Journal of the American Mosquito Control Association,9(4):456-459, 1993

PRELIMINARY OBSERVATIONSON THE CHANGING ROLES OF MALARIA VECTORS IN SOUTHERN BELIZEI

D. R. ROBERTS,' O. CHAN,3 J. PECOR,4E. REIMANKOVA,5 S. MANGUIN,' J. POLANCO3 er.roL. J. LEGTERS'

ABSTRACT. A survey for larval and adnlt Anop&e/esmosquitoes was conducted in Toledo District of southern during August-September1992. We surveyedfor larvae in 145 habitats and conducted paired indoor--outdoor collections of adult mosquitoes landing on humans at 6 houses. In 1940-41, Kumm and Ram reported Anopheles darlingi females to be the most common Anopheles mosquitoes inside houses and reported no specimens ofl nophelesvestitipenrus in southern Belize. In our I 992 survey we found no An. darlingi mosquitoes either as adults or larvae. More An. vestitipennls females were captured indoors than outdoors, whereas most Anopheles albimanus and Anopheles apicimacula females were captured outdoon. All 3 species were represented occasionally in 145 collections of larvae from diverse habitats. Anopheles vestitipennis now appears to be a potentially important vector of malaria during the wet seasonin Toledo District.

The presenceof Anopheles darlingi Root in The Kumm and Ram survey was conducted Belize was first reported by Komp in 1940. The more than 50 yearsago, prior to the use of DDT identity of Komp's original An. darlingi speci- in the malaria control program in Belize. Since mens was recently verified by Linthicum (1988). their survey, no comparabledata have beenpub- In 1941. Kumm and Ram documented the oc- lished for the Toledo District. Thirty yearslater, currenceof house-frequentingpopulations of An. Bertram (1971)did not encountera singlespec- darlingi in the Toledo and Stann Creek districts imen of An. darlingi in an extensive survey of ofBelize (Fig. l). Kumm and Ram relied heavily adult mosquitoes in northern Belize, including on searchesof housesfor resting mosquitoes as areasof . Although few col- their primary survey melhod. Anopheles darlingi lections have been conducted in Belize. the last was found in 3 of 7 localities surveyedin Toledo documentedappearance of An. darlingi in and in 5 of 7 localities in Stann Creek was at Serra de Aqua in June 1946 (Linthicum District. The larvae of An. darlingl were also 1988). collected in both districts. Anophelesvestitipen- We initiated a malaria vector research pro- nis Dyar and Knab was not collected as adults gram in Belize in I 990 and conductedextensive or larvae in Toledo District, but was collectedas larval surveys in northern Belize, including Co- adults at 3 of7 localities and as larvae in Stann rozal, OrangeWalk, BelizeCity, Cayo, and Stann Creek District. Anopheles albimanus Wiede- Creek districts. No An. darlingi or An. vestiti- mann was the most widely distributed species, pennis larvae were collected in those surveys being presentat 9 of 14 localities from both dis- (Rejmankova et al. 1993).In a recent wet season tricts. Iarvae of An. albimanus were also col- survey in Toledo District we included nighttime, lected in both Toledo and Stann Creek districts. paired indoor{utdoor landing collections from Malaria sporozoiteswere found in the salivary humans to increase the likelihood of detecting glands of An. darlingi and An. vestitipennis,btti the presenceof An. darlingiandAn. vestitipennis. not An. albimanus. Anophelesdarlingi was the These collections were performed by capturing dominant indoor anopheline,representing more mosquitoes as they landed on the exposed legs than 700/oof the anophelinescaught indoors in and feet of 2-4 collectors.Paired indoor

0 30km

@ Elevations>1000 fe€t O SampleSites

t District I a

ttttt": Distributionof SampleSites cw

Dangriga

PuntaGorda

Fig. l. Map of Belize depicting the distribution of collection sites in Toledo district, near Punta Gorda, Belize.

1900 h) in activity for An. darlingi populations collectingsites. Although the collection data were in Honduras. Although we employed uniform not strongly quantitative, observations on rela- collecting methods, we could not control for the tive composition of indoor versus outdoor col- numbers of children and adults who gathered lectionsseemed valid. Most houseshad dirt floors around the collectorsat both indoor and outdoor and were constructedwith loose-fitting wood or 458 JouRr.rer-or rnn ArrrEnrcANMosqurro Covrnor Assocnrtox Vol.9, No.4

Table l. Numbers of anophelinescollected from humans in paired indoor+utdoor landing collections in the Punta Gorda area of southern Belize during August and Septembet 1992. Anopheles Anopheles Anopheles albimanus vestitipennis apicimacula Collection site Inside/outside Inside/outside Inside/outside Jacinto Landing 2/3 27/4 0/o Santa Helena 2/l 0/0 0/2 Crique Mafredi' 5/2 39/7 7/29 Crique Trosar o/2 9/O 9/33 Punta Gorda 0/2 2/r o/o Big Fall t/25 o/r 0/0 Totals to/35 77/13 r6/64 I Housesnot sprayed with DDT. palm slatsand a thatch roof. Two housesin Pun- vey. This speciesseemed undeterred by DDT ta Gorda and Big Fall were solid and tightly en- residues becauselarge numberc of An. vestitt closed. pennis femaleswere captured inside both DDT- Collections were performed at 6 sites in the sprayedand unsprayedhouses. Another intrigu- area of Punta Gorda (Toledo District) in south- ing aspectofthis species'behaviorrelates to our ern Belize (Fig. l). Although collections were capturing many more inside houses than were conductedin the samegeneral areas surveyed by capturedoutdoors. The opennessofmany native Kumm and Ram (1941), due to demographic houses in rural southern Belize probably facili- changesduring the intervening 5l years, their tates this indoor-feeding behavior. In contrast, specificsites were no longer in existence.In ad- the host-seekingfemales of An. apicimacula,like dition to the nighttime landing collections, we An. albimanus, were much more abundant out- conducted larval collections at 145 sites in To- doors. Exophagicbehavior ofthe latter 2 species ledo District. Specimensfrom all categoriesof should serve to diminish their overall vectorial collections have been depositedin the mosquito capacity. collectionofthe Walter ReedBiosystematics Unit Recent studiesby Loyola et al. (199 l) and Pa- at the Smithsonian Institution. dilla et al. (1992) have incriminated An. vestiti- No larvae or adults of An. darlingi were en- pennis as a vector of human malaria in the Mar- counteredduring our survey. The dominant spe- qu6s de Comillas region of southern Mexico and ciescollected biting indoors wasln. vestitipennis in 2 communities in northern Guatemala, re- (Table l). Both an. albimanus and An. vestiti- spectively. The latter studies, in combination with pennis were widely distributed, being presentat recent data compiled by Padilla in Guatemala 6 and 5 sites, respectively. Overall, larger pro- (personalcommunication), show An. vestitipen- portions of An. albimanus (78o/o)and, Anopheles nis to be endophagic, but not as strongly endo- apicimacula Dyar and Knab (80o/o)were collect- phagicas indicated by our data from Toledo Dis- ed outdoors than indoors. In contrast, 860/oofall trict. Consequently, a greater sampling effort An. vestitipennlsfemales were collected indoors. covering the entire nighttime interval will pos- All 3 Anophelesspecies collected in landing cap- sibly show a greaterrelative tendencyof An. ves- tures were also representedin the larval collec- titipennis females to feed outdoors in Toledo tions. District. Collection data presentedherein indicate that The presenceand abundance ofmalaria vec- An. darlingi is either restricted to specific local- tors are under the control of dynamic environ- ities that we did not sample, very uncommon, mental variables, as well as human interven- or possibly absent in Toledo District. In April tions. This report emphasizes the need to and May 1993, we finally encounteredpopula- continuously study the changingroles of malaria tions of ln. darlingi adults in riverine environ- vectors in different geographicalareas. Based on ments of Stann CreekDistrict. As a consequence, the published literature, we can expectAn. dar- we think,4n. darlingi is probably presentin To- lingi-transmitted malaria to respond favorably ledo District, but is limited to specific riverine to a DDT house-sprayprogram (Rozendaal et localities. al. 1989,Roberts and Alecrim l99l). However' Although An. vestitipenniswas not encoun- these expectations must be reevaluated lf An. tered in Toledo District during 1940, it was nu- vestitipennishas become the primary vector of merically dominant inside housesduring our sur- malaria in nonriverine areasof Toledo District. DncsMBEn1993 OpsRATroNer.exo ScrsNTrFrcNorEs 459

Hopefully, this report will be the precursor of section of the subgenusNyssariyn chus of Anopheles more definitive studies on vector responses to (Diptera: Culicidae). Mosq. Syst. 20:98-271. DDT-sprayed houses and on vectorial roles in I-oyola, E. G., J. I. Arredondo, M. H. Rodriguez, D. different ecological zones of Belize. N. Brown and M. A. Vaca-Marin. 1991. Anopheles vestitipennis, probable We thank the staffof the Belize/United States the vector of Plasmodium vi- yaxin the l-acandon forestofChiapas, M6xico. Trans. Epidemiological Research Center for support and R. Soc. Trop. Med. Hyg. 85:17l-174. assistance. In particular we want to thank Robert Padilla, N., P. Molina, J. Juarez. D. Brown and C. Miller and Shilpa Hakre for their direct assis- Cordon-Rosales. 1992. Potential malaria vectors tance. We thank Ralph Harbach for reviewing in northern Guatemala(Vectores potenciales de ma- the manuscript. Special thanks are due larry laria in la regi6n norte de Guatemala). J. Am. Mosq. Barber and his staffat the Voice of America in- Control Assoc. 8:307-308. stallation in Punta Gorda, Belize, for providing Rejmankova, 8., D. R. Roberts, R. E. Harbach, J. pe- laboratory space and continuous assistance dur- cor, E. L. Peyton, S. Manguin, R. Krieg, J. Polanco ing our field work in the Punta Gorda area. and L. kgters. 1993. Environmental and regional determinants of Anopheles larval distribution in northern Belize. J. Environ. Entomol. (in press). Roberts, D. REFERENCES CITED R. and W. D. Alecrim. 1991. Behavioral respnse of Anophelesdarlingi toDDT-sprayed house Bertram, D. S. 1971. Mosquitoes of British Hon- walls in Amazonia. PAHO Bull. 25:2lO-217. duras, with some comments on malaria. and on ar- Roberts, D. R., W. D. Alecrim, A. M. Tavaresand M. bovirus antibodies in man and equines. Trans. R. G. Radke. 1987. Thehouse-frequening, host-seek- Soc. Trop. Med. Hyg. 65:742-762. ing and resting behavior of Anopheles darlingi in Komp. W. H. W. 1940. The occurrenceof Anopheles southeasternAmazonas, Brazil. J. Am. Mosq. Con- darlingi Root in and Guatemala. trol Assoc.3:433441. Science9l:522-523. Rozendaal,J. A., J. P. M. Van Hoof, J. Voorham and Kumm, H. W. and L. M. Ram. 1941. Observations B. F. J. Oostburg. 1989. Behavioral responsesof onthe Anophelesof British Honduras. Am. J. Trop. Anopheles darlin3z in Suriname to DDT residues on Med. Hyg. 2l:559-566. house walls. J. Am. Mosq. Control Assoc. 5:339- Linthicum, K. J. 1988. A revision of the Argyritarsis 350.