density-dependent processes in the transmission of human

409

Density-dependent processes in the transmission of humanonchocerciasis relationship between the numbers ofmicrofilariae ingested and successful larval development inthe simuliid vector

M G BASANEZ17 J H F REMME2 E S ALLEY3 O BAIN4 A J SHELLEY5G F MEDLEY6 and R M ANDERSON7

1 Department of Biology Imperial College of Science Technology and Medicine Prince Consort RoadLondon SW7 2BB UK2 UNDPWorld BankWHO Special Programme for Research and Training in Tropical Diseases World HealthOrganization 20 Av Appia CH-1211 Geneva 27 Switzerland3 Onchocerciasis Control Programme in West Africa PO Box 549 Ouagadougou Burkina Faso4 Museum dHistoire Naturelle 61 rue Buff on 75231 Paris Cedex 05 France5 Natural History Museum Cromwell Road London SW7 5BD UK6 Department of Biological Sciences University of Warwick Coventry CV4 7AL UK7 Department of Zoology University of Oxford South Parks Road Oxford OX1 3PS UK

(Received 27 July 1994 revised 7 October 1994 accepted 7 October 1994)

SUMMARY

A previous paper reported that the intake of Onchocerca volvulus microfilariae (mff) by different species of Simulium isessentially proportional to the parasite load in the skin of infected carriers This paper examines the fate of the ingestedmff in susceptible vectors to assess the relationship between parasite intake and infective larval output in blackfly specieswith and without well-developed cibarial armatures Analysis is based on data from 3 onchocerciasis endemic areasGuatemala (S ochraceum sL) West Africa (S damnosum slS sirbanum) and the Amazonian focus between SouthVenezuela and Northern Brazil (S guianense and S oyapockense sL) The data which include published and uneditedinformation collected in the field record experimental studies of parasite uptake by wild flies maintained in captivity untilthe completion of the extrinsic incubation period The relationship between L3 output (measured as the mean numberof successful larvaefly or as the proportion of flies with infective larvae) and average microfilarial intake was stronglynon-linear This non-linearity was best represented by a sigmoid function in case of armed simuliids (S ochraceum sLS oyapockense sL) or by a hyperbolic expression in that of unarmed flies (S damnosum sL S guianense) These resultsare compatible respectively with the patterns of initial facilitation and limitation described in culicid vectors oflymphatic filariases A maximum mean number of 1mdash3 L3fly was observed in all 4 vectors It is concluded that O volvuluslarval development to the infective stage is regulated by density-dependent mechanisms acting at the early phase ofmicrofilarial migration out of the blackflys bloodmeal Damage by the bucco-pharyngeal armature may also be densitydependent A hypothesis based on this density dependence is forwarded to explain initial facilitation so far only recordedin vectors with well-developed cibarial teeth Our results provide quantitative support for the conjecture that chemo-therapy alone is likely to have a greater impact on reducing onchocerciasis transmission in endemic areas where the mainvector has a toothed fore-gut than in foci where the vectors have unarmed cibaria

Key words Onchocerca volvulus simuliid vectors larval development density dependence limitation facilitation

INTRODUCTION

The life-cycle of filarial nematodes consists ofpopulations of parasites within the vertebrate hostand the insect host linked by parasite stagestransmitted in both directions The regulation ofthese parasite populations is mediated by density-dependent mechanisms acting at one or severalpoints of the two-host life cycle (Dietz 1982 Plaisieret al 1990 1991)

Reprint requests to Maria-Gloria Basafiez Departmentof Zoology University of Oxford South Parks RoadOxford OX1 3PS

The demographic dynamics of macroparasitessuch as the filarial worms depends on the magnitudesof the birth death immigration and emigration rateswithin either or both hosts (Anderson amp Gordon1982) Reproduction takes place only in the definitivehost and there is no direct multiplication of parasiteswithin either vertebrate or vector host Densitydependence translates into one or more of these ratesbeing a function of the population density of theappropriate parasite stages (Dietz 1988) Thisfunction may result in decreasing (negative feed-back) or increasing (positive feed-back) rates ofpopulation growth as density rises Although onlynegative feed-back processes are instrumental in

Parasitology (1995) 110 409^27 Copyright copy 1995 Cambridge University Press

httpswwwcambridgeorgcoreterms httpsdoiorg101017S0031182000064751Downloaded from httpswwwcambridgeorgcore University of Basel Library on 11 Jul 2017 at 095109 subject to the Cambridge Core terms of use available at

M G Basanez and others

Table 1 Sources of data used in this work

410

Vector species LocalitySuccessfullarvae

Type ofmean used References

Simulium ochraceum sl

Simulium damnosum sl (savanna spp) West Africa5 damnosum ss and 5 sirbanum

Simulium guianense

Simulium oyapockense sl

Guatemala

West AfricaUpper VoltaCote dlvoire

Southern Ghana

Southern Venezuela

Southern Venezuela

Northern Brazil

L3Hmcl or

Thorf mffThor mff and

L3

L3Hmcl orThor mff

Hmcl mff

Thor mff andL3

Thor mff andT 7

Thor larvaejand L3

WMSQRTMsect

SQRTMSQRTM

WMAMsectAMsectAMAM

SQRTMsectSQRTMsectSQRTMsect

SQRTMsect

De Leon amp Duke (1966)Bain Durette-Desset ampDe Leon (1974)

Omar amp Garms (1975)Collins et al (1977)

De Leon amp Duke (1966)Bain (1971)Philippon amp Bain (1972)Philippon (1977)OCP (1989)Alley et al (1994)Takaoka et al (1984 ft)This workThis work

Shelley et al (1987)

Haemocoelic microfilariae Calculated from flies dissected between 0-5 and 12 h pe (see text)f Thoracic microfilariae Calculated from flies dissected between 0-5 and 12 h pe (see text) Thoracic larvae computed from flies dissected 24 h pe (see text)sect Means calculated from raw data

constraining population growth positive feed-backmay also play a role in population dynamics byenhancing the chances of otherwise inefficient trans-mission mechanisms (Hairston amp De Meillon 1968)

In the part of the life-cycle taking place in thevector host density dependence may act on theelements of vector competence (Spielman amp Rossig-nol 1985 Dye 1992a) More specifically it mayinfluence the input of microfilariae (mff) into theinsects the survival and development of the larvaethroughout the extrinsic incubation period and theoutput of infective (L3) stages More indirectly itmay act by affecting the survival behaviour etc ofthe vector In the particular case of human oncho-cerciasis a previous paper (Basanez et al 1994)explored the first component of the transmission ofOnchocerca volvulus from the human host to thesimuliid vector ie the microfilarial intake by theflies from the skin of infected carriers Data fromthree different endemic areas in Africa and LatinAmerica showed an almost linear relationship be-tween dermal and ingested parasites in the ranges ofmicrofilarial loads explored However it has beenobserved that the numbers of larvae successfullydeveloping to the infective stage in filariasis vectorsgenerally do not increase in the same proportionalfashion (Jordan amp Goatly 1962 Obiamiwe 1977Rosen 1955) On the contrary in most cases thereis a significant loss of parasites throughout the periodrequired for larval maturation

Among the factors that can affect the survival anddevelopment of the parasites within the vector are (1)

genetically determined host susceptibility or re-fractoriness to infection (Curtis amp Graves 1983) (2)the damage caused to ingested parasites by thecibarialbuccopharyngeal armature of the flies whenit is present (Coluzzi amp Trabucchi 1968 Omar ampGarms 1975 McGreevy et al 1978 Reid 19781994) (3) the dynamics of formation and structure ofthe peritrophic membrane secreted around thebloodmeal (Lewis 1950 1953 Duke amp Lewis 1964Reid amp Lehane 1984 Miller amp Lehane 1993) and(4) the mounting of specific and non-specific insectdefences to parasite invasion (Townson amp Chai-thong 1991 Ham 1992) Some or all of thesefactors may be influenced by the density of thecorresponding parasite stages

The study of the events involved in the passage ofingested mff towards the site of final larval de-velopment in several vectors has led to the rec-ognition of 3 main patterns for which the termslimitation facilitation and proportionality havebeen coined They refer respectively to the fractionof successful larvae relative to the total parasiteintake being a decreasing an increasing or a constantfunction of the number of mff ingested (Bain 19711976 Pichon 1974 Pichon Perrault amp Laigret1974 Prodhon et al 1980 Southgate amp Bryan1992) They correspond to negative feed-back (therate of successful incorporation of larvae within theinsects decreases with parasite intake) positive feed-back (success increases along with density) anddensity independence respectively In these studies successful mff have been defined as those larvae


Onchocerca volvulus larval development in simuliid vectors 411

managing to get out of the bloodmeal into thehaemocoele or into the flys location where matu-ration occurs or actually completing their devel-opment to the infective stage

It has been suggested that the non-linearitiesinherent in the concept of facilitation and lim-itation may have different and important impli-cations for the control of these infections byrespectively creating or not creating a transmissionthreshold below which the parasite will not persist inthe host population (Brengues amp Bain 1972 Pichonet al 1974 Bain 1976 Webber 1991 Southgate ampBryan 1992) Such ideas are somewhat speculativeat present due to the lack of a quantitative model forthe population dynamics of the lymphatic filarialworms in both the definitive and vector hosts (Dye19926 1994 Dye amp Williams 1994) However ifthese patterns are found to be equally applicable tothe relationship O volvulus-Simulium a fasterprogress in testing these hypotheses could be madedue to the more advanced stage of development ofsuch models in human onchocerciasis (Dietz 1982Davies Weidhaas amp Haile 1987 Plaisier et al 1990Anderson amp May 1991 Davies 1993)

This paper investigates the evidence for and natureof density dependence in the second step of theincorporation of O volvulus parasites by the simuliidvectors namely the rate at which mff from aninfected bloodmeal succeed in reaching the flysthoracic muscles where they undergo further de-velopment We examine several parasite-vectorcombinations present in endemic areas of humanonchocerciasis These combinations include vectorswith well-developed cibarial teeth (S ochraceum slfrom Guatemala and S oyapockense sl from theVenezuelan-Brazilian border) as well as specieswithout such an armature (S damnosum sl fromWest Africa and S guianense from South Venezuela)Analyses are based upon published and unpublisheddata sets and on new field studies conducted inLatin America

We present the analyses of three different types ofdata sets First the relationship is explored betweenthe mean numbers of mff ingested and the resultingmean numbers of successful larvae in blackfly specieswith and without a cibarial armature A functionalrelationship between microfilarial input and larvaloutput is thus obtained Second the fraction of fliesharbouring successful larvae for different meansuccessful larval loads is studied in order to obtain anestimate of the degree of larval aggregation amongthese flies Third these estimates of larval ag-gregation are used together with the functionalrelationships calculated previously in order to exam-ine the prevalence of flies carrying successful larvaeas a function of the mean microfilarial intake

This procedure provides a way (1) to compare theoutput of successful larvae (measured as meannumbers per fly and as fractions of potentially

infective flies) for different levels of microfilarialinput in species possessing and not possessingcibarial teeth (2) to investigate the resulting patternsand to compare them with those proposed by thehypotheses of limitation or facilitation and (3) sincethe approximate proportionality between parasiteintake and dermal load has already been demon-strated (Campbell et al 1980 Alley et al 1994Basanez et al 1994) to advance hypotheses about theefficacy of various control measures in areas witheither type of vector

MATERIALS AND METHODS

Sources of data

Table 1 summarizes the sources of data analysed inthis paper and the criteria for success used in thedifferent studies The definition of the categorysuccessful larvae comprises those mff that reachthe haemolymph of the insect on their way to thethoracic muscles (haemocoelic mff) those thatactually penetrate the muscle cells (thoracic larvae)and those that manage to develop to the infectivestage (L3)

Parasitological and entomological procedures

All the results presented here are based on fly-feeding experiments in which samples of wild fliesto the main local vector species were engorged torepletion on subjects harbouring different intensitiesof skin infection and who consequently providedthe flies with different mean microfilarial intakes ofthe local population of O volvulus

Assessment of the number of mff ingested in thehaemocoele and in the thorax of the flies and of theproportion of flies with larvae In the different studiesgroups of wild flies were fed on particular bodyregions of the various subjects participating in theinvestigation in an attempt to reduce heterogeneityin the intakes The results comprise the figuresobtained for different individual carriers andordifferent local parasite densities along the body ofthose carriers The procedures for fly collection anddissection are the same as those described by Basafiezet al (1994) A subsample of the flies fed on eachvolunteer was dissected between 0-5 and 12 h post-engorgement (pe) in order to score the number ofingested parasites (total microfilarial count withinthe body of the insect) microfilarial migration out ofthe bloodmeal (those mff that escaped imprisonmentby the peritrophic membrane and were found eitherin the haemocoelemdashhaemocoelic mffmdashor in the tho-racic muscles of the fly-thoracic mffmdash) and thefraction of flies harbouring successful mff amongthose dissected The period between 0-5 and 12 h pewas chosen because a feature present in the publisheddata sets and in our own observations was that mffstarted to appear in the thorax about 0-5 h after the


M G Basdnez and others 412

Table 2 Successful larvae of Onchocerca volvulus from differentmicrofilarial intakes by Simulium ochraceum sl in Guatemala

Studyparticipants

Patientcode nof

cuC2J

DIaJ

C3J

C4J

C5J

C6J

C7J

C8J

C 9 |

oB2tsect(Rodoiro)

C10|

DII |

B3Jsect(Feliciano)

B1Jsect(Aparicio)

DI-bt

Mean no

No offlies

24

24

69

20

20

15

20

24

24

20

25

134

20

67

14

221

64

of larvaefly

Ingestedmff()

1-70

2-60

900

12-50

24-20

26-90

4310

44-70

5110

82-20

91-41

115-99

117-80

17000

222-57

265-84

39000

Thor orHmcl mff(y)

000

000

mdash

000

017

0-30

017

0-56

0-30

0-99

1 41

1-51

1-76

mdash

1-69

2-43

mdash

No offlies

38

48

68

53

62

37

70

65

53

44

47

99

70

L3larvae(y)

004

002

019

002

017

0-12

017

0-17

0-51

104

1-69

2-07

2-53

Prevalenceof flies withsuccessfullarvae()

0005-26000208

mdashmdash

0001-89

1500806

200013-51200018-5712-509-23

25-0032-08400027-27mdash

39-38

55-005319mdashmdash64-29

76-49

mdashmdash

See Table 1 for type of mean usedf D De Leon amp Duke (1966) B Bain Durette-Desset ampraquoDe Leon (1974)O Omar amp Garms (1975) C Collins et al (1977)Body regions on which flies were fed as follows Back sect Fore-arms

infected meal In those data sets in which micro-filarial intakes were recorded immediately (0 h) afterengorgement (S oyapockense sl from nor thernBrazil) it was too early to detect any larvae outsidethe b loodmeal and therefore the figures shown forthoracic larvae were calculated from different sub -samples of flies dissected at 24 h pe

Fly maintenance and dissection for the assessment ofL3output When infective larval loads and proportion ofinfective flies were to be recorded the remaininginsects collected from each participant were keptalive until the completion of the extrinsic incubationperiod (within 6mdash7 days pe depending on theconditions under which flies were maintained incaptivity) Although the methods of post-prandial

transportation and maintenance of flies were slightlydifferent across studies all shared the commonprocedures of collection of the simuliids in individualtubes generally lined with filter paper provision of asingle bloodmeal at the beginning of the experiment(the putative infected meal) more regular provisionof sugar solution usually with antibiotics andmaintenance under relatively constant conditions oftemperature high humidity and darkness achievedby keeping the insects inside insulated cages or ice-boxes (Figueroa Collins amp Kozek 1977 Takaoka etal 1982) The results presented here correspond tomean worm burdens and percentages of flies withtotal L3 larvae (from head thorax and abdomen)harboured by specimens dissected from 144 h (6days pe) onwards It has been observed that


Onchocerca volvulus larval development in sitnuliid vectors 413

Table 3 Successful larvae of Onchocerca volvulus from differentmicrofilarial intakes by Simulium oyapockense sl in SouthernVenezuelan and Northern Brazilian Amazonas

Studyparticipants

Patientcode nof

Carroay-a4

Parima2-3

Catrimani-a2

Catrimani-b3

Cauame2

Carroay-b3

Mean no

No offlies

41

1821

22

26

33

of larvaefly

Ingestedmff()

4-59

10-9015-70

2509

34-43

172-46

Thoracicmff(y)

010

011mdash

mdash

mdash

2-46

No offlies

4671

2569

1180

1937

62

Thorlarvaeor L3(y)

009005

0000-27

016013

000007

1-41

Prevalenceof flies withsuccessfullarvae o)

4-884-354-235-56

00020-29

9091000

000811

48-4846-77

See Table 1 for type of mean usedBody regions on which flies were fed as followsf 2 Shoulders and Back 3 Iliac zone and Buttocks 4 CalvesCatrimani and Cauame from Shelley et al (1987) Carroay and Parima from thiswork

infective larvae can be recruited to the proboscisfrom any location in the insect during the blood feed(Duke 1973 Philippon 1977 Renz 1987)

Analysis of data

Measures of the average number of larvae in thebloodtneal in the haemocoele or thorax and of infectivestages per fly Due to the various ways mean intakesand mean larval loads per fly are reported in the datasources it was not possible to select a single measureof central tendency The means most widely used inthe published literature were the arithmetic mean(AM) the geometric mean of Williams (WM) andthe square root transformed mean (SQRTM) All ofthem are computed taking into account infected anduninfected specimens The justification and con-ditions for their usage as well as their formulae havebeen described by Basanez et al (1994)

Since the mean numbers of successful larvaeappeared to level off along with increasing micro-filarial intakes in all data sets examined the re-lationship between these two variables was exploredby non-linear regression methods (Quasi-Newtonand Simplex) with weighted least squares estimationof the parameters (weight = no flies in each sub-sample) These analyses were performed with CSSStatisticareg (Complete Statistical System) software

Prevalence versus intensity data The results of thenon-linear estimations were subsequently used to

examine the relationship between the percentage offlies with successful larvae and the intensity of theinitial input of mff The fraction of simuliidsharbouring potential or realized infective stages canbe related to the mean larval burden per fly (y) by thefollowing expression

Proportion of infective flies = 1 mdash 1 +yk (1)

where k varies inversely with the degree of larvalaggregation in the flies As k^-oo the distributiontends to Poisson whilst as k^-0 the distribution ishighly contagious or aggregated (variance gt mean)and can be described in terms of the negativebinomial (Anderson amp May 1985)

The estimates of k obtained by this fittingprocedure can then be used to predict the prevalenceof flies carrying successful larvae as a function of themean microfilarial intake assuming different func-tional relationships between the mean L3 output (y)and the initial average input of mff (x)

Proportion of infective flies = 1 mdash1 +y(x)k~k (2)

where y (x) are the predictions of the different non-linear regression models aforementioned The com-putation of k values and of their asymptotic con-fidence limits from prevalence-intensity data wasperformed by maximum likelihood estimation pro-cedures (Cox amp Hinkley 1974 Guyatt et al 1990)Different assumptions about the relationship be-tween the parameter k and the mean were tested by



Table 4 Successful larvae of Onchocerca volvulus from differentmicrofilarial intakes by Simulium damnosum sl in West Africa

Study participants

Patient code no-f

OCP28OCP37OCP41OCP46OCP56OCP54DD1OCP17OCP55OCP47OCP18OCP57OCP39OCP49OCP36OCP53OCP20OCP34OCP26OCP21OCP33OCP22OCP61OCP29OCP60OCP30OCP58OCP40OCP35PH612

OCP23PH624

OCP31OVP32OCP38OCP59PH2b12

OCP19OCP45PB124

BA112

OCP42OCP43OCP48OCP52OCP44OCP25PB112

PH2b24

OCP27OCP24OCP51PH3b12

PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200

Thor or L3Hmcl mff larvae(y) iy)

000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73

PrevalenceCT T11 C H7itrlUl IHCo W1LI1

successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500

See Table 1 for type of mean usedAll flies engorged upon the legs of the volunteers DD De Leon amp Duke (1966)BA Bain (1971) PB Philippon amp Bain (1972) (Subject Maurice TiembaLeraba raw data provided by Dr Odile Bain) PH Philippon (1977) (Subjects 3band 6 from Samandeni 2b from Leraba) OCP Onchocerciasis Control Pro-gramme (data from Alley et al 1994)f 12 Flies dissected up to 12 h pe 24 Flies dissected at 24 h pe All flies fromOCP dissected between 4 and 10 h pe



Table 5 Successful larvae of Onchocerca volvulus from differentmicrofilarial intakes by Simulium guianense in Southern VenezuelanAmazonas

Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)

000010mdashmdash0-24mdash0-70mdashmdashmdash0-93mdash1091-33111mdash

No offlies

mdash1110mdash55mdash153722mdash46mdashmdashmdash13

L3larvae(y)

mdash0-460-67mdash101mdash1-691-271-04mdash117mdashmdashmdash1-60


0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15

See Table 1 for type of mean usedf Body regions on which flies were fed as follows2 Back 3 Iliac zone 4 Buttocks 5 Calves 6 AnklesDavid-b and David-f from Takaoka et al (19846) All other volunteers from thiswork

means of the likelihood ratio statistic which isapproximately chi-squared distributed with degreesof freedom equal to the difference between thenumber of parameters in the models being tested(Sokal amp Rohlf 1981 Cox amp Oakes 1984 Armitageamp Berry 1987) The fitting procedure was thenrepeated with k as a function of microfilarial intake

RESULTS

Tables 2 3 4 and 5 summarize respectively thedata sets from Guatemala Northern Brazil WestAfrica and South Venezuela

The examination of raw scatter plots of the meannumbers of successful larvae against mean intakesfor the 5 ochraceum sL S damnosum sl andS guianense data sets revealed that in the 3 blackflyspecies the output of successful larvae tended toreach a plateau as the input of mff increased As aconsequence the relationship between mean numberof successful larvae (y) and mean microfilarial intake(x) was explored by fitting the model y(x) =ax +cx) to all 3 data sets This function takes thevalue of zero when no mfT are ingested describes aninitially linear relationship with slope a for verysmall values of x and encompasses a process ofdensity-dependent saturation in the number of larvae

that can successfully develop within the fly (lim-itation) where c is a measure of the strength ofdensity dependence Under the null hypothesis (c =0) the model represents a linear relationship (pro-portionality) between the two variables in questionHowever the initial shape of this relationshipappeared to be non-linear in S ochraceum sl whencompared with that in S1 damnosum sl andS guianense There was a slower and initiallyexponential increase in the numbers of successfullarvae harboured by the Guatemalan simuliid asopposed to a steeper and convex rise in the Africanand Venezuelan flies

In order to test the hypothesis that the initialshape of the relationship between larval output andmicrofilarial input could be related to the presence orabsence of a well-developed cibarial armature in thesimuliid foregut an independent data set describingthe proportion of injured parasites found in blood-meals with varying microfilarial densities taken upby S ochraceum sl (Bain Durette-Desset amp DeLeon 1974) was analysed This proportion wasobserved to decrease with increasing microfilarialintake up to a certain level (approximately up to40) This density-dependent reduction of therupture of ingested parasites by the cibarial teethcould translate into an initially increasing proportionof viable mff capable of migrating out of the abdomen


M G Basdnez and others

1

416

100 150 200Number of ingested mfffly

250 300

Fig 1 Scatter plot of the proportion of unscathedmicrofilariae found in bloodmeals of GuatemalanSimulium ochraceum sl containing increasing numbersof mff The line is the non-linear regression fit obtainedwhen the function z = ltz(l mdash exp( mdash fix)) was applied tothe data points z is the fraction of uninjured parasitesand x the parasite intake Parameter values are shown inTable 6 Data from Bain Durette-Desset amp De Leon(1974) The 95 confidence intervals around theobserved proportions p were calculated using the exactmethod based on the F distribution for small values ofx (px lt 20) and by the normal approximation (p + 1-96Vp(l mdashp)x for larger values of x (px 20) (Armitage ampBerry 1985)

100 200 300Mean No Ingested mfffly

400


400

Table 6 Results of the non-linear model fitted tothe data of Fig 1 describing the fraction (z) ofingested mff that remain unscathed by the cibarialarmature of Simulium ochraceum sl as a functionof parasite intake (x) (Data from Bain Durette-Desset amp De Leon (1974))

Model = a(l-exp (-))

nCorrelationProportion of variance explainedaSE (a)t(n-2)PPSE ()tin - 2 )P

210-79170-62680-607100756803130000000129000363-527300022

z is the expected proportion of intact mff in thebloodmeal containing x ingested parasites a is themaximum fraction of unscathed parasites to be attainedThe parameter is a measure of the sensitivity to densitydependence in the relationship between microfilarial intakeand damage by the cibarial armature

(initial facilitation) that would be compatible withthe pattern observed in the Guatemalan blackfly Asa result the function z = a(l mdashe~^x) describing theexpected fraction of unscathed mff was fitted to thedata of Bain (op cit) plotted as the fraction of mffremaining undamaged by the cibarial armature ontheir way to the stomach of the insects (Fig 1) In

0 100 200 300 400Mean Noingested mfffly

Fig 2 (A) Scatter plot of the mean numbers ofsuccessful larvae (measured as mff escaping fromimprisonment by the peritrophic membrane ordeveloping to the infective stage) versus the meanmicrofilarial intake for Simulium ochraceum sl fromGuatemala The dotted line corresponds to thelimitation model y = ax (I +cx) whilst the solid linecorresponds to the expression y = aa( mdash exp( mdash))x( +cx) where y is the mean number of successfulparasites and x the mean microfilarial intake Theparameter values for the hyperbolic function are a =00159 + 0-0017 c = 00030 +0-0007 ac (saturationlevel) = 5-3352 r = 0-95 The parameter values for thesigmoid equation are presented in Table 7 Data pointsas follows (A) L3 larvae De Leon amp Duke (1966)(A) haemocoelic mff Bain Durette-Desset amp De Leon(1974) (O) thoracic mff Omar amp Garms (1975)( bull ) L3 larvae and ( bull ) thoracic mff Collins et al (1977)(B) Scatter plot of the mean numbers of successfullarvae versus the mean microfilarial intake for Sdamnosum ssS sirbanum from West Africa The fitted



Table 7 Results of non-linear regression and correlation analyses ofthe mean number of successful larvae on mean microfilarial intakes forthree Onchocerca volvulus-Simulium spp combinations

Cibarial armature

Model

nCorrelation

Proportion ofvariance explaineda1

sE (a)tn-2)Pc1

SE (c)t(n-2)Pa = aa1

fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119

Vector species5 damnosum

y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182

1 The sigmoid model corresponding to the S ochraceum sl data was fitted usingthe parameter values for a and already estimated from the data shown inFig 1 and Table 6 a = 0-6071 and fi = 001292 The model for S ochraceum sl was fitted estimating all three parameters a =ax fj and c from the data set shown in Fig 2A and Table 2

this expression is a measure of the degree ofdensity-dependent decrease in the damage to in-gested mff caused by the cibarial armature of thevector and a is the maximum proportion of parasitesleft uninjured as intake increases When = 0 all

model is the equation y = ax +cx) where y and xhave been defined in (A) Parameter values are given inTable 7 (A) L3 larvae De Leon amp Duke (1966) (A)haemocoelic mff Bain (1971) (O) haemocoelic mffPhilippon amp Bain (1972) ( bull ) haemocoelic mffPhilippon (1977) ( bull ) exo-peritrophic mff Alley et al(1994) (C) Scatter plot of the mean numbers ofsuccessful larvae versus mean microfilarial intake for5 guianense from South Venezuela The line correspondsto the expression y = ax( +cx) fitted to the datapoints Parameter values are indicated in Table 7 (A)L3 larvae Takaoka et al (19846) ( bull ) L3 larvae and( bull ) thoracic mff this work

ingested parasites are lesioned (z = 0) When -gtooa constant density-independent proportion of mff isleft unscathed (z = a) The fitting of the model to thedata of Fig 1 provided a value of which was smallbut significantly different from zero (Table 6)suggesting the existence of a certain degree ofprotection against cibarial damage dependent on thedensity of microfilarial intake Since under the nullhypothesis of no facilitation z = a for all values oi xthe existence of initial non-linearity was tested byplotting the 95 confidence limits around theobserved proportions of undamaged mff in Fig 1 Itcan be seen that for the lower values of parasiteintake the confidence intervals do not include theestimated value of a = O6071 (95 CL =0-4489-0-7653) supporting the hypothesis of initialfacilitation The expression for z was thus incor-porated into the hyperbolic function previouslydescribed in order to obtain an alternative model for



1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp

05 1 15 2 25Mean No successful larvaefly

100

001 01 1Mean No successful larvaefly

10

Fig 3 (A) Scatter plot of the percentage of fliesharbouring successful larvae versus the mean larval loadper fly for all simuliid species included in this studyThe line is the maximum likelihood fit applied to thisprevalence versus intensity data set assuming therelationship in equation (1) proportion of potential orrealized infective flies = 1 mdash [1 +yk]~k where y is themean number of successful larvae and k an inversemeasure of the degree of larval aggregation (k = 0-3950see text)- Data points are (A) Simulium ochraceum sL(A) S oyapockense sL ( bull ) S damnosum sL ( bull )bullS guianense (B) Percentage of flies with successfullarvae plotted against the logarithm of the mean numberof larvae per insect Line and data points are asdescribed in (A)

5 ochraceum sL in which y(x) = aa( mdash e px)x(l+cx) The inclusion of a positive feed-backoperating at low densities in otherwise hyperbolicexpressions has led to satisfactory sigmoid fittings ofobserved functional responses in other biologicalsituations (Hassell Lawton amp Beddington 1977)

Two approaches were taken to estimate theparameters of the sigmoid model fitted to theGuatemalan blackfly In the former the values of theparameters a and obtained from the fit applied tothe data shown in Fig 1 were fed into the alternativeequation describing the relationship between suc-cessful and ingested larvae for 5 ochraceum sLwhilst in the latter all the parameters in this equationwere freely estimated from the data set depicted inFig 2A The proportion of the variance explained bythe hyperbolic limitation model was then compared

with that resulting from the sigmoid functionincluding initial facilitation

The scatter plots of the data points correspond-ing to S ochraceum sL S damnosum sL andS guianense and the curves fitted to them are shownrespectively in Fig 2A B and C A visual inspectionof the curves in Fig 2A suggests that the sigmoidfunction confers a better fit to the Guatemalan dataset than the hyperbolic expression The proportionof the variation accounted for by the limitationmodel alone (0-9024) did increase when the ex-pression for facilitation was added to the function(0-9468 in the case of a and being estimated fromFig 1 and 0-9471 when all parameter values werecomputed from Fig 2A) In addition a betterbehaviour of the residuals was observed with thesigmoid function and the fitting of the hyperbolicequation to S ochraceum sL provided an upper limitof 5-34 successful larvae per fly well in excess ofobserved values As a consequence we concentratedon the model with initial facilitation for S ochraceumsl and with limitation alone for 5 damnosum sL andS guianense These models the parameter valuesand associated statistics are presented in Table 7 Itcan be seen that for S ochraceum sL the twoprocedures used to fit the sigmoid model providedvery similar results The 95 CL around theparameter fi were 00052-00205 and 00070-0-0209in each case

The crucial test for detecting density-dependentlimitation of the numbers of larvae that succeed indeveloping within the 3 blackfly species in theframework of the chosen functions depends on theparameter c being significantly greater than zeroThe 95 confidence intervals around this parameterdid not include zero in any of the data sets examinedThe limits were 00063-00330 for S damnosum sland 0-0205-0-1875 for S guianense In the sigmoidmodels fitted to S ochraceum sl the values were0-0070mdash0-0153 when the information derived fromFig 1 was used and 00024-0-0175 when all theparameters of the proposed equation were estimatedfrom the data points in Fig 2A The fittedexpressions predict a maximum number of larvaeequal to ac in the case of S damnosum sl (2-35) andS guianense (1-35) and equal to aocc for S ochraceumsL (3-32)

Fig 3A and B show the percentage of simuliidscarrying successful larvae plotted against the meanlarval burden per fly in arithmetic and logarithmicscales respectively for all blackfly species includedin this study The line is the maximum likelihood fitto these data using the relationship in equation (1)with k independent of the mean larval load (y) Thefit provided an estimate of k = 03950 with 95asymptotic confidence limits = 0-3579-0-4367 Thelikelihood ratio statistic showed no significant dif-ference between the model with constant k and thosewith k as a linear (k(y) = 8 + ltfgty) power (k(y) =



100

100 200 300Mean No ingested mfffly

400

100 T


400

100

001 01 1 10 100Mean No ingested mfffly

1000

Fig 4 (A) Percentage of flies with successful larvaeplotted against mean microfilarial intake in simuliidspecies possessing a well-developed cibarial armatureThe line was obtained by applying equation (2)proportion of infective flies = 1 mdash [1 +y(x)k]~k where y(the mean no of successful larvae) is a function of x(parasite intake) The function is described in Fig 2(A)(solid line) parameter values are those for Simuliumochraceum sl (Table 7) and k value is that of Fig 3(A) S ochraceum sl (A) S oyapockense sl (B)Percentage of flies with successful larvae versus meanmicrofilarial intake in simuliid species with unarmedcibarium Line as described in (A) where y(x) followsthe expression in Fig 2B and C Parameter values arethose for S damnosum sl (Table 7) and k value is thatof Fig 3 ( bull ) S damnosum sl ( bull ) S guianense (C)Percentage of flies with successful larvae plotted againstthe logarithm of the mean parasite intake in armed andunarmed blackflies The logarithmic scale permits abetter appreciation of the difference between both kindsof vectors particularly at low microfilarial intakes

and exponential (k(y) = Se^) function of the meanworm burden per fly so the results of these modelsare not presented A common value of k wascomputed to measure the degree of aggregation ofpotentially or realized infective larvae in both armedand unarmed simuliids Our argument in favour of acommon k was based upon the fact that this parasitestage has already evaded the effect of the cibarialarmature when present and also because a shared kvalue provided a better description of the proportionof flies with successful larvae as a function of themean microfilarial intake per fly (see below) Thelow value of k suggests a high degree of larvalaggregation which was also observed in the dis-tributions fitted to the direct larval counts per insectThe latter did not depart significantly from thenegative binomial when tested with the goodness offit chi-squared test (data not shown) The average kin this case was 0-3334 ranging from 0-1811 to0-5381 in agreement with the value estimated fromequation (1)

The values of k thus obtained were used to fit therelationship between the percentage of infective fliesand the mean microfilarial intake (x) by means ofequation (2) and the sigmoid expression y(x) =altx( mdash e~Px)x(l + cx) for the species with cibarialarmature (5 ochraceum sl and S oyapockense slFig 4A) whilst equation (2) plus the hyperbolicfunction y(x) = ax +cx) were used for the un-armed species (S damnosum sl and S guianenseFig 4B) In the case of the data set combining thespecies with cibarial teeth the parameter values ofy(x) were those of S ochraceum sl For the dataconcerning the species without armature the par-ameter values of 5 damnosum sl were chosen (thefigures estimated for S guianense tended to under-estimate the proportion of infective flies) Fig 4Ccompares on a logarithmic scale the data and fittedlines of the percentage of flies with successful larvaeas the numbers of mff ingested increase in the armedand unarmed simuliids This proportion increasedmore slowly and tended to level off between 50 and60 for flies ingesting more than 150 mff on averagein the former group of flies whilst the increase wasfaster and the maximum (between 40 and 50)reached earlier (intakes of 50 mff upwards) in thelatter group

DISCUSSION

Basanez et al (1994) found little evidence of densitydependence in the acquisition of skin microfilariaeby flies feeding on O volvulus carriers in 3 differentendemic areas where S ochraceum sl S damnosum

Dotted line is as described in (A) solid line as in (B)(A) Simuliids with cibarial armature ( bull ) simuliidswithout armature



sl and 5 guianense were the predominant vectorsThe evidence for density dependence seems to bestronger when the fate of the ingested parasites isexamined in the same blackfly species The meannumbers of larvae succeeding in reaching thethoracic muscles of the flies and undergoing larvaldevelopment to the infective stage reached a plateauof about 1mdash3 larvae per fly in the 3 simuliid speciesThis non-linearity was statistically significant in alldata sets examined Limitation had already beendemonstrated in S damnosum sl (Pichon 1974Bain 1976 Boussinesq 1991) but not studied indetail in any other blackflies Apart from species-specific differences in vector competence the lowersaturation value found for S guianense may be due tothe fact that in general the Venezuelan fliesdissected in order to detect thoracic mff wereexamined probably too early (average time rangingfrom 115 to 511 h pe) with the result of loweredlarval loads being observed The Guatemalan andWest African specimens were dissected between 4and 10 h pe when most if not all microfilarialmigration would have taken place (Laurence 1966Philippon 1977) This conjecture finds some sup-port in the better description of the relationshipbetween the proportion of flies with larvae versusmean intake which was achieved when the parametervalues obtained for the African species were usedinstead of those derived from the Venezuelansimuliid in equation (2)

The numerical investigation of the patterns oflimitation facilitation and proportionality occurringin several vectors of human and animal filariases hasso far been based upon the statistical examination ofthe relationship between the so-called parasiteyield (ratio of successful larvae to microfilarialintake yx) or the inverse of the parasite yield (xy)and the microfilarial intake (x) In species withlimitation yx tends to decrease with x and there is alinearly increasing relationship between xy and xThe opposite is to be expected in species withfacilitation Proportionality translates into either theparasite yield or its inverse being independent of x(Pichon 1974 Prodhon et al 1980 Southgate ampBryan 1992) However since the variable x appearsin both axes the interpretation of the trends followedby the above-mentioned ratios is not necessarilystraightforward on statistical grounds alone (Dr CDye personal communication) Another more gen-eral consideration to be taken into account whendrawing conclusions about the patterns emergingfrom the analyses of variables that constitute meanvalues is that a great deal of the variation in the datasets explored may be concealed ie both the averagenumber of successful larvae and the mean micro-filarial intake per fly are subject to random variation

The hyperbolic function fitted to the African dataset is simpler than the empirical models previouslyused by Alley et al (1994) Pichon (1974) and

Plaisier et al (1991) yet it fits the data well and theestimation of the saturation level (around 2-3 larvaeper fly) is very similar The more complex functionalform of the relationship between microfilarial inputand infective larval output advocated by Plaisier etal (1991) was found however to be necessary toallow those authors to explore the dynamics ofrecrudescence after cessation of vector control andthe impact of ivermectin on transmission at lowdermal parasite loads (Alley et al 1994)

The results of the feeding studies analysed here inwhich a single bloodmeal was provided at thebeginning of the experiment indicate that theaverage numbers of mff that reach the haemocoele orthe thorax of the insects are generally good predictorsof the numbers of L3 larvae to be found in the fliesonce the extrinsic incubation period is completedThis suggests that there is probably little larval lossduring intra-thoracic development in susceptibleflies as has previously been reported by otherauthors (Duke 1962 a Collins et al 1977 Philippon1977) However comparative analyses with naturalinfection rates and larval burdens are necessary Onthe one hand naturally infected females will havehad the opportunity to ingest blood more than once(it has been shown that Onchocerca larval devel-opment is dependent on blood intake Ham amp Gale(1984)) but also will have experienced some larvalloss through feeding (Duke 1973 Philippon 1977Renz 1987) and different survival rates On theother hand these analyses could also throw somelight on the relevance of possible mechanisms ofacquired resistance by the biting fly population innatural settings Until now the presence of suchmechanisms has only been demonstrated in ex-perimental OnchocercamdashSimulium systems based onintra-thoracic inoculation of mff (Ham 1986 1992)although there is an early observation describing thatin wild host-seeking females the presence of aprevious infection in the fly seems to reduce thenumber of parasites that can develop from asubsequent infection (Duke 1968)

The most important regulatory processes directlyaffecting Onchocerca survival and establishmentwithin the flies (ie not considering indirect effectssuch as parasite-induced vector mortality) seem tobe taking place early in the parasitic phase thatoccurs in the vector namely during the migration ofthe mff out of the mid-gut The peritrophic mem-brane which is secreted in response to the ingestionof blood has been considered to act as an effectivebarrier to microfilarial migration in simuliids (Lewis1950 1953 Duke amp Lewis 1964 Bain amp Philippon1969 Reid amp Lehane 1984 Eichner et al 1991)However the evidence is scarce and the mechanismsremain largely unknown Bain et al (1976) haveproposed that density-dependent changes in thethickness and rate of formation of the peritrophicmembrane may explain the limitation observed in



the numbers of ingested parasites that gain access tothe thoracic muscles in savanna members of the 5damnosum complex Other possible explanationsinclude the activation of insect defences such as thelectin-like molecules acting at the level of the mid-gut epithelium that have been claimed to interferewith microfilarial migration in other filaria-culicidsystems (Phiri amp Ham 1990) This activation may bedependent on the density of the microfilariae

The hypothesis that the level of saturation foundin the system OnchocercamdashSimulium is the result ofprocesses acting early upon microfilarial migrationrather than of some crowding effect due to theovercoming of a larval carrying capacity as describedin other filaria-vector combinations such asWuchereriamdashAedes (Rosen 1955) and BrugiamdashMansonia (Wharton 1957 a b) is supported by theresults of experimental injections of mff directly intothe thorax of the flies This procedure permits thecircumvention of the barriers associated with theinfection per os and has resulted in the demonstrationof a nearly proportional relationship between larvaloutput and microfilarial input in intrinsically sus-ceptible flies (Lok et al 1980 Ham amp Bianco 1983Eichnereia 1991)

Although limitation acting at high parasite in-takes was common to the 3 blackfly species studied(expressed in the term 1(1 +cx) of the models fittedand in the statistical significance of the parameter c)the pattern of initial exponential increase in thenumbers of mff reaching the haemocoele or thorax ofthe flies present in the Guatemalan data set is morecompatible with initial facilitation Bain et al(1974) interpreted their data on S ochraceum sLplotted on logarithmic scales on bothgtgt and x axes asevidence of proportionality Our analysis suggeststhat density-dependent damage of the ingested mffby the cibarial armature in S ochraceum sL mayproduce an initial positive feed-back between micro-filarial intake and the number of parasites which areviable and hence capable of migrating out of thebloodmeal

The hypothesis of an association between thepresence of a cibarial armature in the insect host andthe pattern of initial facilitation requires furtherinvestigation in other filaria-vector combinationsSo far the causal mechanism that has been claimedto operate in relation to facilitation involves his-tological changes in the abdominal epithelium (Bainamp Brengues 1972) However it is interesting to notethat facilitation has solely been reported for vectorspossessing cibarial teeth such as Anopheles mos-quitoes of the subgenus Cellia (McGreevy et al 1978)including An gambiae ss An arabiensis and Anfunestus (Brengues amp Bain 1972 Southgate amp Bryan1992) although the statistical evidence for itsdemonstration in some of these cases has recentlybeen questioned (Dye 1994) Table 8 summarizesthe patterns reported in the literature for several

vectormdashfilaria combinations concerning the relation-ship between microfilarial input and larval output Itcan be seen that only limitation or proportionalityhave been described for vectors with poorly de-veloped armatures or totally unarmed cibaria (CulexAedes Mansonia S damnosum sL)

A possible explanation for the decline in theproportion of ingested mff that are injured by thecibarial armature as intake increases is that it may bemore likely for a few mff all to be ruptured by theteeth as they are pumped in with the bloodmealwhilst as more larvae are ingested the ones thatbecome entangled may protect the remainder Asimilar decrease in the fraction of ingested parasitesruptured by the cibarial teeth has been found inWuchereria bancroftimdashAn funestus but not in WbancroftimdashAn gambiae sL (Bryan amp Southgate 1988Bryan McMahon amp Barnes 1990)

Although the proportion of the variance in theGuatemalan data explained by the function incor-porating density-dependent damage by the cibarialarmature of S ochraceum sL is very high we can onlysuggest that this mechanism may contribute to thepattern observed without the exclusion of otherpossible underlying processes More probably thephenomena of facilitation and limitation are theresult of the combination of a variety of factorswhose net effect is still to be elucidated Amongothers these factors include the interaction of theingested mff with the cibarial teeth with theperitrophic membrane clotting agents and digestiveenzymes acting in the bloodmeal cellular reactionsin the epithelium of the stomach and insect defences(Denham amp McGreevy 1977 Townson amp Chai-thong 1991) For example Bryan et al (1990) havereported that not all the undamaged mff of Wbancrofti observed in the bloodmeals of anophelinevectors in East Africa were able to reach the thoracicmuscles of the mosquitoes

The value of the clumping parameter k obtainedfrom the proportion of flies with successful larvaeversus mean larval burden indicates a high degree oflarval aggregation in the insect host The magnitudeof this aggregation (k = 0-395) appears to be greaterthan that computed for ingested mff by means of thesame kind of analysis (k = 0-588-0-632 Basanez et al(1994)) reflecting the fact that not all the flies thatingest mff become subsequently infected with de-veloping and developed larvae This high degree ofcontagion is compatible with an overdisperseddistribution such as the negative binomial which wasfound to be an adequate description of the dis-tribution of the raw larval counts (data not shown)The trend towards a decline in the value of k withincreasing age of infection in the flies has also beenobserved when the negative binomial distributionhas been fitted to the numbers of larvae per fly foundin naturally infected S damnosum sL populations(Cheke Garms amp Kerner 1982 Garms amp Cheke



Table 8 Patterns observed for the relationship between microfilarial input and larval output in severalvector-filaria combinations

Vector spp Parasite spp PatternCibarial

armature in thevector

Reference

Wuchereriabancrofti

Setarialabiatopapillosa

Aedes aegypti

Skrjabinofilariaskrjabini

Dipetalonemadessetae

Aedes W bancroftipolynesiensis

Aedes togoi Brugia malayi

Culexquinquefasciatus W bancrofti

( =C pipiensfatigans)

Simulium O volvulusdamnosum sl

Mansonia divesbonneae Brugia malayi

(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent

Very poorlydeveloped

Absent

Absent

Present

Bain (1971)Brengues amp Bain(1972)Pichon

(1974)

Bain (1971)Bain amp Chabaud

(1974)

Bain (1976)Chabaud et al

(1986)

Bain (1976)

Rosen (1955)Pichon(1974)

Southgate ampBryan (1992)

Jordan amp Goatly(1962)

Pichon(1974)Southgate ampBryan (1992)

Bain (1971)Philippon ampBain (1972)

Pichon(1974)

Wharton(1957)Southgate ampBryan (1992)

Brengues ampBain (1972)Southgate ampBryan (1992)


1985) The opposite a lower degree of aggregationis to be expected in the presence of density-dependent parasite-induced vector mortality (An-derson amp Gordon 1982) However this assumesthat this process operates uniformly at all stages oflarval development lopping the tails of skeweddistributions We have evidence that Onchocerca-induced Simulium mortality acts predominantlyduring the early phases of infection ie during

microfilarial ingestion The aspects of density-dependent mortality of infected vectors and parasiteaggregation in both human and fly hosts will be thesubject of subsequent publications

The non-linear regression models describing lar-val output as a function of microfilarial input in thecases of the simuliids with and without cibarialarmature were used together with this value of k inorder to investigate the relationship between the



fraction of flies with successful larvae and the averagemicrofilarial intake The fits obtained appeared to bereasonable descriptions of the data in both casesThe sigmoid equation derived for the Guatemalanblackfly species seemed adequate to explain the dataof S ochraceum sl and S oyapockense sl whilst thehyperbolic model was suitable for 5 damnosum sland 5 guianense The results show that simuliidswith armed foreguts require higher average parasiteintakes than those by unarmed species in order toacquire equivalent levels of effective infectionmeasured as mean larval burdens per fly and aspercentage of flies with larvae Nevertheless thesesame indicators suggest an increased vectorial ef-ficiency of the armed species over the species withoutarmature at heavier microfilarial intakes in agree-ment with the remarks made by Shelley (19881991) These differences will be further explored inthe light of parasite-induced vector mortalitieselsewhere Elevated fly mortalities following theingestion of large numbers of parasites have beenreported to affect the species without the protectionafforded by the cibarial teeth (Lewis 1953 Duke1962 a 1966 Omar amp Garms 1977 Takaoka et al19846) in contrast to the lower death rates recordedfor species with toothed fore-guts (De Leon amp Duke1966 Collins et al 1977 Takaoka et al 1984aShelley et al 1987) However the amount ofparasite-related vector loss which really takes placein endemic areas will depend on how frequent is theacquisition of high parasite intakes in the field

In conclusion our analyses suggest the existenceof initial facilitation in blackfly species like5 ochraceum sl and S oyapockense sL and confirmthe occurrence of limitation in the savanna mem-bers of raquoS damnosum sl the latter also being foundin S guianense The sigmoid nature of the relation-ships depicted in Figs 2A and 4A is compatible withprevious claims that facilitation may provide athreshold for the interruption of transmission inareas where vector species with this pattern prevail(Pichon et al 1974 Bain 1976 Webber 1991Southgate amp Bryan 1992) Given that microfilarialintakes are roughly proportional to skin loads(Campbell et al 1980 Alley et al 1994 Basanez etal 1994) it is possible to postulate that controlcampaigns based solely on the reduction of theparasite reservoir in the human host are likely to bemore successful in foci where the principal vectorshave cibarial armatures whilst a combination ofboth measures against the vector and against theparasite may be necessary in endemic areas main-tained by blackfly species without such armatures(Shelley 1991 1994) The results of ongoingchemotherapy programmes with the microfilaricideivermectin seem to support this conjecture (Remmeetal 1990 Cupp 1992)

However any speculation on the existence of atransmission threshold must be based on formal

stability analysis of dynamic models that explicitlyincorporate the features described above in theequations for the parasite in the vector host as wellas reasonable assumptions about the processesaffecting the parasite in the human host (Dye 1994Dye amp Williams 1994) The existing onchocerciasismodels have been constructed for the epidemio-logical situations arising in West and Central Africansavanna regions (with density-dependent limitationof the number of parasites developing in the flies(Dietz 1982 Plaisier et al 1990 1991 Alley et al1994) or in forest areas (proportionality is assumedby Davies (1993)) A model for the Latin Americanscenarios is greatly needed and is currently beingdeveloped The practical relevance of theoreticalbreakpoints in the complex life-cycles of helminthparasites has been shown to depend on the nature ofthe frequency distributions of the numbers of wormsper definitive and intermediate host (Macdonald1965 May 1977 Anderson 1980) In the case ofhighly overdispersed distributions the parasite in-tensities at which unstable equilibria occur (belowwhich the system is driven to extinction) may be toolow to have a meaningful epidemiological impact(Anderson amp May 1985)

The non-linearities described above need to beevaluated in the context of most field situationswhere overdispersion translates into only a fewindividuals of the host populations harbouring theparasite intensities at which density-dependent con-straints are likely to play a role This applies to mffin the skin of the human host (Basanez amp Yarzabal1989 Boussinesq 1991) as well as to larvae in thesimuliid vector (Cheke et al 1982 Garms amp Cheke1985 Renz 1987 Wenk 1991) The question ofwhether the events comprising the life-cycle of thenematode in the vector are occurring in parasiteranges where the relationships are practically linearwill be explored in a subsequent publication in thelight of the frequency distributions of worm burdensper human and per vector host

We thank the Director of OCP Dr E M Samba forallowing us to make use of the Asubende (Ghana) data setMGB is indebted to CONICIT of Venezuela (ConsejoNacional de Investigaciones Cientificas y Tecnologicas)grant No SI-1473 for financial support of the field workcarried out among the Venezuelan Yanomami and theBritish CouncilFCO for a research training scholarship atImperial College GFM is a Royal Society UniversityResearch Fellow RMA thanks the Wellcome Trust forfinancial support Professor K Dietz Dr C Dye Dr B ASouthgate and Mr J Williams contributed with veryhelpful discussions Dr Brian Williams provided statisticaladvice We also thank Dr J H Bryan for permitting us toexamine her raw data on microfilarial damage by Anophelesgambiae sl and An funestus The assistance of HFrontado N J Villamizar M Bolivar J A Gomez INarbaiza V Park and H Suzuki during the field work inVenezuela and of the field staff of the EpidemiologicalEvaluation Unit and of the Vector Control Unit of OCP isgratefully acknowledged



REFERENCES

ALLEY E S AGOUA H BISSAN Y DOUCOURE K

PLAISIER A P REMME J H F amp QILLEVERE D (1994)

Infectivity of Simulium damnosum ss and S sirbanumin relation to intensity of onchocerciasis infection ofthe human host in Asubende Ghana TropicalMedicine and Parasitology (in the Press)

ANDERSON R M (1980) The dynamics and control ofdirect life cycle helminth parasites Lecture Notes inBiomathematics 39 278-322

ANDERSON R M amp GORDON D M (1982) Processesinfluencing the distribution of parasite numberswithin host populations with special emphasis onparasite-induced host mortalities Parasitology 85373-98

ANDERSON R M amp MAY R M (1985) Helminthinfections of humans mathematical modelspopulation dynamics and control Advances inParasitology 24 1-101

ANDERSON R M amp MAY R M (1991) Infectious Diseases

of Humans Dynamics and Control Oxford OxfordUniversity Press

ARMITAGE p amp BERRY G (1987) Statistical Methods inMedical Research Oxford Blackwell ScientificPublications

BAIN o (1971) Transmission des filarioses Limitationdes passages des microfilaires ingerees verslhemocele du vecteur interpretation Annales deParasitologie Paris) 46 613-31

BAIN o (1976) Traversee de la paroi stomacale duvecteur par les microfilaires techniques detudeutilisees importance epidemiologique Bulletin of theWorld Health Organization 54 397-401

BAIN o amp BRENGUES j (1972) Transmission de lawuchereriose et de la setariose bovine etudehistologique de la traversee de la paroi stomacaledAnopheles gambiae A et dAedes aegypti par lesmicrofilaires Annales de Parasitologie (Paris) 47399-412

BAIN O DURETTE-DESSET M C amp DE LEON R (1974)

Onchocercose au Guatemala lingestion desmicrofilaires par Simulium ochraceum et leur passagedans lhemocele de ce vecteur Annales de Parasitologie(Paris) 49 467-87

BAIN o amp PHILIPPON B (1969) Mecanisme du passagedes microfilaires a travers la paroi stomacale duvecteur son importance dans la transmission delonchocercose Comptes Rendus de V AcademieScientifique de Paris 269 1081-3

BAIN O PHILIPPON B SECHAN Y amp CASSONE J (1976)

Correlation entre le nombre de microfilaires ingereeset lepaisseur de la membrane peritrophique duvecteur dans lonchocercose de savane africaineComptes Rendus de VAcademie Scientifique de Paris283 391-2

BASANEZ M G BOUSSINESQ M PRODHON J FRONTADO

H VILLAMIZAR N J MEDLEY G F amp ANDERSON R M

(1994) Density-dependent processes in thetransmission of human onchocerciasis intensity ofmicrofilariae in the skin and their uptake by thesimuliid host Parasitology 108 115-27

BASANEZ M G amp YARZABAL L (1989) Onchocerciasis inthe Sierra Parima and Upper Orinoco regions

Federal Territory of Amazonas Venezuela InParasitic Diseases Treatment and Control (ed MillerM J amp Love E J) pp 231-56 Boca Raton FLCRC Press

BOUSSINESQ M (1991) Etude epidemiologique delonchocercose en zone de savane camerounaise Effetsdun traitement de masse par linvermectine PhDthesis University of Montpellier II

BRENGUES j amp BAIN o (1972) Passage des microfilairesde lestomac vers Ihemocele du vector dans lescouples Wuchereria bancrofti-Anopheles gambiae A WbancroftimdashAedes aegypti et Setaria labiatopapillosa-Aaegypti Cahiers ORSTOM serie Entomologiemedicale et Parasitologie 10 235mdash49

BRYAN j H MCMAHON p amp BARNES A (1990) Factorsaffecting transmission of Wuchereria bancrofti byanopheline mosquitoes 3 Uptake and damage toingested microfilariae by Anopheles gambiae Anarabiensis An merus and An funestus in East AfricaTransactions of the Royal Society of Tropical Medicineand Hygiene 84 265-8

BRYAN j H amp SOUTHGATE B A (1988) Factors affectingtransmission of Wuchereria bancrofti by anophelinemosquitoes 1 Uptake of microfilariae Transactions ofthe Royal Society of Tropical Medicine and Hygiene 82128-37

CAMPBELL C C COLLINS R C HUONG A Y amp

FIGUEROA-MARROQUIN H (1980) Quantitative aspectsof the infection of Simulium ochraceum by Onchocercavolvulus the relation of skin microfilarial density tovector infection Tropenmedizin und Parasitologie 31475-8

CHABAUD A BAIN O LANDAU I amp PETIT G (1986) La

transmission des parasites par vecteurs hematophagesrichesse des phenomenes adaptatifs La Vie desSciences Comptes rendus serie generale 3 469-84

CHEKE R A GARMS R amp KERNER M (1982) T h e

fecundity of Simulium damnosum sl in northern Togoand infections with Onchocerca spp Annals of TropicalMedicine and Parasitology 76 561-8

COLLINS R C CAMPBELL C C WILTON D P amp NEWTON

L (1977) Quantitative aspects of the infection ofSimulium ochraceum by Onchocerca volvulusTropenmedizin und Parasitologie 28 235mdash43

COLUZZI M amp TRABUCCHI R (1968) Importanza dellarmatura bucco-faringea in Anopheles e Culex inrelazione alle infezioni con Dirofilaria Parassitologia10 47-59

cox D R amp HINKLEY D v (1974) Theoretical StatisticsLondon Chapman and Hall

cox D R amp OAKES D (1984) Analysis of Survival DataLondon Chapman and Hall

CUPP E w (1992) Treatment of onchocerciasis withIvermectin in Central America Parasitology Today 8212-14

CURTIS C F amp GRAVES P M (1983) Genetic variation inthe ability of insects to transmit filariae trypanosomesand malarial parasites Current Topics in VectorResearch 1 31-62

DAVIES j B (1993) Description of a computer model offorest onchocerciasis transmission and its applicationto field scenarios of vector control and chemotherapyAnnals of Tropical Medicine and Parasitology 8741-63



DAVIES J B WEIDHAAS D E amp HAILE D G ( 1 9 8 7 )

Models as aids to understanding onchocerciasis InBlack-flies Ecology Population Management andAnnotated World List (ed Kim K C amp MerritR W) pp 396-407 University Park and LondonPennsylvania State University

DE LEON j amp DUKE B o L (1966) Experimental studieson the transmission of Guatemalan and West Africanstrains of Onchocerca volvulus by Simulium ochraceumS metallicum and 5 callidum Transactions of theRoyal Society of Tropical Medicine and Hygiene 60735-52

DENHAM D A amp McGREEVY P B (1977) Brugianfilariasis epidemiological and experimental studiesAdvances in Parasitology 15 243-309

DIETZ K (1982) The population dynamics ofonchocerciasis In Population Dynamics of InfectiousDiseases (ed Anderson R M) pp 209-41 LondonChapman and Hall

DIETZ K (1988) Density-dependence in parasitetransmission dynamics Parasitology Today 4 91mdash7

DUKE B o L (1962a) Studies on factors influencing thetransmission of onchocerciasis I The survival rate ofSimulium damnosum under laboratory conditions andthe effect upon it of Onchocerca volvulus Annals ofTropical Medicine and Parasitology 56 130-5

DUKE B O L (19626) Studies on factors influencing thetransmission of onchocerciasis II The intake ofOnchocerca volvulus microfilariae by Simuliumdamnosum and the survival of the parasites in the flyunder laboratory conditions Annals of TropicalMedicine and Parasitology 56 255-63

DUKE B o L (1966) Onchocerca-Simulium complexesIII The survival of Simulium damnosum after highintakes of microfilariae of incompatible strains ofOnchocerca volvulus and the survival of the parasite inthe fly Annals of Tropical Medicine and Parasitology60 495-500

DUKE B o L (1968) Studies on factors influencing thetransmission of onchocerciasis V The stages ofOnchocerca volvulus in wild forest Simuliumdamnosum the fate of the parasites in the fly and theage-distribution of the biting population Annals ofTropical Medicine and Parasitology 62 107-16

DUKE B o L (1973) Studies on factors influencing thetransmission of onchocerciasis VIII The escape ofinfective Onchocerca volvulus larvae from feedingforest Simulium damnosum Annals of TropicalMedicine and Parasitology 67 95-9

DUKE B o L amp LEWIS D j (1964) Studies on factorsinfluencing the transmission of onchocerciasis IIIObservations on the effect of the peritrophicmembrane in limiting the development on Onchocercavolvulus microfilariae in Simulium damnosum Annalsof Tropical Medicine and Parasitology 58 83-8

DYE c (1992 a) The analysis of parasite transmission bybloodsucking insects Annual Review of Entomology37 1-19

DYE c (19926) Does facilitation imply a threshold forthe eradication of lymphatic filariasis ParasitologyToday 8 109-10

DYE c (1994) The epidemiological context of vectorcontrol Transactions of the Royal Society of TropicalMedicine and Hygiene 88 132-5

DYE c amp WILLIAMS B G (1994) Non-linearities in thedynamics of indirectly-transmitted infections (or doeshaving a vector make a difference) In Ecology ofInfectious Diseases in Natural Populations (ed DobsonA amp Grenfell B T) Cambridge Isaac NewtonInstitute for Mathematical Sciences (in the Press)

EICHNER M RENZ A WAHL G amp ENYONG P ( 1 9 9 1 )

Development of Onchocerca volvulus microfilariaeinjected into Simulium species from CameroonMedical and Veterinary Entomology 5 293-7

FIGUEROA H COLLINS R C amp KOZEK W J ( 1 9 7 7 ) PoSt -

prandial transportation and maintenance of Simuliumochraceum infected with Onchocerca volvulus AmericanJournal of Tropical Medicine and Hygiene 26 75-9

GARMS R amp CHEKE R A (1985) Infections withOnchocerca volvulus in different members of theSimulium damnosum complex in Togo and BeninZeitschrift fur angewandte Zoologie 72 479-95

GUYATT H L BUNDY D A P MEDLEY G F amp GRENFELL

B T (1990) The relationship between the frequencydistribution of Ascaris lumbricoides and the prevalenceand intensity of infection in human communitiesParasitology 101 139-43

HAIRSTON N G amp DE MEILLON B (1968) On theinefficiency of transmission of Wuchereria bancroftifrom mosquito to human host Bulletin of the WorldHealth Organization 38 935^1

HAM p j (1986) Acquired resistance to Onchocercalienalis infections in Simulium ornatum Meigen andSimulium lineatum Meigen following passive transferof haemolymph from previously infected simuliids(Diptera Simuliidae) Parasitology 92 269-77

HAM P j (1992) Immunity in haemotophagous insectvectors of parasitic infection Advances in DiseaseVector Research 9 101-^9

HAM P j amp BIANCO A E (1983) Screening of someBritish Simuliids for susceptibility to experimentalOnchocerca lienalis infection Zeitschrift furParasitenkunde 69 765-72

HAM P j amp GALE C L (1984) Blood meal enhancedOnchocerca development and its correlation withfecundity in laboratory reared blackflies (DipteraSimuliidae) Tropenmedizin und Parasitologie 35212-16

HASSELL M P LAWTON J H amp BEDDINGTON J R ( 1 9 7 7 )

Sigmoid functional responses by invertebratepredators and parasitoids Journal of Animal Ecology46 249-62

JORDAN P amp GOATLY K D (1962) Bancroftian filariasisin Tanganyika a quantitative study of the uptake fateand development of microfilariae of Wuchereriabancrofti in Culex fatigans Annals of TropicalMedicine and Parasitology 56 173-87

LAURENCE B R (1966) Intake and migration of themicrofilariae of Onchocerca volvulus (Leuckart) inSimulium damnosum Theobald Journal ofHelminthology 40 337-42

LEWIS D j (1950) A peritrophic membrane inSimulium Nature London 165 978

LEWIS D j (1953) Simulium damnosum and its relationto onchocerciasis in the Anglo-Egyptian SudanBulletin of Entomological Research 43 597mdash644

LOK J B CUPP E W BRAIDE E I amp BERNARDO M J

(1980) The development of Onchocerca spp in



Simulium decorum Walker and Simulium pictipesHagen Tropenmedizin und Parasitologie 31 498-506

MACDONALD G (1965) The dynamics of helminthinfections with special reference to schistosomesTransactions of the Royal Society of Tropical Medicineand Hygiene 59 489-506

MAY R M (1977) Togetherness among schistosomes itseffects on the dynamics of the infection MathematicalBiosciences 35 301-43

McGREEVY P B BRYAN J H OOTHUMAN P amp KOLSTRUP

N (1978) On the lethal effects of the cibarial andpharyngeal armatures of mosquitoes on microfilariaeTransactions of the Royal Society of Tropical Medicineand Hygiene 72 361-8

MILLER N amp LEHANE M j (1993) Petritrophicmembranes cell surface molecules and parasitetropisms within arthropods Parasitology Today 945-50

OBIAMIWE B A (1977) Relationship betweenmicrofilarial density the number of microfilariaeingested by mosquitoes and the proportion ofmosquitoes with larvae Annals of Tropical Medicineand Parasitology 71 491-500

OMAR M s amp GARMS R (1975) The fate and migrationof microfilariae of a Guatemalan strain of Onchocercavolvulus in Simulium ochraceum and S metallicum andthe role of the buccopharyneal armature in thedestruction of microfilariae Tropenmedizin undParasitologie 26 183-90

OMAR M s amp GARMS R (1977) Lethal damage toSimulium metallicum following high intakes ofOnchocerca volvulus microfilariae in GuatemalaTropenmedizin und Parasitologie 28 109-19

PHILIPPON B (1977) Etude de la transmissiond Onchocerca volvulus (Leuckart 1893) (NematodaOnchocercidae) par Simulium damnosum Theobald1903 (Diptera Simuliidae) en Afrique tropicaleTravaux et Documents de VORSTOM (Paris) No63

PHIRI j amp HAM P j (1990) Enhanced migration ofBrugia pahangi microfilariae through the mosquitomidgut following iV-acetyl-D-glucosamine ingestionTransactions of the Royal Society of Tropical Medicineand Hygiene 84 462

PICHON G (1974) Relations mathematiques entre lenombre des microfilaires ingerees et le nombre desparasites chez differents vecteurs naturelles ouexperimentaux de filarioses Cahiers ORSTOMserie Entomologie medicale et Parasitologie 12 199-216

PICHON G PERRAULT G amp LAIGRET J (1974)

Rendement parasitaire chez les vecteurs de filariosesBulletin of the World Health Organization 51 517-24

PLAISIER A P VAN OORTMARSSEN G J HABBEMA J D F

REMME J amp ALLEY E S (1990) O N C H O S I M amodel and computer simulation program for thetransmission and control of onchocerciasis ComputerMethods and Programs in Biomedicine 31 43-56

PLAISIER A P VAN OORTMARSSEN G J REMME J ALLEY

E s amp HABBEMA j D F (1991) The risk anddynamics of onchocerciasis recrudescence aftercessation of vector control Bulletin of the WorldHealth Organization 69 169-78

PRODHON J PICHON G RIVIERE F DE JARDIN J GERY

M- DOUE F FAUGERE C amp VERNEUIL M P ( 1 9 8 0 )

Etude quantitative de la reduction parasitairestomacale chez les vecteurs de filarioses CahiersORSTOM serie Entomologie medicale etParasitologie 18 13-25

REID G D F (1978) Cibarial armature of Simuliumvectors of onchocerciasis Transactions of the RoyalSociety of Tropical Medicine and Hygiene 72 438

REID G D F (1994) Structure and function of thecibarial armature in Simuliidae Medical andVeterinary Entomology 8 295-301

REID G D F amp LEHANE M j (1984) Peritrophicmembrane formation in three temperate simuliidsSimulium ornatum S equinum and S lineatum withrespect to the migration of onchocercal microfilariaeAnnals of Tropical Medicine and Parasitology 78527-39

REMME J DE SOLE G DADZIE K Y ALLEY E S BAKER

R H A HABBEMA J D F PLAISIER A P VAN

OORTMARSSEN G j amp SAMBA E M (1990) Large scaleivermectin distribution and its epidemiologicalconsequences Ada Leidensia 59 177-91

RENZ A (1987) Studies on the dynamics of transmissionof onchocerciasis in a Sudan-savanna area of NorthCameroon III Infection rates of the Simuliumvectors and Onchocerca volvulus transmissionpotentials Annals of Tropical Medicine andParasitology 81 239-52

ROSEN L (1955) Observations on the epidemiology ofhuman filariasis in French Oceania American Journalof Hygiene 61 219-48

SHELLEY A j (1988) Vector aspects of the epidemiologyof onchocerciasis in Latin America Annual Review ofEntomology 30 337-66

SHELLEY A j (1991) Simuliidae and the transmissionand control of human onchocerciasis in LatinAmerica Cadernos de Saiide Publica RJ 7 310-27

SHELLEY A j (1994) Factors affecting filarialtransmission by simuliids Advances in Disease VectorResearch 10 183-214

SHELLEY A J LUNA DIAS A P A MORAES M A P amp

PROCUNIER w s (1987) The status of Simuliumoyapockense and S limbatum as vectors of humanonchocerciasis in Brazilian Amazonia Medical andVeterinary Entomology 1 219-34

SOKAL R R amp ROHLF F J (1981) Biometry ThePrinciples and Practice of Statistics in BiologicalResearch New York W H Freeman and Company

SOUTHGATE B A amp BRYAN j H (1992) Factors affectingtransmission of Wuchereria bancrofti by anophelinemosquitoes 4 Facilitation limitation proportionalityand their epidemiological significance Transactions ofthe Royal Society of Tropical Medicine and Hygiene 86523-30

SPIELMAN A amp ROSSIGNOL p A (1985) Insect vectorsIn Tropical and Geographical Medicine (edMahumoud A amp Warren K) pp 172-5 SingaporeMcGraw Hill International Student Edition

TAKAOKA H OCHOA J O JUAREZ E L amp HANSEN K M

(1982) Effects of temperature on development ofOnchocerca volvulus in Simulium ochraceum andlongevity of the simuliid vector Journal ofParasitology 68 478-83

TAKAOKA H SUZUKI H NODA S OCHOA J O amp TADA I



(1984a) The intake migration and development ofOnchocerca volvulus microfilariae in Simuliumhaematopotum in Guatemala Japanese Journal ofSanitary Zoology 35 121-7

TAKAOKA H SUZUKI H NODA S TADA I BASANEZ

M G amp YARZABAL L (19846) Development of

Onchocerca volvulus larvae in Simulium pintoi in theAmazonas region of Venezuela American Journal ofTropical Medicine and Hygiene 33 414mdash19

TOWNSON H amp CHAITHONG u (1991) Mosquito hostinfluences on development of filariae Annals ofTropical Medicine and Parasitology 85 149-63

WEBBER R H (1991) Can anopheline-transmitted

filariasis be eradicated Journal of Tropical Medicineand Hygiene 94 241-4

WENK p (1991) The vector host link in filariasisAnnals of Tropical Medicine and Parasitology 85139-47

WHARTON R H (1957a) Studies on filariasis in Malayaobservations on the development of Wuchereria malayiin Mansonia (Mansonioides) longipalpis Annals ofTropical Medicine and Parasitology 51 278-96

WHARTON R H (19576) Studies on filariasis in Malayathe efficiency of Mansonia longipalpis as anexperimental vector of Wuchereria malayi Annals ofTropical Medicine and Parasitology 51 422mdash39


M G Basanez and others

Table 1 Sources of data used in this work

410

Vector species LocalitySuccessfullarvae

Type ofmean used References

Simulium ochraceum sl

Simulium damnosum sl (savanna spp) West Africa5 damnosum ss and 5 sirbanum

Simulium guianense

Simulium oyapockense sl

Guatemala

West AfricaUpper VoltaCote dlvoire

Southern Ghana

Southern Venezuela

Southern Venezuela

Northern Brazil

L3Hmcl or

Thorf mffThor mff and

L3

L3Hmcl orThor mff

Hmcl mff

Thor mff andL3

Thor mff andT 7

Thor larvaejand L3

WMSQRTMsect

SQRTMSQRTM

WMAMsectAMsectAMAM

SQRTMsectSQRTMsectSQRTMsect

SQRTMsect

De Leon amp Duke (1966)Bain Durette-Desset ampDe Leon (1974)

Omar amp Garms (1975)Collins et al (1977)

De Leon amp Duke (1966)Bain (1971)Philippon amp Bain (1972)Philippon (1977)OCP (1989)Alley et al (1994)Takaoka et al (1984 ft)This workThis work

Shelley et al (1987)

Haemocoelic microfilariae Calculated from flies dissected between 0-5 and 12 h pe (see text)f Thoracic microfilariae Calculated from flies dissected between 0-5 and 12 h pe (see text) Thoracic larvae computed from flies dissected 24 h pe (see text)sect Means calculated from raw data

constraining population growth positive feed-backmay also play a role in population dynamics byenhancing the chances of otherwise inefficient trans-mission mechanisms (Hairston amp De Meillon 1968)

In the part of the life-cycle taking place in thevector host density dependence may act on theelements of vector competence (Spielman amp Rossig-nol 1985 Dye 1992a) More specifically it mayinfluence the input of microfilariae (mff) into theinsects the survival and development of the larvaethroughout the extrinsic incubation period and theoutput of infective (L3) stages More indirectly itmay act by affecting the survival behaviour etc ofthe vector In the particular case of human oncho-cerciasis a previous paper (Basanez et al 1994)explored the first component of the transmission ofOnchocerca volvulus from the human host to thesimuliid vector ie the microfilarial intake by theflies from the skin of infected carriers Data fromthree different endemic areas in Africa and LatinAmerica showed an almost linear relationship be-tween dermal and ingested parasites in the ranges ofmicrofilarial loads explored However it has beenobserved that the numbers of larvae successfullydeveloping to the infective stage in filariasis vectorsgenerally do not increase in the same proportionalfashion (Jordan amp Goatly 1962 Obiamiwe 1977Rosen 1955) On the contrary in most cases thereis a significant loss of parasites throughout the periodrequired for larval maturation

Among the factors that can affect the survival anddevelopment of the parasites within the vector are (1)

genetically determined host susceptibility or re-fractoriness to infection (Curtis amp Graves 1983) (2)the damage caused to ingested parasites by thecibarialbuccopharyngeal armature of the flies whenit is present (Coluzzi amp Trabucchi 1968 Omar ampGarms 1975 McGreevy et al 1978 Reid 19781994) (3) the dynamics of formation and structure ofthe peritrophic membrane secreted around thebloodmeal (Lewis 1950 1953 Duke amp Lewis 1964Reid amp Lehane 1984 Miller amp Lehane 1993) and(4) the mounting of specific and non-specific insectdefences to parasite invasion (Townson amp Chai-thong 1991 Ham 1992) Some or all of thesefactors may be influenced by the density of thecorresponding parasite stages

The study of the events involved in the passage ofingested mff towards the site of final larval de-velopment in several vectors has led to the rec-ognition of 3 main patterns for which the termslimitation facilitation and proportionality havebeen coined They refer respectively to the fractionof successful larvae relative to the total parasiteintake being a decreasing an increasing or a constantfunction of the number of mff ingested (Bain 19711976 Pichon 1974 Pichon Perrault amp Laigret1974 Prodhon et al 1980 Southgate amp Bryan1992) They correspond to negative feed-back (therate of successful incorporation of larvae within theinsects decreases with parasite intake) positive feed-back (success increases along with density) anddensity independence respectively In these studies successful mff have been defined as those larvae










Sources of data








Studyparticipants

Patientcode nof

cuC2J

DIaJ

C3J

C4J

C5J

C6J

C7J

C8J

C 9 |

oB2tsect(Rodoiro)

C10|

DII |

B3Jsect(Feliciano)

B1Jsect(Aparicio)

DI-bt

Mean no

No offlies

24

24

69

20

20

15

20

24

24

20

25

134

20

67

14

221

64

of larvaefly

Ingestedmff()

1-70

2-60

900

12-50

24-20

26-90

4310

44-70

5110

82-20

91-41

115-99

117-80

17000

222-57

265-84

39000

Thor orHmcl mff(y)

000

000

mdash

000

017

0-30

017

0-56

0-30

0-99

1 41

1-51

1-76

mdash

1-69

2-43

mdash

No offlies

38

48

68

53

62

37

70

65

53

44

47

99

70

L3larvae(y)

004

002

019

002

017

0-12

017

0-17

0-51

104

1-69

2-07

2-53


0005-26000208

mdashmdash

0001-89

1500806

200013-51200018-5712-509-23

25-0032-08400027-27mdash

39-38


76-49

mdashmdash








Studyparticipants

Patientcode nof

Carroay-a4

Parima2-3

Catrimani-a2

Catrimani-b3

Cauame2

Carroay-b3

Mean no

No offlies

41

1821

22

26

33

of larvaefly

Ingestedmff()

4-59

10-9015-70

2509

34-43

172-46

Thoracicmff(y)

010

011mdash

mdash

mdash

2-46

No offlies

4671

2569

1180

1937

62

Thorlarvaeor L3(y)

009005

0000-27

016013

000007

1-41


4-884-354-235-56

00020-29

9091000

000811

48-4846-77



Analysis of data













Study participants

Patient code no-f


OCP23PH624


OCP19OCP45PB124

BA112


PH2b24


PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200


000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73


successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500





Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES














































































































































Sources of data








Studyparticipants

Patientcode nof

cuC2J

DIaJ

C3J

C4J

C5J

C6J

C7J

C8J

C 9 |

oB2tsect(Rodoiro)

C10|

DII |

B3Jsect(Feliciano)

B1Jsect(Aparicio)

DI-bt

Mean no

No offlies

24

24

69

20

20

15

20

24

24

20

25

134

20

67

14

221

64

of larvaefly

Ingestedmff()

1-70

2-60

900

12-50

24-20

26-90

4310

44-70

5110

82-20

91-41

115-99

117-80

17000

222-57

265-84

39000

Thor orHmcl mff(y)

000

000

mdash

000

017

0-30

017

0-56

0-30

0-99

1 41

1-51

1-76

mdash

1-69

2-43

mdash

No offlies

38

48

68

53

62

37

70

65

53

44

47

99

70

L3larvae(y)

004

002

019

002

017

0-12

017

0-17

0-51

104

1-69

2-07

2-53


0005-26000208

mdashmdash

0001-89

1500806

200013-51200018-5712-509-23

25-0032-08400027-27mdash

39-38


76-49

mdashmdash








Studyparticipants

Patientcode nof

Carroay-a4

Parima2-3

Catrimani-a2

Catrimani-b3

Cauame2

Carroay-b3

Mean no

No offlies

41

1821

22

26

33

of larvaefly

Ingestedmff()

4-59

10-9015-70

2509

34-43

172-46

Thoracicmff(y)

010

011mdash

mdash

mdash

2-46

No offlies

4671

2569

1180

1937

62

Thorlarvaeor L3(y)

009005

0000-27

016013

000007

1-41


4-884-354-235-56

00020-29

9091000

000811

48-4846-77



Analysis of data













Study participants

Patient code no-f


OCP23PH624


OCP19OCP45PB124

BA112


PH2b24


PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200


000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73


successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500





Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES








































































































































Studyparticipants

Patientcode nof

cuC2J

DIaJ

C3J

C4J

C5J

C6J

C7J

C8J

C 9 |

oB2tsect(Rodoiro)

C10|

DII |

B3Jsect(Feliciano)

B1Jsect(Aparicio)

DI-bt

Mean no

No offlies

24

24

69

20

20

15

20

24

24

20

25

134

20

67

14

221

64

of larvaefly

Ingestedmff()

1-70

2-60

900

12-50

24-20

26-90

4310

44-70

5110

82-20

91-41

115-99

117-80

17000

222-57

265-84

39000

Thor orHmcl mff(y)

000

000

mdash

000

017

0-30

017

0-56

0-30

0-99

1 41

1-51

1-76

mdash

1-69

2-43

mdash

No offlies

38

48

68

53

62

37

70

65

53

44

47

99

70

L3larvae(y)

004

002

019

002

017

0-12

017

0-17

0-51

104

1-69

2-07

2-53


0005-26000208

mdashmdash

0001-89

1500806

200013-51200018-5712-509-23

25-0032-08400027-27mdash

39-38


76-49

mdashmdash








Studyparticipants

Patientcode nof

Carroay-a4

Parima2-3

Catrimani-a2

Catrimani-b3

Cauame2

Carroay-b3

Mean no

No offlies

41

1821

22

26

33

of larvaefly

Ingestedmff()

4-59

10-9015-70

2509

34-43

172-46

Thoracicmff(y)

010

011mdash

mdash

mdash

2-46

No offlies

4671

2569

1180

1937

62

Thorlarvaeor L3(y)

009005

0000-27

016013

000007

1-41


4-884-354-235-56

00020-29

9091000

000811

48-4846-77



Analysis of data













Study participants

Patient code no-f


OCP23PH624


OCP19OCP45PB124

BA112


PH2b24


PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200


000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73


successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500





Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES








































































































































Studyparticipants

Patientcode nof

Carroay-a4

Parima2-3

Catrimani-a2

Catrimani-b3

Cauame2

Carroay-b3

Mean no

No offlies

41

1821

22

26

33

of larvaefly

Ingestedmff()

4-59

10-9015-70

2509

34-43

172-46

Thoracicmff(y)

010

011mdash

mdash

mdash

2-46

No offlies

4671

2569

1180

1937

62

Thorlarvaeor L3(y)

009005

0000-27

016013

000007

1-41


4-884-354-235-56

00020-29

9091000

000811

48-4846-77



Analysis of data













Study participants

Patient code no-f


OCP23PH624


OCP19OCP45PB124

BA112


PH2b24


PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200


000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73


successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500





Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES








































































































































Study participants

Patient code no-f


OCP23PH624


OCP19OCP45PB124

BA112


PH2b24


PH3b24

Mean no

No offlies

5050505050509350505052505050505050505050505050505050505050

172505849505050

3085050

270102

505049495050

462307494949

17740

of larvaefly

Ingestedmff()

0-020-300-480-660-840-901101-821-962-402-482-804-505-246126-507-349-56

10-4212-6213-5414-3214-9215-9416-2016-56171619-2021-8822-3125-76260327-2730-2430-3234-2638-7843-6843-72461050-3450-6458-34591470-5771-5290-3092-42

100-60103-51114-7912016250-6935200


000008004000002012mdash 0-400-140-240-300130-320-260-300-460-340-460-500-440-440-360-340-280-981001060-500-660-680-530-400-620-800-640-861-840-561001 581-811-561121001-411-391121121-391141-571-662-572-391-73


successfullarvae()

000600400000200

1000mdash10001400180011-542400160020002600320030003800300030002200280018005000560042-00300042-00320029-652800310342-86360042-00440025-004400500041-2659-8038004600530655-103800380028-5748-86571448-9851-0239-555500





Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES








































































































































Studyparticipants

Patientcode nof

Coy-11 a6

David-a6

Joonafesi5

Cipriano5

Coy-14

David-b5

Coy-IIb4

Cecilio-a3

Cecilio-b4

Mayuba5

Coy-III4

Cecilio-c5

Coy-IV4

David-c4

David-e2

David-f3

Mean no

No offlies

93010101819179

1310151031312611

of larvaefly

Ingestedmff()

1-614-124-624-788-32

12-7721-54430057-91580183-338500

114-95123-15165-76253-05

Thoracicmff(y)


No offlies


L3larvae(y)



0003-33

1818400016-67400035-2946-6737-8454-55600041-3061-2941-94461546-15



RESULTS







1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES







































































































































1

416


250 300



400


400




210-79170-62680-607100756803130000000129000363-527300022








Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES








































































































































Cibarial armature

Model

nCorrelation


sE (a)tn-2)Pc1


fiasE (a)t(n-2)P

SE 08)t(n-2)Pc2

SE (c)t(n-2)P

S ochraceum sl

Present

aa(-e~^)xy (+cx)

270-97301

0-97322

0-94681

0-94712

00612000797-78170000000112000205-500200000003720012900339000983-4718000200013900034413750000400100000342-720200119


y

540-8740

0-7639

00463000974-75630000000196000672-952100047

sl S guianense

Absent

ax

(1+ex)16

0-8517

0-7255

014060-0452310790007701040003892-672500182







1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES







































































































































1UU

oa| 80

pounda 60

m 40

Ioa 20S

0

A

A A = =

A ^zA^f^

ampT A

~ bull a ~ s

A -

A

amp


100


10










100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES







































































































































100


400

100 T


400

100


1000




DISCUSSION


























Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES




























































































































































Reference

Wuchereriabancrofti


Aedes aegypti









(=M longipalpis)

Anopheles gambiaess

W bancrofti

An arabiensis

An funestus

Limitation

Proportionality

Facilitation

Absent


Absent

Absent

Present


(1974)


(1974)


(1986)

Bain (1976)






Pichon(1974)

















REFERENCES















































































































































REFERENCES




































































































































