Some observations on African horsesickness including an account
of an outbreak amongst dogs. East afr. agric. J. 17: 6244 (1951).
44
PILO-MORON, E.; VINCENT, J.; AIT MESBAH, 0. et F~RTHOMME,
G.: Origine de la
peste équine en Afrique du Nord: résultats d’une enquête sur les ânes du Sahara
algérien. Arch. Inst. Pasteur Algérie 47: 105-118 (1969).
45
PILO-MORON,
E. ; VINCENT, J. et SUREAU, P. : Présence de virus de la peste équine type
9 en République Algérienne. Identification des souches de virus isolées en 1965-1966.
Arch. Inst. Pasteur Algérie 44: 78-101 (1966).
46
PIRANI, A.: Solla peste equine in Erytrea. Nuova Vet. 5: 69-82 (1930).
47
POSTIGLIONE,
E. : Il servizio veterinario e le pui gravi malltra diffusibili del bestiame
nelle nostre colonie dell’Africa orientale. Clin. Vet. 58: 614-690 (1935).
48
RAFYI, A.: La peste équine. Bull. 0.1. E. 56: 170-215 (1961).
49
REID, N. R.: African horsesickness. The 1959-1969 epizootic of African horsesickness
in the Near East region and India and a note on technical assistance provided by the
F.A.O. of the United Nations. Brit. vet. J. 117: 137-142 (1961).
50
ROBIN, Y. et BRES, P.: Recherches effectuées sur l’écologie des arbovirus au Sénégal.
WHO Annual Report. (WHO, Geneva 1965-1969).
5 1
ROBIN, Y. et GONIDEC, C. LE: Recherches effectuées sur l’écologie des arbovirus au
Sénégal. WHO Annual Report (WHO Geneva 1970).
52
SCHUMBERG, A. und KUHN, P.: Uber die Ubertragung von Krankheiten durch
einheimische stechende Insekten. Arb. Gesundh. Amt 40: 209-234 (1912).
53
SCHWABE, C. E.: Veterinary medicine and human health, 2nd ed., pp. 160-228
(Williams & Wilkins, Baltimore 1969).
54
SERS, J. L.: La peste équine en Algérie; Etude sur l’épizootie d’automne 1965; thèse
doct. vét., Alfort (1965).
5 5
SHAH, K. V. : Investigation of African horsesickness in India. 1. Study of the natural
disease and the virus. Ind. J. vet. Sci. 34: 1-14 (1964).
56
STELLMANN, C.; MIRCHAMSY, H.; GILBERT H. et SANTUCCI, J.: La peste équine
africaine. Arch. Inst. Razi 22: 57-101 (1970).
B O U R D I N
30
51
STELLMANN, C.; M IRCHAMSY, H.; GILBERT, H. et SANTUCCI, J. : La peste équine
africaine. Bull. 0.1. E. 67: 887-947 (1967).
58
TAUFFLIEB, R.; CORNET, M. et CAMICAS, J. L.: Recherches effectuées sur l’écologie
des arbovirus au Sénégal. WHO Annual Reports 0. (WHO, Geneva 1965-1970).
59
THEILER, A.: Uber sud-afrikanische Zoonosen. Die Pferdeseuchen. 1. Dunparden-
ziekte. Schweiz. Arch. Tierheilk. 35: 145-162, 202-219 (1893).
60
THEILER, A.: Transmission of horsesickness into dogs. Rep. Govt. Vet. Bacterio-
logist, Transvaal, 1905-1906, pp. 192-213 (1907).
6 1
THEILER, A.: African horsesickness. Sth Africa Dept. Agric. Bull. 99: (1921).
62
THEILER, A.: The susceptibility of the dog to African horsesickness. J. camp. Path.
23: 315-325 (1910).
63
THEILER, A. : African horsesickness. In: A system of bacteriology in relation to medi-
cine, pp. 362-373 (HMSO, London 1930).
64
THEILER, A. and STOCKMANN, S.: On the correlation of virus diseases in stock in
South Africa. J. camp. Path. 18: 155-160 (1905).
6 5
VAN SACEGHEM: La peste du cheval ou horsesickness au Congo Belge: Bull. Soc.
Path. exp. II: 423-432 (1918).
66
VINCENT, J. : Persona1 communication (1972).
67
W ETZEL, H.; NEVILL, E. M., and ERASMUS, B. J.: Studies on the transmission of
African horsesickness. Onderstepoort J. vet. Res. 37: 165-168 (1970).
Author’s address: Dr. P. BOURDIN, I.E.M.V.T. Laboratoire de recherches vétérinaires,
B.P. 2057, Dakar (Senegal)