Host feeding of mosquitoes (diptera culicidae) associated with the recurrence of rift valley fever in egypt

Mixed blood meals from humans and RVF virus susceptible animals were identiÞed in the predominant mosquitoes.. KEY WORDS mosquito, blood meal, Rift Valley fever, Egypt RIFTVALLEY FEVERRV

https://www.researchgate.net/publication/12708424

Host Feeding of Mosquitoes (Diptera: Culicidae) Associated with the

Recurrence of Rift Valley Fever in

Egypt

Article in Journal of Medical Entomology · December 1999

DOI: 10.1093/jmedent/36.6.709 · Source: PubMed

CITATIONS

29

READS 20

8 authors , including:

Abd-Alla Gad

National Authority for Remote Se…

93 PUBLICATIONS 973 CITATIONS

SEE PROFILE

Reda Ramzy

National Nutrition Institute

95 PUBLICATIONS 2,078 CITATIONS SEE PROFILE

Steven M Presley

Texas Tech University

61 PUBLICATIONS 490 CITATIONS

SEE PROFILE

Ali Hassan

Institute of Environmental Studie…

38 PUBLICATIONS 341 CITATIONS SEE PROFILE

All content following this page was uploaded by Ali Hassan on 26 February 2017.

The user has requested enhancement of the downloaded file

Recurrence of Rift Valley Fever in Egypt

ADEL M GAD,1HODA A FARID,1REDA R M RAMZY,1MAHMOUD B RIAD,1

STEVEN M PRESLEY,2, 3STANTON E COPE,2, 4MOSSAD M HASSAN,1 ANDALI N HASSAN1

J Med Entomol 36(6): 709Ð714 (1999)

ABSTRACT In 1993, Rift Valley fever (RVF) virus reappeared in Egypt We determined the

prevalence and feeding patterns of mosquitoes in 5 villages where the virus was active Of 10 species

recovered, Aedes caspius (Pallas), Culex pipiens L., Cx antennatus (Becker), and Cx perexiguus

Theobald constituted 99% of ⬎35,000 mosquitoes captured in dry ice-baited CDC light traps Ae.

caspius was most prevalent, except at NagÕ El Hagar where it was replaced by Cx perexiguus Cx.

pipiens ranked 2nd, except at NagÕ El Ghuneimiya, where it was replaced by Cx antennatus Most

blood meals analyzed by an enzyme-linked immunosorbent assay reacted to ⱖ1 antiserum Cx.

pipienswas mainly anthropophagic, and therefore may have been the main vector of RVF virus

among humans Ae caspius feeds were chießy from humans, bovines, and equines Cx antennatus and

Cx perexiguusfed generally on bovines Mixed blood meals from humans and RVF virus susceptible

animals were identiÞed in the predominant mosquitoes Prevalence and host selection, as well as

predicted probability for a blood meal being interrupted, indicated that Ae caspius may have served

as a bridge vector between humans and bovines in 4 of the villages Cx perexiguus may have played

this role at NagÕ El Hagar Because potential vectors are abundant, susceptible domestic animals are

associated closely with humans, and surveillance of imported livestock is not systematic, we conclude

that RVF virus sporadically will recur in Egypt

KEY WORDS mosquito, blood meal, Rift Valley fever, Egypt

RIFTVALLEY FEVER(RVF) virus was Þrst introduced

into southern Egypt in 1977 from where it rapidly

spread northward to the Nile Delta causing an

esti-mated 600 human deaths as well as abortion and death

in sheep and cattle (Hoogstraal et al 1979, Meegan

1979) The virus persisted during 1978 After a 12-yr

absence, RVF virus reappeared in May 1993 in

south-ern Egypt, with an estimated 600Ð1,500 human

infec-tions and widespread aborinfec-tions and deaths in domestic

livestock Sera collected from sheep, goats, buffalo,

and cattle in the Aswan governorate were positive for

antibodies to RVF virus (Arthur et al 1993) Evidence

of RVF virus activity also was detected in the Nile

Delta and the Faiyum Oasis (Corwin et al 1993, WHO

1994)

To acquire RVF virus, mosquitoes must feed on

viremic mammals; therefore, it is important to

under-stand mosquito host-feeding patterns During the

1977Ð1978 epidemics, Culex pipiens L was implicated

as the primary vector of RVF virus, based on a single

virus isolation from an unengorged female

(Hoog-straal et al 1979) and its blood feeding habits (Kenawy

et al 1987, Gad et al 1995) Aedes caspius (Pallas) and

Cx antennatus(Becker) were suspected of dissemi-nating the virus among livestock, based on feeding patterns (Kenawy et al 1987, Gad et al 1995) and vector competence (Gad et al 1987) However, these studies of host selection by mosquitoes (Zimmerman

et al 1985) were carried out after RVF virus was last documented in 1981 (Imam et al 1981)

During the RVF virus outbreak in 1993, we initiated

a multidisciplinary study in Aswan governorate En-tomological efforts determined the prevalence and host feeding patterns of mosquitoes at 5 villages where human or livestock disease was occurring

Materials and Methods Mosquito Collections Mosquitoes were sampled

from 14 to 24 August 1993 in the Aswan governorate

on the southernmost part of the Nile Valley in 5 ag-ricultural villages north of Aswan City (Fig 1) where animal and human RVF virus infections were occur-ring The villages were environmentally and socioeco-nomically similar Houses generally were associated with animal sheds where domestic animals (horses, donkeys, cattle, buffalo, goats, sheep, and poultry) were kept at night Epidemiological investigations in NagÕ El Hagar and Sabil Abu El Magd revealed that the human population exceeded 2,000 in each village An-imal host censuses included only those anAn-imals

sus-ceptible to RVF virus In NagÕ El Hagar (n ⫽ 44

1 Research and Training Center on Vectors of Diseases, Ain Shams

University, Cairo, Egypt.

2 Naval Medical Research Unit No 3, Abbassia, Cairo, Egypt.

3 Current address: Training Programs Br C 462, Marine Corps

Com-bat Development Command, 3300 Russell Road, Quantico, VA 22134Ð

5001.

4 Current address: Navy Environmental Health Center, 2510

Walmer Avenue, Norfolk, VA 23513.

0022-2585/99/0709Ð0714$02.00/0 䉷 1999 Entomological Society of America

households), bovines (cattle only) and ovines (sheep

and goats) constituted 13.5 and 86.5%, respectively, of

the livestock (n ⫽ 252) Host populations consisted of

7.4 humans, 0.8 bovines, and 5.0 ovines per household

In Sabil Abu El Magd (n ⫽ 52 households), bovines

(cattle and buffaloes) and ovines accounted for 36.5

and 63.5%, respectively, of the livestock (n ⫽ 214).

There were 7.0 humans, 1.5 bovines, and 2.6 ovines per

household Houses mostly were surrounded by

sug-arcane Þelds except in NagÕ El Hagar Mosquitoes

were collected by dry ice-baited traps (without light)

operated from sunset to sunrise Blood-fed mosquitoes

were recovered from these traps, and on a few

occa-sions collected resting inside buildings From 7 to 28

traps per village per night were hung outdoors on trees

or streetlights near houses on the edges of cultivated

Þelds during 6 consecutive nights in NagÕ El Hagar, 4

in Sabil Abu El Magd, and 2 in El Raghama NagÕ El

Ghuneimiya and El Naghaghra were each sampled

once, because these villages had been treated with

insecticides after the 1st night Traps were checked

every 2 h and the catching net replaced when full of

insects Specimens were identiÞed on dry ice

accord-ing to the keys of Gad (1963) and Harbach (1985) Blood-fed mosquitoes were separated by species and location, each group placed in a labeled screwcap vial and stored at ⫺70⬚C until tested for blood meal source

at Ain Shams University

Blood Meal Identification Only mosquitoes

cap-tured in traps were tested Blood meals were identiÞed using the direct enzyme immunoassay (EIA) devel-oped by Beier et al (1988), with slight modiÞcation Brießy, individual mosquitoes were homogenized in phosphate buffered saline Polyvinyl chloride plates (Falcon 3912 microtiter plates, Becton Dickinson, Ox-nard, CA) were sensitized by incubating each mos-quito homogenate in duplicate wells Plates then were washed with PBS containing 0.05% Tween 20, and 50

␮l of host-speciÞc peroxidase conjugate (human, bo-vine, obo-vine, equine, canine, cat, rat, and chicken, Sigma, St Louis, MO), diluted at 1:2000 (or 1:250 for bovine) in boiled casein, were added to each well Plates were incubated, washed and optical densities at

410 nm determined with an enzyme-linked immu-nosorbent assay (ELISA) reader (MR 4000 Dynatech, Alexandria, VA) 15 min after the addition of ABTS peroxidase substrate (Kirkegaard & Perry, Gaithers-burg, MD) Positive controls consisting of mixed unfed mosquitoes and host blood from our sera bank, and negative controls consisting of unfed mosquitoes were included on each microtiter plate To decrease cross reactivity, heterologous serum (8 hosts) was added to the conjugate solution for each test Positive samples were those with absorbance values exceeding mean plus 3 times the standard deviation of the negative controls Multiple feedings were detected by repeat-ing the assay for the whole range of hosts tested The probability of interrupted feeding was determined as

in Burkot et al (1988) and assumed that the proba-bility of interruption was the same for human and nonhuman hosts Only meals containing blood from humans and ovines or bovines were used to evaluate mosquito feeding behavior regarding hosts susceptible

to RVF virus (Ghoneim and Woods 1983)

Results

Prevalence of adult mosquitoes varied among the 5 study villages (Table 1) From 4 to 10 species were

recovered from each village Ae caspius, Cx pipiens,

Cx antennatus, and Cx perexiguus were collected from

all 5 villages where they constituted 98.7% of the 35,289 specimens Of the specimens collected, 664 were blood-fed to repletion Mosquito species each

with ⬍1% prevalence were Cx poicilipes (Theobald), Uranotaenia unguiculata Edwards, Anopheles pharoen-sis Theobald, An tenebrosus Doenitz, An sergentii (Theobald), and An multicolor Cambouliu Two to 4 species were abundant (at least 5%) in each village Ae caspiuspredominated, except at NagÕ El Hagar where

Cx perexiguus was most prevalent Cx pipiens ranked

2nd in prevalence, except at NagÕ El Ghuneimiya

where it was replaced by Cx antennatus The pattern

of mosquito abundance in the study villages followed that observed for species prevalence An average 30.3Ð

Fig 1 Mosquito collection sites in Southern Egypt

(numbered)

17.1 female Ae caspius were collected per trap-night

per village Other predominant mosquitoes were less

abundant: with 5.9Ð58.7 female per trap-night per

vil-lage for Cx pipiens, 0.5Ð43.8 for Cx antennatus and

⬍0.1Ð95.6 for Cx perexiguus Average mosquito

den-sity was lowest at El Naghaghra and highest at NagÕ El

Hagar (Table 1)

Of 664 blood engorged mosquitoes analyzed by EIA,

92.2% reacted to ⱖ1 antiserum (Table 2) and 14.9% of

these reacted to ⱖ2 antisera Large mammals

(hu-mans, bovines and equids) constituted most hosts

identiÞed for the predominant mosquitoes (Table 2)

Ae caspius fed frequently on human, bovine and

equine hosts, although this pattern varied among

vil-lages (Table 3) (␹2⫽ 46.7, df ⫽ 16, P ⬍ 0.01) Frequent

human feeds by this mosquito were observed in

Raghama and NagÕ El Hagar In Sabil Abu El Magd

females fed predominantly on large mammals Ovine

feeds by this mosquito were more frequent in

Raghama than in the other 2 villages Cx pipiens was

mainly anthropophagic (Table 3) The proportional

meals consisting of human, bovine, equine and ovine

blood did not vary among villages for this species (␹2⫽

10.1, df ⫽ 6, P ⬎ 0.05) Cx antennatus and Cx

perex-iguusfed mostly on bovines (Table 2) Ovine feeds by

predominant mosquitoes were 2.8% Dog, rat, cat and

chicken feeds amounted to 3.4% for Ae caspius, 5.4%

for Cx pipiens, 8.3% for Cx antennatus, and 9.0% for Cx.

perexiguus.All 4 mosquitoes fed on unidentiÞed hosts

Of 28 An tenebrosus, 89.3 and 10.7% fed on humans and

bovines, respectively Two human and 1 equine feeds

were detected for 3 An pharoensis One An multicolor

fed on horse blood, whereas 1 unidentiÞed blood meal

was from U unguiculata.

Mixed meals were recognized in the 4 predominant

mosquitoes from all 5 villages, and also in An

tenebro-sus (Table 4) Of 99 mixed blood meals, 91% were

double feeds and 9% were triple feeds Mosquito feeds containing blood from humans and RVF virus suscep-tible animals (bovines or ovines) constituted 28.3% of multiple feeds and 4.2% of all blood meals Such feeds

were 35.3 and 14.8% of mixed meals by Ae caspius from NagÕ El Hagar (n ⫽ 17) and Sabil Abu El Magd (n ⫽ 27), respectively, and 40.7% of those by Cx pipiens from NagÕ El Hagar (n ⫽ 27) Mosquito feeds

con-taining human blood and blood from nonsusceptible RVF virus animals were 26.3 and 3.9% of multiple and total meals, respectively Such mixed blood meal

groups constituted 58.8 and 3.7% of Ae caspius meals

from NagÕ El Hagar and Sabil Abu El Magd

respec-tively, and 22.2% of Cx pipiens meals from NagÕ El

Hagar Overall, the probability of a blood meal being interrupted was 0.10 In villages where ⬎100 meals were tested, the proportion of mixed meals estimated

the probability of interruption of a meal by Ae caspius

(Table 5) The predicted probability for interrupted meals by this species was 0.096 in NagÕ El Hagar and 0.043 in Sabil Abu El Magd

Discussion

Of 14 mosquito species known to occur in the Aswan governorate (Kenawy et al 1987), 10 were recovered from villages where cases of RVF virus infections were reported Serological evidence of in-fections came from blood samples collected during the same period and region from humans and livestock

(Arthur et al 1993) Ae caspius, Cx pipiens, Cx per-exiguus, and Cx antennatus predominated in outdoor

trap collections No systematic indoor collections

were made However, only Ae caspius (n ⫽ 102) were

found resting in 2 bedrooms at Sabil Abu El Magd

Females (n ⫽ 14) of this species were recovered in 1

bedroom at NagÕ El Hagar as well Other sampling

Table 2 Single blood meal hosts of mosquitoes collected from Aswan governorate, 14 –24 August 1993

Mosquito species n Blood meal hosts (% total)

Human Bovine Ovine Equine Dog Cat Rat Chicken Nonreactor

Table 1 Mosquito species composition determined by CO 2 -baited CDC traps (without light) from rural villages in Aswan governorate,

14 –24 August 1993

Mosquito species

% of total females NagÕ El Ghuneimiya

(18) El Raghama(42) El Naghaghra(7) Sabil Abu El Magd(68) NagÕ El Hagar(89)

Numbers in parentheses represent total trap-nights.

a Including mosquito species with ⬍1% prevalence each (Cx poicilipes, U unguiculata, An pharoensis, An tenebrosus, An sergentii, An.

multicolor).

methods might have produced a different pattern.

Indeed, an earlier study carried out in the Aswan

governorate (Kenawy et al 1987) revealed that

al-though Ae caspius prevailed outdoors, and Cx pipiens

represented 95% of indoor collections

The predominant species fed on large mammalian

hosts including humans, bovines, equines, and to a

much lesser extent, ovines Vector competence

stud-ies of mosquitoes collected in epizootic areas during

the 1993 outbreak (Turell et al 1996) revealed that the

4 predominant species were susceptible to infection

and able to transmit RVF virus We suggest that these

mosquitoes were the chief vectors of RVF virus in the

5 villages studied Earlier reports of mosquito blood

feeding in Egypt were based on precipitin tests and

failed to detect mixed meals (Gad et al 1995) The EIA

currently used revealed that mixed meals including

human and bovine or ovine blood were common in the

predominant species from all 5 villages, particularly

Ae caspius,indicating that a signiÞcant proportion of

the mosquito population in Aswan governorate acted

as a bridge vector of RVF virus between humans and susceptible animals Based on abundance, feeding be-havior, susceptibility to infection, and ability to

trans-mit the virus, Ae caspius may have played an important

role in RVF virus transmission in at least 4 of the villages, serving as the major bridge vector between humans and bovines Furthermore, interrupted feeds between humans and ovines or bovines by this mos-quito indicated frequent host contacts among these

hosts, in villages where it predominated Ae caspius

was suspected of disseminating RVF virus among an-imal populations during the 1977Ð1978 epidemics (Kenawy et al 1987) In the Aswan governorate, it is associated closely with sugar cane plantations In NagÕ

El Hagar, where there were no sugarcane Þelds when

collections were made, Cx perexiguus predominated,

feeding mainly on bovines, but also on humans, and therefore acting as the primary link between animals

and humans in that village Cx pipiens was the 2nd

most abundant mosquito, except at NagÕ El Ghunei-miya Its marked anthropophagy indicated that it may have been the main vector of RVF virus among hu-mans Earlier studies reported similar observations for

Cx pipiensfrom the Aswan (Kenawy et al 1987) and Sharqiya (Gad et al 1995) governorates Zoophagy was reported for females from Gharbiya governorate (Zimmerman et al 1985) In the Faiyum Oasis, this species exhibited an elevated forage ratio for bovines and ovines (Beier et al 1986) Much of the geographic variation in host-feeding by mosquito species in Egypt has been attributed to relative host abundance, which

is largely a reßection of ecological conditions and human customs (Zimmerman et al 1985, Beier et al 1987) Because ovines were abundant in Sabil Abu El Magd and NagÕ El Hagar, the low number of ovine feeds could not be related to low sheep/goat

popu-lations in these villages Cx antennatus constituted

approximately one-third of collected mosquitoes and fed mainly on bovines, as well as humans and other large mammals This mosquito may have been the main link vector for RVF virus between humans and animals in Sharqiya (Gad et al 1995) Interestingly,

⬇42% of Cx antennatus feeds were unidentiÞed In the

Nile Delta, this mosquito fed exclusively on large mammals (Zimmerman et al 1985), with a marked preference for ovines (Gad et al 1995) Its host range

is wider in other parts of Africa, where it appears to feed on large mammals as well as birds (Chandler et

al 1975, 1976) However, few Cx antennatus fed on

Table 4 Multiple blood meal sources of mosquitoes in Aswan

governorate, 14 –24 August 1993

Blood meal

% mosquito blood meal hosts

Ae.

caspius pipiens Cx. perexiguus Cx. speciesAll Human/Bovine 14.6 14.8 10.0 15.2

Human/Equine 20.8 14.8 0.0 15.2

Human/Ovine 4.1 17.7 0.0 8.1

Human/Chicken 0.0 2.9 0.0 1.0

Bovine/Equine 27.1 2.9 30.0 18.2

Bovine/Ovine 2.1 23.6 0.0 9.1

Equine/Ovine 6.3 0.0 20.0 5.1

Human/Bovine/Equine 2.1 0.0 0.0 1.0

Human/Bovine/Ovine 0.0 2.9 0.0 1.0

Human/Equine/Ovine 0.0 2.9 0.0 1.0

Human/Ovine/Dog 0.0 2.9 0.0 1.0

Human/Ovine/Rat 0.0 2.9 0.0 1.0

Bovine/Equine/Ovine 2.1 0.0 0.0 1.0

Bovine/Ovine/Cat 2.1 0.0 0.0 1.0

Bovine/Ovine/Rat 0.0 2.9 10.0 2.0

a Including blood meals by Cx antennatus (1 human/bovine, 2

human/rat, 1 bovine/equine, 1 bovine/dog) and An tenebrosus (1

human/bovine, 1 human/rat).

Table 3 Single blood meal hosts of mosquitoes collected from villages in Aswan governorate, 14 –24 August 1993

Mosquito tested in

village

Blood meal hosts (% total)

n Human Bovine Equine Ovine Dog Cat Rat Chicken Nonreactor

Ae caspius

Sabil Abu El Magd 157 28.0 31.2 32.6 3.2 1.9 0.6 0.6 0.0 1.9

Cx pipiens

Sabil Abu El Magd 15 46.7 33.3 13.3 0.0 0.0 0.0 0.0 0.0 6.7

domestic birds in Lower Egypt and unidentiÞed feeds

by this mosquito did not react with general bird

an-tisera (Gad et al 1995) Therefore, it is doubtful that

most nonreacting blood meals by Cx antennatus were

from wild birds Our Þndings demonstrated that the 4

dominant mosquito species feed on a wide variety of

hosts, and thus are quite opportunistic in their feeding

behavior, conÞrming earlier reports (Kenawy et al

1987, Gad et al 1995) This is important because these

mosquitoes may have served as a bridge vector

be-tween humans and domestic animals

The basic transmission cycle of RVF virus in Egypt,

although poorly understood, differs from that in

sub-Saharan Africa where infections recur after

particu-larly wet seasons No true ßoodwater mosquito species

occur in Egypt, where the agricultural land is irrigated

regularly and therefore not exposed to drought

periods Moreover, although Ae caspius lays

drought-resistant eggs (unpublished data), vertical

transmis-sion of RVF virus by this species is questionable,

be-cause no transovarial transmission seems to occur in

aedines (Macintosh and Jupp 1981) or in Cx pipiens

(Turell et al 1984, Romoser et al 1992) Furthermore,

no animal reservoir host has been demonstrated

(Hoogstraal et al 1979) Therefore, it is most likely

that RVF virus does not become established, but

rather has to be reintroduced into Egypt Camels

(Hoogstraal et al 1979) or sheep (Gad et al 1986)

smuggled from Sudan are thought to have introduced

the virus into Egypt in 1977, and the same scenario

may have been repeated

Several possible modes of RVF virus transmission

have been contemplated Biological transmission of

RVF virus from viremic livestock to vectors to other

vertebrates has been generally admitted High

vire-mias occurred in infected humans and livestock

(Hoogstraal et al 1979, Meegan 1979) and were

suf-Þcient to infect potential vectors (Turell et al 1996)

Therefore, both humans and domestic animals may

have served as an important source of infection

Fur-thermore, because of high titered viremias in

verte-brate hosts, mechanical transmission by mosquitoes

was suspected during the 1977 epizootic/epidemic

(Hoogstraal et al 1979, Meegan and Bailey 1988)

Mechanical transmission has been demonstrated

ex-perimentally in Cx pipiens (Hoch et al 1985), which

were able to transmit RVF virus by probing, even if

they failed to feed (Gargan et al 1983) This,

com-bined with abundance in areas where RVF virus

oc-curs, may extend the importance of interrupted

feed-ing by infected Cx pipiens, because it would increase

its chances of transmitting RVF virus However, no data exist regarding mechanical transmission by other potential vectors, so that interrupted feeding by such mosquitoes only indicated their potential for in-creased transmission Aerosol transmission from ver-tebrates to humans by inhalation of virus during ani-mal slaughtering (Hoogstraal et al 1979) and transmission through contact with infected animal tis-sues (Van Velden et al 1977) have been documented Therefore, as long as humans live in close association with susceptible domestic animals, and without sys-tematic surveillance of livestock introduced through Sudan, RVF virus sporadically will be imported into Egypt This statement is supported by our study, which demonstrated that 12 yr after the 1st RVF virus epidemic, potential vectors that feed on both humans and animals remain available, and could contribute to the dissemination of RVF virus in Egypt

Acknowledgments

This study was supported in part by the Research and Training Center on Vectors of Diseases, Ain Shams Univer-sity

References Cited

Arthur, R R., M S El-Sharkawy, S E Cope, B A Botros, S Oun, J C Morrill, R E Shope, R G Hibbs, M A Darwish, and I.Z.E Imam 1993 Recurrence of Rift

Val-ley fever in Egypt Lancet (Lond.) 342: 1149Ð1150

Beier, J C., M A Kenawy, S El Said, and A I Merdan 1986.

Vector potential of culicine mosquitoes in Faiyum Gov-ernorate, Egypt J Am Mosq Control Assoc 2: 164Ð167

Beier, J C., P V Perkins, R A Wirtz, J Koros, D Diggs, T P Gargan II, and D K Koech 1988 Blood meal

identiÞ-cation by direct enzyme-linked immunosorbent assay

(ELISA), tested on Anopheles (Diptera: Culicidae) in

Kenya J Med Entomol 25: 9Ð16

Beier, J C., J H Zimmerman, M A Kenawy, S El Said, and

M M Abbassy 1987 Host-feeding patterns of the

mos-quito community (Diptera: Culicidae) in two Faiyum Governorate villages, Egypt J Med Entomol 24: 28Ð34

Burkot, T R., P M Graves, R Paru, and M Lagog 1988.

Mixed blood feeding by the malaria vectors in the

Anoph-eles punctatuscomplex (Diptera: Culicidae) J Med En-tomol 25: 205Ð213

Chandler, J A., P.F.L Boreham, R B Highton, and M N Hills 1975 A study of the host selection patterns of the

mosquitoes of the Kisumu area in Kenya Trans R Trop Med Hyg 69: 415Ð425

Table 5. Probability that Ae caspius feeding on humans takes a multiple blood meal

Village No humanmeals human mealsNo mixeda Total

meals Proportion mixedhuman meals

Predicted

Qb IHc

aMeal with human and bovine or ovine blood.

bQ, proportion of meals with only human blood ⫹ (proportion of mixed human meals/2).

cAssuming IH ⫽ IN ⫽ probability of interruption of a human (IH) or nonhuman (IN) blood meal ⫽ proportion of mixed blood meals/2Q (1 Ð Q).

Chandler, J A., R B Highton, and P.F.L Boreham 1976.

Studies on some ornithophilic mosquitoes (Diptera,

Cu-licidae) of the Keno Plain, Kenya Bull Entomol Res 66:

33Ð143

Corwin, A., M Habib, D Watts, J Olson, M Darwish, R.

Hibbs, B Botros, M Kleinosky, R Shope, and M

Kil-patrick 1993 Prevalence of antibody to Rift Valley

fe-ver virus in the Nile rife-ver Delta of Egypt, 13 years after

a major outbreak J Egypt Public Health Assoc 53: 153Ð

162

Gad A M 1963 Insects of medical importance Institute of

Medical Entomology, Ministry of Public Health Dokki,

Cairo, Egypt (in Arabic)

Gad, A M., F M Feinsod, I H Allam, M Eisa, A N Hassan,

B A Soliman, S El Said, and A J Saah 1986 A possible

route for the introduction of Rift Valley fever into Egypt

during 1977 J Trop Med Hyg 89: 233Ð236

Gad, A M., M M Hassan, S El Said, M I Moussa, and O L.

Wood 1987 Rift Valley fever virus transmission by

dif-ferent Egyptian mosquito species Trans R Soc Trop

Med Hyg 81: 694Ð698

Gad, A M., I M Riad, and H A Farid 1995 Host-feeding

patterns of Culex pipiens and Culex antennatus (Diptera:

Culicidae) from a village in Sharqiya Governorate, Egypt

J Med Entomol 32: 573Ð577

Gargan, T P II, C L Bailey, G A Higbee, A Gad, and S El

Said 1983 The effect of laboratory colonization on the

vector-pathogen interactions of Egyptian Culex pipiens

and Rift Valley fever virus Am J Trop Med Hyg 32:

1154Ð1163

Ghoneim, N H., and T G Woods 1983 Rift Valley fever

and its epidemiology in Egypt: a review J Med 14: 55Ð79

Harbach, R E 1985 Pictorial keys to the genera of

mos-quitoes, subgenera of Culex and the species of Culex

(Culex) occurring in southwestern Asia and Egypt, with

a note on the subgeneric placement of Culex deserticola

(Diptera: Culicidae) Mosq Syst 17: 83Ð107

Hoch, A L., T P Gargan, and C L Bailey 1985

Mechan-ical transmission of Rift Valley fever virus by

hematoph-agous Diptera Am J Trop Med Hyg 34: 188Ð193

Hoogstraal, H., J M Meegan, G M Khalil, and F K Adham.

1979 The Rift Valley fever epizootic in Egypt, 1977Ð78.

2 Ecological and entomological studies Trans R Soc Trop Med Hyg 73: 624Ð629

Imam, Z E., R El-Karamany, F Omar, and O El-Kafrawy.

1981 Rift Valley fever in Egypt J Egypt Public Health

Assoc 56: 356Ð383

Kenawy, M A., J C Beier, J H Zimmerman, S El Said, and

M M Abbassy 1987 Host feeding patterns of the

mos-quito community (Diptera: Culicidae) in Aswan Gover-norate, Egypt J Med Entomol 24: 41Ð45

Macintosh, B M., and P G Jupp 1981 Epidemiological

aspects of Rift Valley fever in South Africa with reference

to vectors Control Epidemiol Biostatist 3: 92Ð99

Meegan, J M 1979 The Rift Valley fever epizootic in 1977Ð

78 I Description of the epizootic and virological studies Trans R Soc Trop Med Hyg 73: 618Ð623

Meegan, J M., and C L Bailey 1988 Rift Valley fever, pp.

51Ð75 In T P Monath [ed.], The arboviruses:

epidemi-ology and ecepidemi-ology, vol 4 CRC, Boca Raton, FL

Romoser, W S., M E Faran, C L Bailey, and K Lerdthus-nee 1992 An immunocytochemical study of the

distri-bution of Rift valley fever virus in the mosquito Culex

pipiens.Am J Trop Med Hyg 6: 489Ð501

Turell, M J., T P Gargan, and C L Bailey 1984

Replica-tion and disseminaReplica-tion of Rift Valley fever virus in Culex

pipiens.Am J Trop Med Hyg 33: 176Ð181

Turell M.J., S M Presley, A M Gad, S E Cope, D J Dohm,

J C Morrill, and R R Arthur 1996 Vector competence

of Egyptian mosquitoes for Rift Valley fever virus Am J Trop Med Hyg 54: 136Ð139

WHO 1994 Rift Valley fever WHO Wkly Epidemiol Rec.

69: 74

Van Velden, D.J.J., J D Meyer, J Olivier, J.H.S Gear, and

B Macintosh 1977 Rift Valley fever affecting humans

in South Africa: a clinicopathological study S.A Med J 51: 867Ð871

Zimmerman, J H., H A Hanafi, and M M Abbassy 1985.

Host-feeding patterns of Culex mosquitoes (Diptera:

Cu-licidae) on farms in Gharbiya Governorate, Egypt J Med Entomol 22: 82Ð87

Received for publication 29 January 1998; accepted 8 June 1999.