_Á..-Medica/ and Veterinary Entom%gy (1992) 6, 261-265

LINDA D. JONES, ELIZABETH HODGSON, TREVOR WILLIAMS,STEVE HIGGS* and P A TRICIA A. NUTT ALL Natural Environmental ResearchCouncil, lnstitute of Virology and Environmental Microbiology, Oxford, and *Colorado State University, College ofVeterinary Medicine and Biomedical Sciences, Department of Microbiology, Fort Collins, Colorado, U.S.A.

Abstract. Tick saliva (or salivary gland extract) potentiates the transmission ofThogoto (THO) virus to uninfected ticks feeding on a non-viraemic guinea-pig.This phenomenom has been named saliva activated transmission (SA T). Toinvestigate the potential of different haematophagous arthropods to mediateSA T, guinea-pigs were infested with uninfected R. appendiculatus Neumannnymphs and inoculated with THO virus and salivary gland extract (SGE)derived from a range of ixodid (metastriate and prostriate) or argasid ticks, ormosquitoes; control guinea-pigs were inoculated with virus alone. Enhancementof,THO virus transmission was observed only when SGE was derived frommetastriate ticks. Comparison with the vector potential of these variousarthropod species revealed that enhancement of THO virus transmission wasspecific for ticks which were competent vectors of the virus. The data indicate acorrelation between vector competence and the ability of haematophagousarthropods to mediate SA T of THO virus.

,,/Key words. Thogoto virus, saliva activated transmission, vector competence,haematophagous arthropods, Aedes aegypti, Argasidae, Ixodidae.

Introduction

Arthropod vectors of arboviruses become infected whenIhey feed on the blood of a viraemic host (W.H.O.. 1985).In addition. virus transmission from infected to uninfectedticks cQ-feeding on an uninfected vertebrate has beenobserved. even though the vertebrate host does not developa detectable viraemia (Jones et al.. 1987). 1nvestigationson the mechanism of non-viraemic transmission índicatethat a factor(s) associated with the salivary glands of ticksand secreted in tick saliva potentiates this novel mode ofarbovirus transmission. hence the terro 'saliva activatedtransmi!Osion' (SAT) (Jones ¿t al.. 19H9b. 1992a).

Experiments on SA T were conducted with Thogoto(~HO) virus. an arbovirus recently classilied in the familyOrthomyxoviridae (Davies et al.. 1986: Francki et al..1991). The virus was originally isolated from a mixed pool of

Boophilus and Rhipicephalus ticks removed from sentinelcattle in Thogoto forest near Nairobi, Kenya (Haig et al.,1965), and has been isolated subsequently from variousixodid ticks: in sub-Saharan Africa from Amblyomma~'ariegatum Fabricius, B.decoloratu.v Koch, B.annulatu.vSay, Hyalomma truncatum Koch, R.e~'ertsi Neumann andR.appendiculatus Neumann: in Egypt from H.anatolicumanoticulum Koch; in Sicily from R.bursa Canestrini &Fanzago: in Portugal from R.sanguineus Latreille: and inIran from H.a.anarolicum (reviewed by Davies et al.,1986). The frequent isolation ofTHO virus from these tickspecies indicates that they are probably competent naturalvectors oí the virus. Experimentally, R.appendiculatusand .4m. ~'ariegarum are proveo biological vectors (Davieset al.. 1986; Jones et al., 1989a) with similar infectionthresholds when feeding on a viraemic host (Davies et al.,1990).

The natural transmission cycle of THO virus involvesvarious species of smalI mammals, with the adult tickstransmitting ¡he virus to larger mammals. including domesticanimals. The virus has been isolated from cattle and

Corrcspondcn~'C: Dr P. A. NuttalI. N .E.R.C. Inslitutc ofVirology and Environrncntal Microbi,)logy. Mansl1cld Road.Oxford OXI 3SR. U.K.

26j

262 Linda D. Jones el al.

camels (Kemp el al., 1973), sheep (Davies el al., 1984) andman (Moore el al., 1975), and neutralizing antibodies havebeen demonstrated in the sera of cattle, sheep, donkeys,buffaJoes, camels, rats and man (Davies el al., 1986). Insheep, the virus causes pyrexia (Haig el al., 1965) andabortions (Davies el al., 1984), and in the two reportedhuman cases, THO virus infection was associated withoptic neuritis and a fatal meningitis (Moore el al., 1975).

The significance of SA T in nature and the relativeimportance of different tick species in the epizootiology ofTHO virus are unknown. To date, potentiation of SAT ofTHO virus has been demonstrated only with salivary glandsderived from partially red female R.appendiculalus, Am.variegalum and B.microplus ticks (Jones el al., 1992b). Inthis paper we assess the potential of salivary glands derivedfrom a range of haematophagous arthropods to mediateSA T and compare this with their vector potential forTHO virus. '

Materials and Methods

Cells and virus. BHK-21 and Yero cell cultures werepropagated in modified Eagle's medium (EMEM) supple-mented with 10% newbor:n bovine serum (NBS). TheSicilian (SiAr 126) isolate oí THO virus (obtained as aninfected suckling mouse brain extract) was plaque clonedin Yero cells and virus stocks derived by passage in BHK-21 cells (Davies el al., 1986).

Ticks. Laboratory colonies of the three-host ixodid tickspecies, R.appendiculalus, Am.hebraeum Koch and Am.cajennense Fabricius. were established by feeding all threestages of R.appendiculalus and the larval and nymphalstages of the Amblyomma tick species on Dunkin Hartleyguinea-pigs. Adult Amblyomma ticks were fed on NewZealand White (NZW) rabbits (Jones el al., 1988). Similarfeeding protocols were undertaken for the maintenance ofH.dromedarii Koch. H.m.rufipes Koch, R.everlsi, lxodesricinus Linnaeus and l.hexagonus Leach. However, thelarval and nymphal stages oí these species were fed onhamsters and the adults on NZW rabbits. During theinterval between feeding, ficks of each species were main-tained in perforated tubes, held inside a desiccator, at atemperature of 21-260<::: and at 85% relative humidity(r.h.). ..-

Colonies of the soft tick species Ornilhodoros maritimusYermeil & Marguet and Argas monolaken.\"is Schwan,Corwin & Brown were maintained as previously described.Briefiy, the larval stage was fed on a hamster and thenymphal and adult stages on a membrane feeding systemwhich consisted of a glass reservoir of 2 - 3 mI volumearound which a circulatory system of warm water (28°C)was constructed. A membrane prepared from a ".-day-oldchick ski n was cut to size and stretched over the lip of thereservoir and secured with a rubb~r bando Defribinatedgoose blood (4ml) was introduced joto the reservoir usinga 5 mI syringe. Ticks were then placed on the membraneand a retaining gauze cap clamped over the reservoir tocontain the ticks (Jones et al., 1988).

Mosquitoes. Aedes aegypti (L.) was maintained in OUTinsectary by standard procedures. Ae.aegypti females werefed on an adult Pathology Oxford (PO) strain mouse (12weeks of age) using standard procedures. The mouse wasanaesthetized with pentabarbitone sodium (at the recom-mended dose) and placed on top of a gauze-covered jarcontaining the unfed mosquitoes. Both unfed and fedmosquitoes were maintained in gauze covered jars at atemperature of 21°C and a r.h. of 55-65%. Healthy adultAnopheles stephensi Giles mosquitoes were kindly suppliedby Dr C. R. Davies, London School of Hygiene andTropical Medicine.

Virus assay. Ticks and mosquitoes were homogenizedindividually in 1 mI and 0.2 mi respectively, of EMEMcontaining 10% NBS and appropriate antibiotics to inhibitbacterial growth. Blood samples were obtained on selecteddays post attachment of ticks or mosquitoes, by cardiacpuncture from anaesthetized animals. Titration of materialderived from blood or ticks or mosquitoes, and plaqueneutralization assays, were undertaken as described byDavies et al. (1986).

Per os infection of ixodid and argasid tick species and themosquito species Ae.aegypti. Hamsters inoculated withTHO virus developed a high titre viraemia of up to8.210g¡o PFU/ml blood, 3 days post-inoculation of THOvirus (Davies et al., 1986). For per os infection, the ticksand mosquitoes were fed on viraemic hamsters. The timingof tick and mosquito application, with respect to virusinoculation of the host, was determined by the tick andmosquito feeding behaviour. The nymphal stages of Am.hebraeum and Am.cajennense were allowed to attach 3-4days prior to inoculation of the halnSters with 5()(XJ PFUTHO virus. R.evertsi, H.m.rufipes. H.dromedarii, J.ricinusand J.hexagonus 1 dar prior to inoculation, and O.mari-timus and adult Ae.aeg}'pti at the time of peak viraemia.i.e. 3 days post inoculation with THO virus.

Assay of SA T factor activity. Salivary glands were dis-sected out from uninfected adult female Am.cajennense,Am.hebraeum, H.dromedarii, H.m.rufipes and R.evertsiticks which had fed for a period of 6 days, J.ricinllS andJ.hexagonus female ticks which had fed for a period of 5days, and from partially replete females of O.maritimusand Ar.monolakell.\'is, bloodfed Ae.aegypti and unfedAn.stephensi. The dissected salivary glands were placed inphosphate buffered saline (PBS) pH 7.2, extracted byhomogenization and low-speed centrifugation, and frozenat -20°C.

Previous studies have demonstrated that there is nosignificant difference in the vector efficiency of nymphs asrecipients of SA T, at least in the case of Am. variegatumand R.appendiculatus (Jones et al.. 1990). Thus, exper-iments were standardized by infesting guinea-pigs (two persalivary gland sample) with approximately fifty uninfectedR.appendiculatus nymphs. Each guinea-pig was inoculatedsubcutaneously with 5()(XJ PFU THO virus mixed withsalivary gland extract SGE (40 f!g protejo per animal)derived from partially fed ticks or mosquitoes; controlguinea-pigs were inoculated with virus alone. For assayof SA T factor activity, recipient nymphs were titrated

Tick saliva activated transmission 263

Replication and transmission of THO virusfor virus 12 days post-engorgement (the time of max-imum virus titre; Davies et al., 1986). Virus transmissionwas measured by the number of uninfected ticks thatacquired virus.

Statistical analysis. Virus titres in ticks which had fed onviraemic hamsters were analysed using Student's t-test(Table 1). No significant difference was observed in virustitres in ticks which red on duplicate hamsters. Thus, al!titres were pooled for each tick species. Further statisticalanalyses were undertaken to determine if a relationshipexisted between virus titres in ticks and viraemic titres inthe hamster (Table 2). Data were analysed by fitting aseries of generalized linear models, using a GLIM program(Baker & Nedler, 1978).

Results

Uptake o[ THO virus in a bloodmeal

To determine the vector potential of different arthropodspecies for THO virus, approximately thirty nymphs of theticks Am.hebraeum, Am.cajennense, R.evertsi, H.m.rufipes,H.dromedarii, l.hexagonous, l.ricinus and O.maritimus.and adult female Ae.aegypti mosquitoes, were red onviraemic hamsters. Vectorcompeten~e of Ar.monolakensisand An.stephen.\'i was not assessed due to the limitednumbers of ticks and mosquitoes which were available fortesting. Blood titres of 6.7-8.310g¡o PFU/ml blood wererecorded on day 3 post-inoculation of hamsters. Virus wasdetected in a high proportion of nymphs, when assayedfor virus immediatcly alter feeding (Table 1). Statisticalanalysis of virus titres in ticks following engorgementrevealed no significant differences between species exceptfor l.ricinus nymphs which had lower titres than themajority of the other ixodid tick species: O.maritimustitres were not tested due to the limited number of ticks.

Transtadial persistence and virus transmission for theabove tick species was assessed. Virus transmission wasexamined by placing groups of ten adults (with an equalsex ratio) on sixteen hamsters, 42 days following per osinfection at the nymphal stage (Table 2). Viraemia wasdetected in all of the hamsters on which R.evertsi, Am.cajennense, Am.hebraeum, H.m.rufipes and H.dromedariiadult ticks fed (H 16-25; viraemic hamsters were founddead or sick and killed on days 4-8 following tick at-tachment). Maximum blood titres ranged from 6.5 to8.310g10 PFU Iml blood, and were highest on the day beforedeath; no significant relationship was observed betweenvirus titres in ticks and viraemia titres in the hamsters.

Thogoto virus transmission was not observed whenl.ricinus, I.hexagonus or O.maritimus adults were fed onhamsters (Table 2; H 26-30). Following engorgementnone ofthe ticks contained virus, and there was no evidenceof viraemia «I00PFU/ml blood) or seroconversion inthe hamsters.

The experiment was repeated using Ae.aegypti mos-quitoes. Forty uninfected Ae.aegypti fe males were al-lowed to feed on a viraemic hamster (maximum titre8.310g10 PFU Iml blood). Owing tothe limited number ofmosquitoes which fed on the hamster (a total of fourteen),none were assayed for virus (following engorgement).Fourteen days post-feeding, mosquitoes were re-fed on anuninfected hamster (Table 2, H31). Following engorge-mento none of the mosquitoes were found to contain THOvirus «lOPFU/mosquito) and there was no..itvidence ofviraemia or seroconversion in the hamster.

These results indicate that. of the species tested. R.evertsi.Am.hebraeum, Am.cajennense. H.m.rufipes and H.dro-medarii are competent vectors of THO virus.

Table 1. Per os infection of ixodid and argasid tick species with THO virus.

No. infcctcdlno. tcstcd

Mean titreIOgll) PFU/tickt

Titre(PFU/mI blood)* Tick speciesHam~ler

HlH2H3H4H5H6H7H8H9HI0HllH12H13H14H15

7.26.98.07.66.77.68.07.47.98.17.78.16.97.38.0

R.evertsiR.evertsiA.hebraeumA.hebraeumA.cajennenseA.cajewlenseH.m.rufipesH.m.rufipesH.dromedariiH.dromedariiI.ricinusI.ricinus

I.he.tagonusI.he.tagonusO.maritimus

7/108/109/107/106/109/10

lO/lO8/107/109/107/109/106/105/103/5

3.2 (2.1-4.5)3.5 (2.9-4.7)3.4 (3.1-3.8)3.1 (2.5-4.2)3.3 (2.2-3.9)3.8 (3.1-4.9)3.7 (3.2-4.3)4.0 (~.3-5.1)3.6 (2.9-4.2)3.9 (3..1-4.9)3.1 (2.6-3.5)3.3 (2.9-3.6)3.8 (2.9-5.1)3.6 (3.0-4.2)2.9 (2.1-3.4)

* Blood taken on day 3 pt)st inoculation of hamsters on which the indicated

uninfccted tick species red., Individual ticks were as.'!ayed for virus on day O ~)st-engorgemcm. The means

and range were calculated from those ticks found to .:omain virus.

264 Linda D. Jones el al.

Table 2. Transmission of THO virus by selected tick and mosquito species lohamslers. Titres less than 1.0 logIa PFU/ml blood are scored negative: thosethal were moribund are scored positive.

Virus titre (log10 PFU/ml blood) at days:

Species 4 5 6 7 8Hamster

H16H17H18H19H20H21H22H23H24H25H26H27H28H29H3OH31

3.35.74.2

4.67.27.35.68.33.7

6.88.07.33.67.5++

7.6+

6.5

R.evertsiR.evertsiA.cajennenseA.cajennenseA.hebraeumA.hebraeumH.m.rufipesH.m.rufipesH.dromedariiH.dromedariil.ricinusl.ricinus

l.he.tagonusl.he.tagonusO.maritimus

Ae.aegypti

+++

5.9+

-2.9

S.S4.9

S.S

+

+ -

Assay for SA T factor acti~'ity in salivary glands with the control ticks which red on guinea-pigs inoculatedwith virus alone. In contrast, SA T factor activity was notobserved when the inoculum included SGE derived fromJ.ricinus, J.hexagonus. O.maritimus or Ar.monolakensisticks, or the mosquito species, Ae.aegypti or An.stephensi(Table 3; GP 11-21). Virus was not detected in the bloodof any of the guinea-pigs on day 5 post/attachment of ticks(GP 1-24; <10PFU/ml blood).

A mixture of THO virus and SGE derived from eitherpartially fed uninfected female R.evertsi. Am.cajennense.Am.hebraeum. H.m.rufipes or H.dromedarii was inocu-lated into uninfected guinea-pigs infested with uninfectedR.appendiculatus nymphs (Table 3: GP 1-10): controlguinea-pigs were inoculated with virus alone (GP 22-24).Enhancement of virus transmission was observed with allof the above tick species. i.e. a 7-10-fold increase in thenumber of recipient ticks which became infected compared Discussion

The ability of an arthropod to sustain an infection andsubsequently deliver the virus during feeding is a measureof vector competence, 'the combined effect of al! thephysiological and ecological factors of vector, host, patho-gen. and environment that determine the vector status ofa given arthropod population' (McKelvey el al.. 1981).Comparison of the vector potential of a variety of arthropodspecies for THO virus indicated that only metastriateixodid ticks were susceptible to the virus. In general. thenatural vector species of an arbovirus are more competentfor that virus than are potential vector species, which donot overlap in their geographic range with the virus.However. in this study, there was no evidence to suggestdifferences in the vector efficiency of competent tick species.

The feeding preferences of vectors also have significancefor the transmission of arboviruses in nature, in thatdifferent hosts wil! vary in their susceptibility to a givenarbovirus. Furthermore. as in the case of SA T of THOvirus, there is evidence to suggest that the salivary glands(and their secretions) can play an integral role in diseasetransmission with the parasite utilizing the physiologicalactivities of the vector's feeding mechanism to enhance its

Table 3. Thc cffect of SGE dcrivcd from haematophagousarlhropod veclors on Ihe abilily of R.appendiculacus nymph~10 acquirc THO \;rus.

No. infcctedlno. testcd*Guinea-pig % infectedSpccics

GPl-2GP3-4GPS-6GP7-8GP9-10

GPll-12GP13-14GPlS-16GP17-18

GP19GP20-21

GP"..2-24

31/6036/7025/6040/7029/50

2/605/594/403/40

2/205/SO

5/SO

52

51

42

57

58

3

8

10.,

R.evertsiA.cujewlenseA.hebrueumH.m.rufipesH.dromedurii

I.ricinus

l.he.t/lgonusO.muririmusAr. mono/tlkensis

An.srephensiAe.aegypri

Comrolt

106

6

. Indi\iduals wcre assaycd for virus on day 12 post-c:ngorgcmcnI.t Control guinc:I-pigs wcrc inoculated with virus :lJonc.

Tick saliva activated transmission

own relative infectivity (Jones et al., 1992a). Assessmentof the potential of other haematophagous arthropods tomediate SA T of THO virus indicate that the SA T factorsynthesis is specific to the sa1ivary glands of metastriateticks and is not synthesized in the sa1ivary glands ofprostriate and argasid ticks, or mosquitoes (at least thosespecies tested). Thus, an intriguing correlation has emergedin the apparent relatedness of vector competence and thesynthesis of the SAT factor.

The main difference between metastriate and prostriateticks is in the female 's reproductive physiology (Diehletal., 1982). In addition, ofthe family, Ixodidae, prostriateticks have the most primitive forro of attachment and areconsidered to be interrnediate in comp1exity between thatof argasids and the metastriate ixodids (Kemp, 1982). Theobservation that the SA T factor is not synthesized in thesalivary glands of prostriate ticks mar be related to theirsimpler mechanisms of attachment and feeding.

In order to evaluate ful1y the Tole of prostriate andargasid tick species in the epizootiology of THO virus,these studies need to be extended to tick species thatare sympatric in their geographical distribution with thevirus, e.g. O.moubata Murray and various Ixodes species(Hoogstraal, 1956). Furtherrnore, in view of the potentialimpact of climatic change on the distribution and pre-valence ofvector populations (Sutherst & Maywald, 1985),other tick species - competent vectors of THO virus butnot prevalent in afeas from which the virus has beenisolated - mar in the future be of significance in theepizootiology of the virus.

Acknowledgments

The authors thank Ms Pauline Henbest for the rearing ofthe tick colonies and Professor D. H. L. Bishop forhis suppon.

References

Bakcr. R.J. & Ncdlcr, J.E. (1978) GL/.W s_vslem release 3-generali;:ed linear inleraclive modelling. Royal Statistical Society.London.

Oavics. C.R., Joncs, 1,.:0. & Nuttall, P.A. (1986) Experimentalstudies on the transmission cycle of Thogoto virus. a candidateorthomyxovirus. in Rhipicephalus appendiculalus. AmericanJOIlrnalol Tropical ;Wedicine and Hygiene. 3S, 1256-1262.

Oavics, C.R., Jones, L.O. & Nuttall. P.A. (19W) A comparativestudy oC thc infcction thresholds of Thogoto virus Cor Rhipice-phulus appendiculalus and Alnbl.vommu variegalum. Journal 01Tropical .WeJicine anJ Hygiene. 43. 99-103. .

Oavics, F.G.. Soi, R.K. & Wariru, B.N. (1984) Abortion inshl.'Cp causcd by Thogoto virus. Velerinary Record. liS, 654.

Oichl. P.A.. Acschlimann. A. & Obcnchain, F.O. (1982) Tickrcprodul."tion: oogenl.'Sis and oviposition. Ph.vsiology 01 Ticks,Cu"enr Themes in Tropical S~'ience (ed. by F. O. Obenchainand R. Galun). Vol. l. pp. 277-350. Pergamon Press. OxCord.

Francki, R.I.B., Fauquet, C.M., Knudson, D.L. & Brown, F.(1991) Classification and nomenclature of viruses. Fifth reportof the International Committee on Taxonomy of Viruses.Archives o[ Virology, Supplement 2, 45Opp.

Haig, D.A., WoodaIl, J.P. & Danskin, D. (1965) Thogoto virus:a hitheno undescribed agent isolated from ticks in Kenya.Journal o[ General Microbiology, 38, 389-394.

Hoogstraal, H. (1956) African Ixodidae: Ticks of the Sudan.Bureau o[ Medicine and Surgery Technical Reprint series,Department o[ the Navy, Vol. 1.

Jones, L.D., Davies, C.R., Steele, G.M. & Nuttall, P.A. (1987)A novel mode of arbovirus transmission involving a nonvirae-mic host. Science, 237, 775-777.

Jones, L.D., Davies, C.R., Steele, G.M. & Nuttall, P.A. (1988)The rearing and maintcnance oi ixodid and argasid ticks in thelaboratory. Animal Technology, 39, 99-106.

Jones, L.D., Davies, C.R., Steele, G.M. & Nuttall, P.A. (1989a)Vector capacity of Rhipicephalus appendiculatus and Amb-lyomma variegatum for Thogoto and Dhori virus. Medical andVeterinary Entomology, 3, 195-202.

Jones, L.D., Davies, C.R., Williams, T., Cory, J. & NuttaIl, P.A.(1990) Non-viraemic trnnsmission oi Thogoto virus: vectoreificiency of Rhipicephalus appendiculatus and Amblyommavariegatum. Transactions o[ the Ro.val Society o[ Tropical ,\1edi-cine and Hygiene, 84, 846-848.

Jones, L.D., Hodgson, E. & Nuttall, P.A. (1989b) Enhancementof virus transmission by tick salivary glands. Journal o[ GeneralVirology, 70. 1895-1898.

Jones, L.D., Kaufman, W.R. & Nuttall, P.A. (1992a) Feedingsite modification by tick saliva resulting in enhanced virustransmission. Experientia, in press.

Jones, L.D.. Matthewson, M. & Nuttall, P.A. (l992b) Salivaactivated transmission (SA T) oi Thogoto virus: dynamics oiSA T factor activity in the salivary glands of Rhipicl!phalusappendicularus. Ambi.vomma variegatum and Booeililus micro-plus. Expl!riml!ntal anJ Applil!d Acarology. 13. 1'41- 248.

Kemp, D.H. (1982) Tick attachment and feeding: role of themouthparts, feeding appara[us, salivary gland secre[ions. andthe host immune response. Physiology o[Ticks. Current Thernesin Tropical Science (ed. by F. D. Obenchain and R. Galun),Vol. 1, pp. 119-168. Pergamon Press, Oxford.

Kemp, G.E.. Causey, O.R., Moore, D.L. & O'Connor. E.H.(1973) Viral isolates from livestock in Northern Nigeria: 1964-1970. American Journal o[ Tropical .\ledicine and Hygiene,34, 707- 710.

McKelvey. J.J., Jr. Eldridge. B.F. & Maramorosch. K. (1981) In:Vectors o[ Disease Agents (ed. by J. J. McKelvey. Jr. B. F.Eldridge and K. Maramorosch), p. 243. Praeger. New York.

Moore, D.L., Causey, O.R.. Carey, D.E., Reddy, S., Cooke,A.R., Akinkugbe. F.M., David-West, T.S. & Kemp. G.E.(1975) Anhropod-borne viral infeetions of man in Nigeria.1964-1970. Annals o[ Tropical ,\1edicine and Parasitology.69. 49-64.

Sutherst, R.W. & Maywald, G.F. (1985) A computerised systemfor matching climates in et:ology. Agriculrure, Ecosystems andEnvironment. 13, 281- 299.

World Health Organization (1985) Arthropod-borne and rodent-borne viral diseases. World Health Organi:arion Tl!chnicalReprinr series. Vol. 719. World Health Organization. Geneva.

Acccpted 1 Aprill992