5L4 JounNer, oF THE ArupnrcnN Moseurro Cournor, Assocr.rrror Vor.. 6, No. 3

BASELINE DATA ON AEDE S AEGYPTI POPULATIONS IN SANTODOMINGO, DOMINICAN REPUBLIC1

M. A. TIDWELL,2 D. C. WILLIAMS,3 T. CARVALHO TIDWELL,N C. J. PEIiA,' T. A, GWINN.3D. A. FOCKS,5 A.ZAGLIJL2 ero M. MERCEDESe

ABSTRACT. Baseline field studies were conducted from April 1987 to July 1988 on Aedes aegpti inSanto Domingo, an endemic area for dengue fever. Premise, container and Breieau indices were m6aiuredin one treated area and 2 nearby control areas. These indices averaged 69.6, 46.3 and 142.1, respectively.The principal larval habitats of Ae. aegypti were 208-liter (55-gal) concrete-lined drums. The estimateddaily adult production was approximately 60 per house. Adult mosquito populations were monitoredusing oviposition traps and by sweep net collections. There was no correlationbetween adult abundanceand the larval indices. Monitoring the natural adult densities was more efficient for evaluating theimpact of ULV malathion application than the use of standard bioassay procedures.

INTRODUCTION

In April 1987, studies were begun as part of aUnited States Agency for International De-velopment funded project in collaboration withthe Dominican Republic National MalariaEradication Service to establish baseline dataon Aedes aegypti (Linn.) and to evaluate theimpact of adult and larval control measures inselected areas of Santo Domingo, DominicanRepublic.

Santo Domingo, the capital, encompasses ap-proximately 260 km2 and has a population ofover 1.6 million inhabitants (6a/ha). The cli-mate is tropical with temperatures averagingbetween 22'C and,28"C, and rainfall averaging140-150 cm per year (De la Fuente 1975). Rain-fall is usually heaviest during May-October. Asis common in tropical climates, most houses areopen to the outdoors with windows and doorsunscreened.

Highest human population densities are foundin lower middle and lower class sections of thecity. Because many of these areas lack or areirregularly supplied with piped water, 208-liter(55-gal) drums and other types of containers arecommonly used to store water. These containersserve as larval habitats for Ae. aegypti, the vectorof dengue fever in the Dominican Republic. Pre-

1 Supported by USAID Vector Control Project 517-0235.

2 Vector Control Project AID, Pontificia Universi-dad Catolica Madre y Maestra, Apartado 822, San-tiago, Dominican Republic.

3 University of South Carolina, International Cen-ter for Public Health Research, P.O. Box 699, Mc-Clellanville, SC 29458.

a Leave of absence from Gerencia de Controle deVetores, COMLURB, Rio de Janeiro,Brazil.

5 U.S. Department of Agriculture, Agricultural Re-search Service, Insects Affecting Man and AnimalsResearch Laboratory, P.O. Box 14565, Gainesville, FL32604.

6 Servicio Nacional de Erradicacion de la Malaria,Santo Domingo, Dominican Republic.

vious surveys in the Dominican Republic andHaiti (U.S. Department of Health, Educationand Welfare 1981) reported high endemicity fordengue fever with antibody rates of approxi-mately 70% in children under 10 years of age.In April 1988, approximately 40Vo of the bloodsamples from febrile children in Santo Domingotested positive and all 4 dengue serotypes werepresent in the area (EIIen Koenig, SESPASNational Viral Laboratory, Santo Domingo, Do-minican Republic, personal communication).Since June 1988, there have been 4 cases ofdengrre hemorrhagic fever (DHF) with 2 deathsin the Dominican Republic.

In a study designed to gather baseline data onAe. aegypti abundance in Santo Domingo and toevaluate control measures between 1977 and1980, house (premise) indices ranged from 12 to49% and Breteau indices from 25 to 724 (Mooteand Ganan?). Larval abundance was related tothe number of water-filled containers. Therewas a weak negative correlation between rainfalland larval abundance. In 1985 house (premise)indices as high as 100% were reporbed in someareas adjacent to Santo Domingo (Matute G.8).

Pefra and Zaglul (1986) collected Ae. aegyptithroughout Santo Domingo and stated that itwas the most widely distributed and commonlycollected mosquito (48% of all larvae and adults)in the city. The larvae were taken from allclasses of artificial containers and drains as wellas from natural pools, holes in coralline rock,tree holes and bromeliads.

Aed,es aegypti densities have traditionallybeen estimated indirectly by ovitraps and by theuse of the common indices; premise or house(percentage of houses positive fot Ae. aegypti

7 Moore, C. G. and N. Ganan. Aedes aegypti sur-veillance and control measures in Santo Domingo,Dominican Republic. Poster presentation at 29th an-nual meeting of the Anerican Society of TropicalMedicine and Hygiene. Atlanta; 1980 November 4-7.

8 Matute G., J. V. 1985. Unpublished Pan AmericanHealth Organization report.

SEPTEMBER 1990 An. tscvpn rN THE DoMrNtcAN REpuBLrc 515

larvae), container (percentage of water-filledcontainers that are positive), and Breteau (num-ber of positive containers at 100 premises)(Tinker 1967, Brown 1971, Moore et al. 1978,Chan 1985). The use of these methods for esti-mating populations is generally due to the im-practicality of making time consuming bitingcollections or Iabor intensive resting captures(U.S. Department of Health, Education andWelfare 1979, Tinker 1967).

The principal objectives of this preliminarystudy were: 1) to obtain baseline data on Ae.aegyptipopulation dynamics in Santo Domingo,and 2) to develop a practical adult Ae. aegptimonitoring system. The latter was particularlyimportant in order to have a reliable method forassessing the impact of ultra Iow volume insec-ticide applications and other control methodson the adult mosquitoes. This paper includesbaseline data obtained using traditional meansof estimating Ae. aegypti populations as well assweep net captures of adults inside the houses.

DESCRIPTION OF STUDY SITES

Location of the study sites is shown in Fig. 1.Ensanche Espaillat, a lower middle class barrioencompassingca. 4I ha (100 acres) in the north-east sector of Santo Domingo, was selected asthe principal study site. Ensanche Espaillat isrepresentative of many of the older lower middleclass barrios of the city and, in addition, initial

inspections revealed substantial populations ofAe. aegypti. The estimated population densityin this area was 435 persons per ha. The majorityof the dwellings are single story, cement blockconstruction, although an occasional 2 or 3story structure may be present. The houses usu-ally have 4 or 5 rooms and occupy approximately70 m2. This barrio has an average of 6.3 personsper house and 69 houses per ha. Water is pro-vided to this sector several times a week al-though there may be periods of 1-4 weeks whenthe piped water is not available and water mustbe purchased from tank trucks. The small back-yards of adjoining houses (Fig. 2) usually havefruit trees or various shrubs. Mejoramiento So-cial, a nearby barrio with similar characteristics,was selected as the second study site.

A third area, Gualey, contiguous with E. Es-paillat was selected for a comparative study ofadult densities with selected larval indices. Gu-aley, a lower class barrio adjacent to the RioOzama, is characterized by small (48-m'�)wooden/tin shacks. These are densely packed(approximately 174 per ha) (Fig. 3). Theseshacks are loosely constructed and usually haveone access door, few window openings, openeaves and tin roofs. Approximately 6.2 personsreside per house and a rough estimate of thepopulation density for this area is 1,100 personsper ha. Close to the river, few access roads areavailable; entry into the area is usually by foot.Open sewer ditches cross the area and drain intothe river. This barrio is similar to most of the

Santo Domingo

Fig. 1. Map of Santo Domingo, Dominican Republic, with study site indicated.

SEPTEMBER 1990 As. escvpn Iu tsp DourNIcAN REPUBLIc 5t7

daily basis such as kitchen and bath receptacles,etc., were not included. To determine the levelof mosquito productivity in various breedinghabitats, all water was removed from 15 con-tainers on each of 3 days in 1988 (May 13,August 11 and 17) and the total number ofimmatures was recorded. These containers werelocated on randomly selected premises in E.Espaillat.

Twenty-five of 100 houses were selected atrandom for weekly monitoring of adult densitiesand as sites for l-liter ovitraps (1 per house,filled and maintained from a single watersource). Collections began at approximately0900 h and terminated by noon. Adults werecollected by 2 men working together in tandem,one using a 6-volt, hand-held mechanical aspi-rator and the other a 12-inch sweep net, or bothusing nets. They collected continuously for 5min throughout the house-under tables, chairs,beds, in closets, etc. All male and feamle Ae.aegypti and Culex mosquitoes were recorded.The average number of female Ae. aegypti col-lected during 10 man-minutes (2 men x 5 mineach) at each of the 25 houses represents thefemale density index. This index was then com-pared to the house, container and Breteau in-dices, from all 3 study areas on 3 separate dates.In addition to the female density index, an adultpositive house index was used to indicate thepercentage of houses positive for adult Ae. ae-gypti. Statistical analyses (SAS Institute Inc.1985) were conducted to determine if adult den-sity indices were correlated with larval indices.

Rainfall data for Santo Domingo was providedby the national weather service. Regressionanalyses of the various indices and ovitrap col-Iections on rainfall were conducted.

RESULTS

A summary of the container information forthe 3 study areas is provided in Table 1. Surveysofwater holding containers in the 3 study areasrevealed differences in the types and proportionsof receptacles present. In all 3 areas, essentialwater storage containers-drums, smaller tanks,and to a lesser extent basins and cisterns-werethe most numerous. These represented 68% ofall positive larval habitats for Ae. aegypti. Ce-ment, metal and plastic drums with a holdingcapacity of approximately 76-208 liters (20-55gal) represented 5l% of the potential breedingsites and 55 to 64% of these were positive. Out-door containers were more numerous than thosefound indoors except in the categories of flowervases, buckets and earthen jars. In the poorerbarrio, Gualey, no water basins or cisterns wereencountered. There were, however, greater pro-portions of earthen jars (< 4 Iiters and smallcontainers).

From the 45 containers examined (3 collec-tions of 15 containers each), 28,615 immatureswere recorded: first and second instars repre-sented approximately 44Vo, third and fourth in-starc, 47Vo, and pupae 9%. The pupal densityindex (mean number of. Ae. aegyptli pupae perhouse) was 127 and the larval density index(mean number of larvae per house) was 1,379.

The average number of immatures found in30 drums (unlined and concrete-lined) was 807.However, the highest number of immatures in aconcrete-lined drum was 1,943 and the highestnumber of pupae was 417. The average numberof immatures for 6 containers holding 38 liters(10 gal) or less was 263. The average number ofimmatures from 5 positive tires was 206.

Table 1. Prevalence of container types indoors and outdoors with corresponding percent positivity for Aedesaeglpti immatures in 3 barrios of Santo Domingo, Dominican Republic (April 1987, February and June

1988).,

Flower Smallu"a"a "o"iui"". Buckets Earthen Water. ursterns llres uthers

lars Dasrns

Area (Site) No. % pos No. % pos No. % posNo. % posNo. % posNo. % posNo. % posNo. % pos No. % pos

Drums

Espaillat 630(indoor) 78(outdoor) 552

M. Social 484(indoor 89(outdoor) 395

Gualey 175(indoor) 67(outdoor) 108

Total 1289(indoor) 234(outdoor) 1055

55.138.557.455.45L.756.264.062.764.856.450.457;1

1105

IUO1 n

07 l

000

L275

t22

2362r818

168152163430

43840038

49.6 rr2 44.649.5 4 25.050.0 108 45.438.1 45 60.035.5 5 20.062.5 40 65.044.7 68 26.543.3 33 18.250.0 35 34.344.7 225 42.243.8 42 I9.055.3 183 47.5

98 13.3 1152 3.8 046 23.9 rr53 18.9 521 38.1 432 6.3 14t L2.2 2234 IL.8 227 L4.3 0

192 14.6 38107 13.1 2685 16.5 12

36.4 56 41.10 -

36.4 56 41.140.0 L4 42.925.0 0 -100 14 42.936.4 0 -36.4 0 -

0 -31.6 70 41.434.6 0 -4t.7 70 4t.4

30.0 68 39.7 52 25.040.0 425.0 2 50.029.5 64 40.6 50 24.035.3 29 4r.4 9 11.1

0 - 0 -35.3 29 47.4 I 11.1

I 44.4 7 85.73 0.0 4 75.06 66.7 3 100

30.7 106 38.7 68 29.440.0 425.0 6 66.730.3 102 4t.2 62 25.8

t Tabular data from Espaillat, M. Social and Gualey are the result of 382 premises inspected in 6 surveys,215 premises in 3 surveys, and225 premises in 3 surveys, respectively.

JouRNel oF THE ArranRrclN Mosqumo Coivrnol Assoclluorl V o l . 6 , N o . 3

In the productive, concrete-lined drums thepupae generally appeared to be of normal size.However, in one drum (not included in thesedata), which had the only mixed colony (approx-imately 2,500 immatures) of Aedes and Culzx.the emerged male and female Ae. aegypti wereabnormally small and approximately 10% couldpass through the 16 x 18 mesh of a standardcollapsible mosquito cage.

In a comparison of the house, container andBreteau indices with the average number offemales per house (female density index), onlythe container indices were similar among the Bbarrios on the 3 dates (Fig. a). The female den-sity index increased in all 3 areas and variedfrom a low of 1.72 (Gualey) to a high of 9.43 (E.Espaillat). The adult positive house index re-corded at the same time revealed an increase inthe percentage of positive houses in E. Espaillatfrom 88 to 100% and in M. Social from 96 to100% . In Gualey this percentage, 92Vo , rcmained.constant,

The average number of female Ae. aegyptiperhouse for all collection dates in E. Espaillat was4.0 + 6.8 (range 0-134) with most individualcollections containing 1 or 2 female mosquitoes.With 107 collection dates between April 1987and July 1988, the female density index percollection date varied from 1.22 to 15.04. Thehighest number of male Ae. aegypti recorded fora house in E. Espaillat was 77 with an averageof 3.8 + 6.0 (range 0-77). From 148 house col-Iections in Gualey, the highest number of Ae.aegypti females taken from a single dwelling was16. In some houses in E. Espaillat and M. Social,however, high populations of mosquitoes (>20per collection) were present one or more times.From 2,153 house collections in E. Espaillat andM. Social, 53 samples (2.5% of all samples) from18 houses had >20 female Ae. aegypti. Collec-tions of )20 females were made on 2 or moreoccasions from 9 "mosquitogenic" houses and,in one house, 14 collections were made eachhaving more than 20 mosquitoes. As many as134 females were collected from a single dwell-ing. Culex quinquefasciafus Say were also readilycollected: the highest number of females andmales collected from a single house using sweepnets was 68 and 39, respectively. The averagenumbers of females and males collected perhouse per collection date and area were 3.2 +4.8 (range 0-68) and 3.0 + 4.3 (range 0-39),respectively.

There were no sigrrificant relationships be-tween adult densities and any of the 3 indices(Kendall correlation coefficients for adult den-sities with premise, container and Breteau in-dices were 0.36, 0.07 and 0.14, respectively).

Ovitrap collections were made in E. Espaillatfrom April 1987 to July 1988 (except for July

1987 and March, April 1988). During this periodthe average number of eggs collected per ovitrapper day was 12.9 + 23.5 with daily trap coileC-tions ranging from 0 to 210. During all collec-tions >50% ofpaddles were positive. Statisticalanalyses were conducted to determine ifmonthly average female density indices (1,600samples) were correlated with monthly averagesof the number of eggs per ovitrap per day (1,760samples) and if individual premise female den-sities were correlated with number of eggs perovitrap per day from the same premises (24gsamples). Analyses revealed no significant cor-relations between egg numbers and adult collec-tions either on a monthly average basis or adaily basis (Kendall correlation coefficientswere 0.06 and 0.04, respectively).

Average daily rainfall per month in SantoDomingo ranged from 0.51 mm in July 1988 to14.27 mm in December 1987 (an unusually wetDecember). Regression analysis of the differentimmature indices on rainfall accumulated 8 daysprior to assessing the indices (based on theobserved field development time of the larvaeplus pupae, the number of days of rain whichwould contribute to the presence of immatureAe. aegypti on the date the indices were taken)indicated that rainfall did not explain a signifi-cant amount of the variations in the premise,container or Breteau indices (r' : 0.19, 0.25 and0.14, respectively). Regression of averagemonthly adult density indices on average dailyrainfall per month also indicated that rainfallfailed to explain a significant portion of themonthly variations of adult densities (r'� = 0.02).

DISCUSSION

Drums, tanks and other essential water hold-ing containers were the principal breeding hab-itats in the lower middle class and lower classareas where the water supply is inadequate. Inmiddle and upper class areas other methods ofwater storage are more frequently employedwhen necessary, such as the use of cisterns andclosed pressurized systems. While an occasionalbarrel may be found in these areas, preliminarysurveys revealed that water-filled flower vases,tires and other small, usually nonessential waterholding containers were the most frequently en-countered larval habitats of Ae. aegypti.

The concrete tanks, which are usually metaldrums lined with cement as rust hole leaks de-velop, represent a special problem. In additionto being numerous and uncovered, they are alsomuch heavier than the original metal drums.Consequently, they cannot be dumped easily fordraining or cleaning. Thus they retain at leastsome water and presumably offer a nearly con-tinuous oviposition site. Also associated with the

Snprprusnn 1990 Ae. tscvpr:r rN rHn DoNrrNrceN RspusLIc 519

Apri l '1987

E. ESPAILLAT

GUALEY

F€b 1984

M. SOCIAL

Jun€ 1088

Jun€ 1988

1 5 0

A p r i l 1 9 4 7 F e b . 1 9 8 8 June 19BB

I pcevtse - $ Posltve Houses N COrufetruEg - gpciltv€ OonrarnsrsE EHETEAU

J":?Bb";:l?;:." % rtuxtt x1o1# F€m/Hous6)xro

Fig. 4. Larval and adult Aedes oegypti indices from the barrios of Ensanche Espaillat, Gualey and Mejora-miento Social, Santo Domingo, Dominican Republic (April 1987 and June 1988).

r 00

uncleaned drums is an accumulating 0.5-5 cmlayer of decomposing organic debris-a nu-trient-enriched culture for the development ofbacteria, fungi, algae, protozoa and other orga-

nisms which serve as larval food (Riviere 1985).It is possible that the rough, porous concretesurface provides a good substrate for microor-ganisms which serve as food for the mosquito

520 JoURNAL oF THE AunRrcnu MosQulro CorrRol Assocllrlott Vol. 6,

larvae. This would help explain the larger pro-portion of pupae (with subsequent adults) re-corded from concrete containers. These concretetanks not only had high productivity but alsothe highest positivity rates-up to 77Vo con-tained immatures of Ae. aegypti.

Preliminary estimates of adult productivitywere based on pupal counts made during the Ae.aegypti immature census of 45 containers from19 premises in E. Espaillat. A pupal densityindex of 127 pupae was obtained using thecounts from all positive containers. Of 280 ran-domly collected pupae from 5 containerc,97%emerged. Assuming that roughly Vz of the 127pupae emerge as adults each day (observedpupaldevelopment time was 48 h), it is estimated thatapproximately 60 mosquitoes, or 30 females,were produced per premise per day in E. Espail-lat with 83% of these coming from the drums. Itis estimated that approximately 6,24O Ae. ae-gypfi adults were produced daily from a 3-blockarea of this barrio or approximately 4,140 mos-quitoes per hectare. With an average of 6.3 per-sons per house and 69 houses per ha it is esti-mated that approximately 10 mosquitoes areproduced daily for each person in the barrio.

Analyses of the results of surveys conductedin April 1987 and June 1988 in the 3 barriosdemonstrated no significant correlations be-tween the female density index and any of the 3Iarval indices (Fig. a). While the larval indicesdeclined in June, there was an increase in thefemale density index. There was also an increasein the positive house index, except in Gualeywhich remained, at 92%. Although there was areduction in the overall number ofpositive con-tainers. there was an increase in the total num-ber of positive barrels per house in E. Espaillat.The number of positive barrels per house re-mained essentially unchanged in the other 2areas. The decrease in nonessential water stor-age containers was associated with a 2-monthperiod of reduced, scattered rainfall when theaverage monthly precipitation was approxi-mately 60 mm. While the largest increase in theJune 1988 female density index occuned in E.Espaillat and might be attributed, at least inpart, to the increase in positive drums, the in-crease in the female densities in the other 2areas is not as readily explained.

Moore et al. (1978) reported high correlationsbetween seasonal rainfall and larval density in-dices reflecting the importance of rain-filledcontainers, relative to the number of essentialor deliberately filled water storage containers,in more affluent Puerto Rico. During the currentstudy in the Dominican Republic, there wereperiods without rainfall lasting from a few weeksto several months and no significant relation-ship was shown between rain and adult Ae.

aegypti densities. This is probably due to themosquito production from the essential waterstorage containers masking the rain associatedproduction (Tonn et al. 1969, Yasuno and Tonn1970).

These data demonstrate the difficulty in es-timating adult populations based on the prem-ise, container or Breteau larval indices. It issuggested that if adult populations cannot besampled directly then container productivityand the pupal density index would be betterindicators of population fluctuations especiallywhen obtained at regular intervals during theseason. While pupal production may be variabledepending on availability of Iarval food andother factors, repeated weekly samples will helpclarify population estimates.

For adult monitoring, Fox and Specht (1988)suggested that 5-min landing counts by obser-vation may be practical for evaluating the pres-ence of Ae. aegypti in areas of high mosquitodensity. In their study, however, adult monitor-ing was confined to only one site with 2 stationsapproximately 3 m apart and may not haveaccurately reflected population variations fromdifferent habitats in other parts of their studyarea (Morris 1960).

In the current study, the indoor use ofthe 2-man sweep net technique at 25 randomly dis-persed collecting sites in an area, provided anefficient monitoring system for both Ae. aegyptiand Cr. quinquefasciartus. Using this system, var-iation between areas, as well as variation be-tween houses or months, could be detected. Inaddition, monitoring adult populations beforeand after ULV adulticide applications revealedthat a reduction in the natural Ae. oegyptipop-ulations could be readily detected using thesweep net technique. While there was no signif-icant difference in the results obtained usingeither the small hand aspirator or the sweep net(based on a f-test, P < 0.01, comparing side byside samples using aspirator and sweep net), thenet was preferred for its economy since it didnot require the constant replacement of batter-ies.

It was noted that some houses had muchhigher populations of Ae. aegypti females (>20adult females per collection) than other similartypes of houses. During the course of 91 surveysin E. Espaillat and M. Social, approximately36% of the sample houses had at least onecollection with more than2D Ae. aegypti femalesand. 18% of the houses had 2 or more of thesehigh collections. One house had 134 femalesnetted in 10 min. These "mosquitogenic" housesgenerally do not share any obvious characteris-tics such as proximity to each other, specialconstruction, number of inhabitants, or highlyproductive breeding sites on or adjacent to the

SEPTEMBER 1990 An. lnevprr IN rnp DorrarNIcAN REPUBLIc 52r

premises that might account for the increasedpopulations. The ecology of such houses andtheir contribution to disease transmission war-rant further study.

In Gualey, as mentioned above, the houseswere of a different construction, less substantialand smaller. In addition they had 14% fewerbarrels than the other 2 areas. It is possible thatthese factors contributed to the relatively lowAe. aegypti densities as well as the lack of highdensity mosquito houses noted during the 148collections in this barrio.

In a comprehensive 3-year study on the adja-cent island of Puerto Rico, Moore et al. (1978)examined 21,306 premises for lawal breeding,reporting that the average premise, containerand Breteau indices were approximately 17, 9and 26, respectively. In the 3-year period theseauthors noted that an increase in dengue trans-mission occurred when the average Breteau in-dices rose above 20. This corresponds to a den-sity figure of 4 on the scale of 1-9 developed byBrown (1971).

During the current study in both E. Espaillatand M. Social, the premise, container and Bre-teau indices were at times approximately 80, 50and 200 or higher, which would correspond to adensity figure of 9. This is the highest categoryon the scale; as such, it could lead to high adultdensities with the potential for substantialtransmission of dengue. With open houses andcontiguous small shaded back yard areas, mos-quito dispersal among dwellings can readily oc-cur. This could facilitate dissemination of den-gue virus throughout the area.

Hieh Ae. aegypti densities are found through-out the Dominican Republic. Aedes aegypti wasapproximately twice as numerous as other spe-cies (mainly Cr. quinquefosciatus) collected fromthe houses in M. Social and E. Espaillat, whileCx. quinquefasciatus was more commonly col-lected in Gualey. The lack of adequate wastedisposal and the open sewer ditches found inthis area help provide the high Culex popula-tions.

Portions of E. Espaillat and M. Social wereincluded in 1978 and 1979 larval surveys madeby CDC and SNEM staff. During these surveysthe house indices in these areas ranged fromapproximately 30 to 40% (C. G. Moore, Centersfor Disease Control, Fort Collins, CO, personalcommunication). Since that time a deteriorationof the piped water system (T. Santana, SNEM,personal communication) and the increasing useof open containers to store water have contrib-uted substantially to an increase in mosquitopopulations in these as well as other areas ofSanto Domingo.

Aedes aegypti control programs have operatedin Santo Domingo (Cabrera 1979), but due to

political and economic constraints they have notbeen continued. In view of the high populationdensities of Ae. aegypti, the endemicity of all 4dengue serotypes, and the continuing use ofessential water storage containers, it is probablethat Santo Domingo will experience a seriousepidemic of dengue with increased incidence ofdengue hemorrhagic fever. Considering this im-pending threat and the limited resources avail-able for vector control, it is suggested that com-munity education and participation programsdirected toward reducing mosquito populationsbe strongly encouraged.

ACKNOWLEDGMENTS

The authors would like to thank the followingpeople for suggestions, assistance in carrying outthese studies and/or reviewing the manuscript:Yilma Mekuria, David Mandeville, MargaretTidwell, Richard Darsie, Gary Ross, RobertTonn, Allen Wooldridge, Dan Haile, OscarFultz, Chester Moore, Gary Clark, Paul Reiter,Ray Parsons, Fernando Agudelo, SNEM person-nel (Junior Felix, Rogaciones German, TeofiloSantana, Gaurin Tolentino, Juan Rodriguez,Beato Fermin) and the Dominican Republic Na-tional Meterological Service (Rosa Sanchez andCocco Quezada).

REFERENCES CITED

Brown, A. W. A. 1971. World wide surveillance ofAedes aegypti. Pap. Proc. Calif. Mosq. Control As-soc.42:20-25.

Cabrera, F. 1979. Observations on dengue endemicityin the Dominican Republic, pp. 44-45. In: Denguein the Caribbean,1977. Scientific Pub. 375. PanAmerican Health Organization, Washington, DC.

Chan, K. L. 1985. Methods and indices used in thesurveillance of dengu.e vectors. Mosq.-Borne DiseaseBul. 1:79-88.

De la Fuente G., S. 1975. Geografia Dominicana S. deIa Fuente G., Santo Domingo, Dominican Republic.

Fox, I. and P. Specht. 1988. Evaluating ultra-lowvolume ground applications of malathion againstAed,es aegypti using landing counts in Puerto Rico,1980-1984. J. Am. Mosq. Control Assoc. 4:163-167.

Moore, C. G., B. L. Cline, E. Ruiz-Tiben, D. Lee, H.Romney-Joseph and E. Rivera-Correra. 1978. Aed.esaegypti in Puerto Rico: environmental determinantsof larval abundance and relation to dengue virustransmission. Am. J. Trop. Med Hyg.27:1225-123L.

Morris, R. F. 1960. Sampling insect populations.Annu. Rev. Entomol. 5:243-264.

Peffa, C. J. and A. Zaglul. 1986. Los mosquitos de laCiudad de Santo Domingo. Ciencia y Sociedad1 1:176-187.

Riviere, F. 1985. Effects of two predators on commu-nity composition and biological control of Aedesaegypti al:d Aed.es polynesiensis, pp. I2l-135. In:L. P. Lounibos, J. R. Rey, and J. H. Frank (eds.),Ecology of mosquitoes: proceedings of a workshop.

522 JounNar, oF THE AupnrclN Moseurro Cournol AssocrluoN V o l . 6 , N o . 3

Florida Medical Entomology Laboratory, VeroBeach, FL.

SAS Institute Inc. 1985. SAS users guide: statistics,version 5 edition. Cary, NC.

Tinker, M. E. 1967. Measurement of Aedes acglptipopulations. J. Econ. Entomol. 60:634-682.

Tonn, R. J., P. M. Sheppard, W. W. McDonald andY. H. Bang. 1969. Replicate surueys of larval habi-tats of Aedes aegypti in relation to denzue hemor-rhagic fever in Bangkok, Thailand. Bull. W. H. O.40:819-829.

U.S. Department of Health, Education and Welfare.Public Health Service, Centers for Disease Control.1979. Biology and control of Aedes apgypti.VectotTopics No. 4, 39 pp.

U.S. Department of Health, Education and Welfare.Public Health Service, Centers for Disease Control.1981. Dengue surveillance, reference and research1981 report. 12 pp.

Yasuno, M. and R. J. Tonn. 1970. A study of bitinghabits of Aedes aegypti in Bangkok, Thailand. Bull.W. H. O. 43:319-325.