Journal of the American Mosquito Contol Association, ll(4):463467, 1995

Copynghi O 1995 by the American Mosquito Control Association, Inc'

A NESTABLE FIBER POT FOR SAMPLING RESTING MOSQUITOES

MCHOLAS KOMAR' RICHARD J. POLLACK eNo ANDREW SPIELMAN

Department of Tropical Public Healt|t, Haward School of Public Health,

665 Huntington Avenue, Boston, MA 02115

ABSTRACT To enhance the effectiveness of an arbovirus monitoring program, we evaluated a com-

mercially available device for sampling resting vector mosquitoes. Diverse Anopheles, Culiseta, and,Culex

mosquitoes were taken in these nlstable fiber pots. The pots sample about as many Culiseta melanura

mos{uitoes per device as do conventional resting boxes, but fewer than do boxes fitted with expanded

framis. Mori Cs, melnnura, and more bloodfed mosquitoes, but fewer species of mosquitoes are harvested

with fiber pots than with CDC light traps. Fiber pots are more readily used, transported, and stored and

are less expensive than conventional resting box devices or CDC light traps. A monitoring program based

on the use of fiber pots, therefore, expends fewer resources than one using conventional resting boxes

and collects about as many vector mosquitoes,

INTRODUCTION

Various kinds of devices have been used tocapture resting vector mosquitoes (Service

1976). Prt traps (Muirhead-Thomson 1958), bytheir nature, are placed permanently in a site andgenerally require elaborate construction. Wood-en kegs, once used for shipping nails, wereadapted for sampling anopheline mosquitoesduring World War tr (Smith 1942), and othermodifications on this theme have been usedsince. These resting box devices are open on oneside and generally are dark colored (Goodwin1942. Burbutis and Jobbins 1958. Gusciora1961, Moussa et al. 1966). Efficacy is enhancedwhen the apparent area of the opening is en-larged by mqrns of a funnel-like extension (Ed-man et al. 1968). These devices continue to beused in numerous malaria and arbovirus sur-veillance programs.

Although the efficacy and simplicity of thestandard rectangular resting box devices renderthem attractive for routine monitoring, their costand bulk impede their operational use. Becausenails no longer are packed in kegs, wooden rest-ing devices must be specially fabricated. Theirparallel-sided design, howevet prevents themfrom nesting and renders contained mosquitoesdifficult to harvest. A less expensive resting boxdevice that is more convenient to transport andeasier to use may facilitate efforts to monitormosquitoes.

To improve the efficiency of an ongoing ar-bovirus monitoring program, we evaluated theeffectiveness of commercially available, nesta-ble fiber pots for sampling natural populationsof Culiseta melanura (Coq.), a diurnally restingmosquito that is the enzootic vector of easternencephalitis (EE) and Highlands J viruses (Mor-ris 1988). Efficacy was compared among fiberpots, conventional plywood resting boxes, andCDC light traps.

MATERHLS AND METIIODS

Sampling devices: Commercially available fi-ber pots (Western Pulp Products Co., Corvallis,OR), are molded from recycled wood pulp in theform of hollow, truncated pyramids (Fig. 1) witha height of 28 cm. The base (open end) is 28 x28 cm, and the top (closed end) is 15 x 15 cm.The thickness of the fiber walls is I cm. Allsurfaces of the pot are dark brown.

For comparison, we used conventional ply-wood boxes (20 X 38 X 3O cm) fashioned fromplywood and with their interiors painted matteblack. Funnel-shaped entrance frames wereplaced on these boxes during part of the study(Fig. 2), effectively expanding the opening to 6OX 120 cm. Resting boxes and frames were pro-vided by John D. Edman, University of Massa-chusetts. Portable CDC miniature light traps(John W. Hock, Inc., Gainesville, FL) also wereused for comparison (Sudia and Chamberlain1962).

Study sites: Mosquitoes were sampled at 12sites in southeastern Massachusetts (Bristol,Plymouth, and Norfolk counties). Seven of thesesites were located in the Hockomock SwampWildlife Management Area, a location knownfor arbovirus activity, in the towns of Thuntonand Easton (Bristol County) and Bridgewater(Plymouth County). Three additional sites werelocated in Kingston (Plymouth County) and oneeach in Hingham (Plymouth County) andStoughton (Norfolk County). All sites appearedsuitable for capturing resting mosquitoes andwere near avian roosting sites.

Procedures: Fiber pots and resting boxeswere placed on the ground with their openingsoriented at random. Light traps were suspendedl-2 m above the ground at intervals ofabout l0m and operated overnight using 6-V motorcyclebatteries. Mosquitoes in light traps generallywere harvested during morning hours; resting

Frsen Pors ron Slt.,rpLwc ResrINc MoSQUIToES 465Decerrassn 1995

Table 1. Number of mosquitoes harvested in2,500 collections from nestable fiber pots set in

12 southeastern Massachusetts sites.

Mosquitoesper l,0OOcollections

Females

Species

7ablood- No.

No. fed males

Culiseta melanuraCuliseta morsitansAnopheles quadri-

maculatusCulex territansAedes cinereusCulex restuansCulex pipiensAno p he le s p unctip enni sUranotaenia sapphirinaCulex salinariusUnidentified

Total 51 294

to October 3, 1994 (Table 1). Culiseta melanurapredominated, comprising 75Vo of the harvest.Some Anopheles quadrimaculatus Say also wereharvested from these devices. More than half ofthe females harvested contained readily detect-able vertebrate blood. On average, about onemosquito was taken from each pot per day.Highlands J virus was detected in 2 of 95 poolsof pot-harvested Cs. melanura. Nestable fiberpots, therefore, sampled numerous vector mos-quitoes.

The harvest of mosquitoes taken in fiber potswas compared to that taken with conventionalplywood resting boxes and with resting boxesenhanced by an expanded frame. In Test I, 16fiber pots and 15 resting boxes were randomlyplaced within a 0.5-ha wooded space from Sep-tember 1 to September 17, 1994. A similar arrayof devices was monitored from July 18 to Au-glst 23, 1994, in Test II, except that expandedframes were placed on all resting boxes. Similarharvests of mosquitoes were recorded in the fi-ber pots and the frameless resting boxes (Table2). Boxes with expanded frames, however, har-vested far more mosquitoes than did the fiberpots. We conclude that fiber pots sample asmany mosquitoes as do resting boxes, but fewerthan do boxes fitted with expanded frames.

We then comp€ued the diversity of mosquitoestaken with flber pots to that taken with CDClight traps in Kingston, MA, from August 30 to

Table 2. Comparison of number of female

mosquitoes harvested from nestable fiber pots

with that from conventional resting boxes orresting boxes augmented by an expanded

frame.

Mosquitoesper l0O

collections

Culisetamelanura

No.col-lec-tions

48667

60 2063 1 0

3 2 43 2 r

1 9 069 <l62 l0 < l0 00 0o 4 l

2624

65J

2I

< 14

625

Collectiondevice

Totalno.

Voblood-

No. fed

24 132 t 2 l

13 2982 18

9590

u l96

II

Test

Fiber potResting box

Fiber potBox and frame

2830

l 6107

September 19, 1994. Twenty fiber pots wereplaced in a l-ha site near 6 light traps. This sitesupported the roosting of more than 1,000American robins (Turdus mi7ratorius Linn), butcontained no habitat suitable for larval Cs. me-lanura. More than twice as many female Cs. me-Ianura were taken per fiber pot than per lighttrap (Table 3). Our light trap collections includedmany more human-biting mosquitoes than didthe fiber pot collections; no Aedes mosquitoes,Coquillettidia perturbans (Walker), nor Culexsalinarius Coq. were taken in fiber pots. Al-though anopheline mosquitoes were taken withboth kinds ofdevices, light traps collected 3-foldmore than did fiber pots. Overall, more than 6times as many bloodfed mosquitoes were takenin fiber pots than in light traps. More Cs. me-lanura, and more bloodfed mosquitoes, but alesser diversity of mosquitoes, are harvestedwith fiber pots than with light traps.

DISCUSSION

Monitoring systems for vector mosquitoesmust be adapted to the peculiar features of thetarget mosquito and to the prevailing physicaland political environment. The nocturnal flightbehavior of Cs. melanura renders them vulner-able to capture in light traps, and their attractionto dark cavities results in their entry into artifi-cial diurnal resting devices. For various reasonsthe Encephalitis Surveillance Program of theCommonwealth of Massachusetts has beenbased on light trap collections (Edman et al.1993) and that of New Jersey on resting boxes(Crans 1994). A 3rd option, adopted by the stateof Florida, employs serological monitoring of

466 Jounxal or rne AunnrceN Mosqurro CoNrnol AssocterroN Vor . l l , No.4

Table 3. Comparison of number and feeding status of female mosquitoes harvested fromnestable fiber pots with that from CDC light traps.

Mosquitoes per 1,000 collections in

Fiber pots(260 collections)

Light traps(33 collections)

Species MeanVo

bloodfed MeanVo

bloodfed

Culiseta melanuraCuliseta morsitansAnop he I e s q uadrimac ul at u sCulex territansCulex restuansCulex pipiensU ranotae nia s ap p hirinaCulex salinariusAedes cantatorAedes vexansUnidentified

Total

9683o0

80loo

o

1799542I

< l000o

8 l6

1 80

l5J

J

369J

5

r79

2650o

20oo000o

152M 9 l

sentinel chicken flocks (Morris 1988). Eachsampling strategy produces data requiringunique modes of interpretation.

The fiber pots that we evaluated for monitor-ing Cs. melanura in Massachusetts may benefitencephalitis surveillance programs. As many Cs.melanura were harvested by means of theseconvenient and inexpensive devices as with theplywood resting boxes that have conventionallybeen used for this purpose. Although CDC lighttraps are said to collect about l0 times as manyfemale Cs. melanura as do resting boxes (Nascil98O'), we found that nestable fiber pots col-lected more mosquitoes than did unbaited CDClight traps. Larger collections of Cs. melanurafrom resting boxes than from CDC light traps atan upland farm site, but not in forested wetlandsites, have been reported (Joseph and Bickley1969). This finding may be explained by theabundance of vertebrate hosts at upland sites.Perhaps the attraction of a dense nidus of birdsoutcompetes the attraction of a light trap. Mos-quitoes seem to enter our pots most abundantlyearly in the afternoon, perhaps because they ini-tially rested postprandially in the canopy neartheir roosting avian hosts. Harsh daytime envi-ronmental conditions may drive these mosqui-toes toward the ground near our fiber pots. Therelative frequency of blood-engorged Cs. melan-rra distinguishes resting collections from light

'Nasci, R. S. 1980. Vector biology of Culiseta me-lanura (Coquillett) in southeastern Massachusetts.Ph.D. dissertation. Universitv of Massachusetts, Am-herst, MA. lfi) pp.

trap collections. A diversified monitoring pro-gram may be particularly useful because restingdevices sample a different segment of the vectorpopulation than do light traps.

The nearly total absence of Aedes mosquitoesin our fiber pot collections confirms observa-tions with other resting devices (Burbutis andJobbins 1958). The harvesting potential of fiberpots for Cq. perturbans, an important pest andpotential EE epidemic vector (Howitt et al.1949), has yet to be determined because thisspecies' adult population was exceedingly lowduring the period of study at our sites. Althoughfiber pots sample many more nocturnally activeCuliseta and Culex mosquitoes, light traps inMassachusetts have the advantage of samplingdiverse Aedes, as well as Cq. perturbans mos-quitoes. Information on the density and virus in-fection rate in these "bridge-vector" mosquitoesis useful in formulating an intervention decision,and this information will be lacking in a moni-toring program that does not include light traps.

Nestable fiber pots sample as many Cs. me-lanura, including virus-infected females, as doconventional resting boxes. Because diversekinds of mosquitoes, including anophelines, restin these pots, they may be useful for monitoringthe intensity of transmission of other mosquito-borne pathogens, such as human malaria.

We confirm that resting boxes fitted with fun-nel-like collapsible frames, which increase theapparent area of the entrance 9.6-fold, collectabout 6 times more mosquitoes than otherwisewould be collected (Edman et al, 1968). Oneperson, however, can service many more pots

FBsn Pors pon SnuplrNc Rrsrnqc MosQultor's 46'7Dncsl,aen 1995

than frame-mounted boxes in a given period of

time. and at far less cost. The disadvantages of

mounting frames, therefore, appear to outweighbenefits gained in collection efficacy.

Economic factors may influence the choice offiber pots over conventional resting boxes. Weestimate that a resting box can be procured forabout $15 and that a frame costs an additional$15. A fiber pot, in contrast, costs only $1.30(delivered). A 0.5-kg (1-lb) fiber pot weighs sig-nificantly less than a resting box (about 3 kg)and a plywood box with its frame (about 7 kg).The nestable feature of the fiber pots combinedwith their light weight promotes portability andstorage. Indeed, we believe that one person cancomfortably carry either 20 flber pots, 2 restingboxes, or one resting box with its accompanyingcollapsed frame. None of the 295 flber pots thatwe have installed appears to have lost structuralintegrity during 1 year of exposure to an accu-mulated IOO cm of precipitation. In Practice, wefound that mosquitoes can be harvested in halfthe time from a trapezoidal fiber pot than froma cuboidal resting box. Furthermore, we foundthat artiflcial illumination of the interior of theresting device was essential for mosquito collec-tion from plywood boxes, but not from fiberpots. Thus, monitoring programs that use fiberpots expend fewer resources than those usingconventional resting boxes.

Our experience with nestable fiber pots forsampling arbovirus vector mosquitoes encour-ages us to recommend their further evaluationfor use as sampling devices for mosquitoes thattransmit the agents of malaria, filariasis, RiftValley fever, and various other encephalitides.Use of these pots may facilitate routine moni-toring of vector mosquitoes.

ACKNOWLEDGMENTS

This research was supported in part by theMassachusetts Department of Public Health andthe Northeastern Mosquito Control Association.We thank Yvonne Chase of the Massachusetts

Department of Public Health for identification of

virus isolates.

REFERENCES CITED

Burbutis. P. P and D. M. Jobbins. f958. Studies on

the use of a diumal resting box for the collection of

Culiseta melanura (Coquillett). Bull. Brooklyn En-

tomol. Soc. 53:53-58.Crans, W. 1994. F,F,E in New Jersey during 1993-

Proc. N.J. Mosq. Control Assoc. 8l:l18-121.Edman, J. D., E D. S. Evans and J. A. Williams. 1968-

Development of a diurnal resting box to collect Cz-

liseta melanura (Coq.) Am. J. Trop. Med. Hyg. 17:

451456.Edman, J. D., R. Timperi and B. Werner. 1993. Epi-

demiology of eastem equine encephalitis in Massa-

chusetts. J. Fla. Mosq. Control Assoc. 64:84-96.

Goodwin, M. H. 1942. Studies on artificial restingplaces of Anopheles quadrimaculatus Say. J. Natl.Malaria Soc. 1:93-99.

Gusciora, W. R. 1961. The resting box technique forthe sampling of Culiseta melanura (Coquillett).

Proc. N.J. Mosq. Exterm. Assoc.48:.122-125.Howitt, B. F,, H. R. Dodge, L. K. Bishop and R. H.

Gorrie. 1949. Recovery of the virus of easternequine encephalomyelitis from mosquitoes (Man-

sonia perturbans) collected in Georgia. Science 1 10:l4t-142.

Joseph, S. R. and W. E. Bickley. 1969. Culiseta me-lanura (CoquTllett) on the eastem shore of Maryland(Diptera: Culicidae). Bull. 4-161. Univ. Md. Agric.Expt. Stn., College Park, MD.

Morris, C. D. 1988. Eastern equine encephalomyelitis,pp. l--2O. In: T. P Monath (ed.). The arboviruses:epidemiology and ecology, Volume III. CRC Press,Boca Raton, FL.

Moussa, M. A., D. J. Gould, M. P. Nolan, Jr. and D.E. Hayes. 1966. Observations on Culiseta melan'zra (Coquillen) in relation to encephalitis in south-em Maryland. Mosq. News 26:385-393.

Muirhead-Thomson, R. C. 1958. A pit shelter forsampling outdoor mosquito populations. Bull.W .H .O . 19 : l l 16 -1118 .

Service, M. W. 1976. Mosquito ecology. Applied Sci-ence Publ., Ltd., London.

Smith, G. E. 1942. The keg shelter as a diurnal restingplace of Anopheles quadrimaculatus. Am. J. Tlop.lN'Ied.22:257-269.

Sudia, W. D. and R. W. Chamberlain. 1962. Battery-operated light trap, an improved model. Mosq. News22:126-129.