Biting patterns and seasonal densities of Anopheles mosquitoes in the Cayo District, Belize, Central America with emphasis on Anopheles darlingi

Transcription

1 Journal of Vector Ecology 45 Biting patterns and seasonal densities of Anopheles mosquitoes in the Cayo District, Belize, Central America with emphasis on Anopheles darlingi Nicole L. Achee, John P. Grieco 1, Eliska Rejmankova 2, Richard G. Andre 1 Errol Vanzie 3, Jorge Polanco 3, Ireneo Briceno 3, Russell King 3, and Donald R. Roberts 1 1 Department of Preventive Medicine and Biometrics, Uniformed Services University of the Health Sciences, 431 Jones Bridge Road, Bethesda, MD 2814, U.S.A. 2 Department of Environmental Science and Policy, University of California, Davis, CA 95616, U.S.A. 3 Ministry of Health, Belize Received 1 April 25; Accepted 8 November 25 ABSTRACT: The present study utilized an experimental hut to conduct human-baited landing collections for characterizing the all-night biting patterns and seasonal densities of adult Anopheles darlingi in the centrally located Cayo District of Belize, Central America. A total of 25 all-night collections (i.e., sunset to sunrise) were conducted from January 22 to May 23, capturing a total of 18,878 An. darlingi females. Anopheles darlingi exhibited a bimodal nightly biting pattern with one predominate peak occurring three h after sunset and a smaller peak occurring one h prior to sunrise. Biting females were collected throughout the night in higher densities indoors (9,611) than outside (9,267) the experimental hut (O:I=1.:1.4). Seasonal adult collections show An. darlingi densities were highest during the transitional months between the end of the wet and beginning of the dry season (January) and the end of the dry season and beginning of the wet season (May). A total of 2,1 An. darlingi females was captured in 31 two-h, human-baited landing collections performed from January to October 22. Anopheles darlingi monthly population densities were found to have no significant associations with high or low temperatures, precipitation, or river level. However, qualitative data examination indicates an inverse relationship between river level and An. darlingi adult collections suggesting a disturbance of larval habitats. All-night biting and seasonal distribution patterns for other anopheline species are also described. None of the adult specimens collected throughout the entire study tested positive for Plasmodium spp. infection using the VecTest rapid diagnostic kit. Journal of Vector Ecology 31 (1): Keyword Index: Anopheles darlingi, biting behavior, seasonal distribution, malaria, Belize. INTRODUCTION Public health officials in endemic countries must understand the role specific anopheline vectors play in malaria transmission in order to implement successful, cost-effective control methods. This includes defining the nightly biting patterns, indoor/outdoor biting ratios, and seasonal population densities of anopheline species of interest. The public health importance of malaria in Belize has stimulated the investigation of potential vector status of some of the anopheline species found throughout the country including: Anopheles albimanus Weidemann, An. darlingi Root, An. pseudopunctipennis Theobald, and An. vestitipennis Dyar & Knab. All of these species are documented vectors of malaria transmission in the Americas (Loyola et al. 1991, Padilla et al. 1992, Ramsey et al. 1994, Lourenco-de-Oliveira 1989, Klein et al. 1991) and specifically Belize (Achee et al. 2); however, An. darlingi is presently considered one of the most important. This designation is based upon natural sporozoite infections (Davis 1931, Arruda et al. 1986, Herrera et al. 1987, Oliveira-Ferreira et al. 199), host-feeding preferences and indoor biting behavior (Deane et al. 1946, Roberts et al. 1987, Klein and Lima 199; Rozendaal et al. 1989, Grieco 21) as well as positive associations between adult densities and peaks of regional malaria transmission throughout its geographic distribution (Ferraroni and Hayes 1979, Deane 1986, Lourenco-de-Oliveira et al. 1989, Charlwood 198, Rozendaal 1992, Tadei et al. 1998, Tadei and Thatcher 2, Da Silva-Vasconcelos et al. 22). For these reasons, An. darlingi has been the focus of several behavioral studies in Belize in order to understand its role in local malaria transmission (Manguin et al. 1996a, Harbach et al. 1993, Roberts et al. 1996, Roberts et al. 22, Rejmankova et al. 2, Grieco 4 ). Kumm and Ram (1941) first described natural sporozoite infections in An. darlingi from Belize, and recent studies have also detected P. falciparum infections in wild-caught females collected from inside local homes (Achee et al. 2). Anopheles darlingi populations from Belize have demonstrated both high salivary gland (41%) and midgut (53%) susceptibility to a laboratory strain of P. falciparum (Grieco et al. 25a). In addition, results from remote sensing and geographical information 4 Grieco, J.P. 21. The bionomics and vector competence of Anopheles albimanus Wiedemann and Anopheles vestitipennis Dyar and Knab in the Toledo District of southern Belize. Doctoral Dissertation, Uniformed Services University of the Health Sciences. 445 pp.

2 46 Journal of Vector Ecology June 26 system (GIS) studies have shown the highest malaria incidence in Belize to occur within villages located at a distance 1 km of previously defined An. darlingi larval habitats (i.e., freshwater rivers) (Hakre 5 ). Previous studies using human-baited collections have shown that An. darlingi feeds both outdoors and indoors but exhibits a primarily endophagic behavior (Grieco 21, Roberts et al. 1996). Data from these studies, however, were representative of only early evening activity (183-2 h) with no quantification of late-night biting patterns. The temporal patterns of biting may be altered due to host defensive behaviors and environmental parameters that change over the course of the night. Until the present study, the host seeking and feeding behaviors of An. darlingi in Belize during late-night hours were unknown. Seasonal variation studies of An. darlingi in Belize are also limited. Roberts et al. (22) showed An. darlingi females were present in houses located in proximity to rivers during both wet and dry seasons but were only present at upland homes during the wet season. A seasonal distribution study of An. darlingi in the southern Toledo District of Belize indicated adult population densities to be highest during the months of the dry season with lowest rainfall (Grieco 21). Because specific vectors require specific types of aquatic habitats, temporal distributions will vary within different topographical environments as a result of habitat availability. For this reason, further research examining the relationship between seasonal changes and An. darlingi abundance in different regions of Belize was needed. The objectives of the present study were to define the 12-h indoor/outdoor biting activity patterns of An. darlingi 5 Hakre, S. 23. Epidemiology of Malaria in Belize. Doctoral Dissertation, Uniformed Services University of the Health Sciences. 282 pp. in Belize and to determine associations between seasonal adult densities and environmental data on a local scale for the Cayo District. This information will provide further insight into the role of An. darlingi in the transmission of malaria within the region. MATERIALS AND METHODS Study site The study site was established on open pastureland at N and W in the foothills of the Maya Mountain Range within the mid-reaches of the Sibun River Watershed. Located in the central Cayo District of Belize, the study site was approximately 12 miles south of the capital, Belmopan (Figure 1). There are a total of 11 villages within the watershed, the three closest (i.e., within 5 miles) of the study site being Armenia (pop. 4), Caves Branch (pop. 7), and Hershey (pop. 1). The majority of inhabitants are Spanish immigrants with some indigenous Maya and Creole. The Cayo District reported a total of 153 human cases of Plasmodium vivax malaria for the 22 study year, which was slightly less than the 184 reported in 21. There were no cases of P. falciparum reported for either year. Malaria case distributions for Armenia, Caves Branch, and Hershey reported five total cases of P. vivax for the year 22 and seven total cases of P. vivax in 21, according to the Belize Ministry of Health. In 22, two cases occurred in February with one case occurring each in March, June, and September. Experimental hut A house survey of 1 homes within the adjacent village of Armenia was conducted prior to the start of the study to ensure comparability of sleeping area, design, and construction materials between indigenous homes and the experimental hut. The design for the 4 x 4 m experimental hut consisted of untreated plank walls, a corrugated zinc roof, Figure 1. Map of the research site ( ) located within the central Cayo District of Belize (from Roberts et al. 1996).

3 Journal of Vector Ecology 47 and dirt floor. There was one window per wall for a total of three windows, each measuring.67 x.67 m, and one door measuring 2. x 1. m. The windows and door remained open throughout the entire biting collection period. Details of the construction design are in press (Achee et al. 25). All-night mosquito collections The 12-h biting pattern of An. darlingi was characterized using human-baited collections starting 3 min prior to sunset (18 h) and continuing until 3 min. after sunrise (6 h). Collections were conducted when adult densities were highest (January to July). One collector was located inside the hut and another collector was positioned outside, approximately 2 m from the door. All anopheline females landing on the exposed lower legs of the collectors were captured using manual aspirators during a 2-min sampling period each halfhour. Collectors rotated positions (i.e., inside/outside) after each sampling period and were replaced with another set of collectors after 3 h. Mosquitoes were placed into modified cardboard pint cartons and killed each hour by suffocation with acetone vapors within an airtight killing chamber (i.e., insulated cooler). Upon completion of the 12-h collection period, specimens were sorted by species (Wilkerson and Strickman 199), counted, and placed into 1.5 ml vials properly labeled with hour of collection, date, indoor/outdoor station, and species then placed over silica gel in a sealed container until testing for sporozoite infections. The total numbers of each anopheline species captured at both the indoor and outdoor locations for all collections were averaged by collection hour. Seasonal mosquito collections Human-baited landing collections were performed at the same research site as previously described, except that each collection lasted only 2 h from sunset (18-2 h), with collectors rotating indoor and outdoor positions after continuous 3-min sampling periods. Three collections were performed each month except for the months of January (six collections), May (one collection), and July (six collections). No collections were performed in the months of June, November, and December. The total number of each anopheline species captured per collection was recorded and monthly averages calculated. Natural sporozoite infections Sample pools, consisting of 1 to 1 adults, of each anopheline species sorted by hour, biting location (i.e., indoor or outside), and date from both all-night and 2-h seasonal collections were screened for natural malaria sporozoite infections (i.e., P. falciparum, P. vivax VK21, P. vivax VK247) using VecTest rapid diagnostic kits (Medical Analysis Systems, Inc., Camarillo, CA) according to manufacturer s instructions. Data analysis Correlations between the nightly density of An. darlingi captured from 12-h collections and environmental variables, including indoor and outdoor temperature, relativity humidity (gathered using HOBO Pro Series Weatherproof Data Loggers (Forestry Suppliers Inc., Jackson, MS)) and wind speed (gathered using Davis Weather Monitor II (Davis Instruments Inc.)) were analyzed using multiple linear regression and nonparametric bivariate statistics (SPSS version 9., SPSS Inc.). Student s t-tests were performed to compare average indoor and outdoor environmental variables by hour. Nonparametric bivariate correlations and multiple linear regression analyses were performed to define associations between monthly average anopheline densities captured from 2-h seasonal collections and environmental data including temperature, precipitation, and river level obtained from the National Meteorological Service. Table 1. Total numbers of anopheline mosquitoes captured in 25 all-night biting collections conducted indoors and outside an experimental hut from January 22-May 23 as part of a study examining the biting activity pattern of An. darlingi in Belize, Central America. Species Total Collected Indoor Outdoor O:I An. darlingi 18,878 9,611 9,267 1.:1.4 An. albimanus :1: An. pseudopunctipennis :1:88 An. punctimacula :.77 An. vestitipennis :1.2 An. apicimacula :.91 An. gabaldoni An. punctipennis An. crucians Chagasia bathana Aberrant An. darlingi a :.64 a Harbach et al

4 48 Journal of Vector Ecology June 26 Table 2. Results from 31 human-baited landing collections conducted indoors and outside an experimental hut for 2 h after sunset during January-October 22. Total number of collections performed each month include: January=6; February=3; March=3; April=3; May=1; July=6; August=3; September=3; and October=3. January February March April May July August September October Species I O I O I O I O I O I O I O I O I O O:I An. darlingi n=2, b c :.95 An. albimanus n= :.71 An. punctimacula n= :.3 An. apicimacula :.58 n=3 An. vestitipennis n=24 An. pseudopunctipennis n=17 An. gabaldoni n=11 An. crucians n=2 Aberrant An. darlingi a n=14 Monthly An. darlingi Avg. Monthly An. darlingi O:I : : : : : :1.5 1.: : :.61 1.:.94 1.: : :2.45 a Harbach et al b For O:I calculations 94 An. darlingi were removed from the total outside population because no indoor collections were simultaneously conducted during four sampling periods. c For O:I calculations 21 An. darlingi were removed from the total outside population because no indoor collections were simultaneously conducted during two sampling periods.

5 Journal of Vector Ecology 49 RESULTS All-night mosquito collections From 25 all-night collections performed from January 22 to May 23, a total of 18,878 An. darlingi female mosquitoes were captured (Table 1). Other anopheline species collected in decreasing frequency included: 527 An. albimanus Wiedemann, 371 An. pseudopunctipennis Theobald, 46 An. punctimacula Dyar and Knab, 44 An. vestitipennis Dyar and Knab, 21 An. apicimacula Dyar and Knab, 5 An. gabaldoni Vargas, 2 An. punctipennis Say, and 1 An. crucians Wiedemann. In addition, there were a total of 2 Chagasia bathana Dyar and 258 aberrant morpho-types of An. darlingi (Harbach et al. 1993). Anopheles darlingi females were captured during all collection hours and exhibited a bimodal indoor/outdoor biting pattern with one predominate peak occurring 3 h after sunset and a second weak peak defined 1 h prior to sunrise (Figure 2A). The cumulative majority of all females captured both indoors (54%) and outdoors (56%) was collected within 5 h post-sunset although biting continued throughout the night. Anopheles darlingi were endophagic with a higher population density being collected inside the experimental hut (9,611) as compared to outdoors (9,267) giving an outdoor to indoor ratio of 1.:1.4 (Table 1). Biting patterns of other anophelines of medical importance in the country, An. albimanus and An. pseudopunctipennis, were also examined. Anopheles albimanus showed an endophagic behavior with an outdoor to indoor biting ratio of 1.:1. (Table 1). Although the total number of An. albimanus captured was low (527), allnight collections indicated a bimodal biting activity pattern with an early peak occurring within 2 h post-sunset and another pronounced peak occurring 3 h prior to sunrise (Figure 2B). However, the majority of An. albimanus females collected, both indoors (55%) and outdoors (58%), were captured within 3 h after sunset with minimal numbers being collected during the remainder of the night. A strong endophagic behavior was described for An. pseudopunctipennis with 242 females being collected inside compared to 129 outside the experimental hut. This defined an outdoor to indoor biting ratio of 1.:1.88 (Table 1). While not well-defined due to the low numbers of collected specimens (371), females exhibited a peak in indoor biting starting 2 h post-sunset and continuing until 6 h post-sunset with low numbers continuing to bite throughout the night (Figure 2C). Outdoor collections indicated a peak at 5 h postsunset and again 1 h prior to sunrise. The majority of An. pseudopunctipennis both inside (54%) and outside (47%) the experimental hut were not collected until 5 h post-sunset. The total numbers of other female anopheline species collected at the research site were too low to use in extrapolating meaningful biting activity patterns. Weather data recorded during all-night collections indicated temperature readings inside the hut averaged 24 C with a range from 17 C to 31 C, while outside the hut temperatures averaged 23 C and ranged from 16 C to 29 C. The average temperature indoors was consistently higher each hour throughout the night compared to the average outside temperature (Student s t-test, P=<.5). Relative humidity levels recorded indoors ranged from 69 to 1% with the average of 9% being significantly lower than the average humidity level of 95% recorded outside the experimental hut (Student s t-test, P=.2). High wind speeds recorded at the collection site averaged 4 km/h and ranged from to 32 km/ h, with the strongest winds occurring during the first 4 h postsunset. Multiple linear regression analyses of average nightly environmental data with nightly An. darlingi biting densities indicate that the numbers of females collected biting both inside and outdoors were not significantly associated with any of the environmental variables measured. Seasonal biting collections A total of 31 2-h seasonal collections conducted from January to October 22 generated a total of 2,1 An. darlingi (Table 2). Other anopheline species captured in decreasing densities included: 162 An. albimanus, 65 An. punctimacula, 3 An. apicimacula, 24 An. vestitipennis, 17 An. pseudopunctipennis, 11 An. gabaldoni, and 2 An. crucians. In addition, 14 aberrant morpho-types of An. darlingi were also collected (Harbach et al. 1993). The months of January (11), May (31), and July (99) exhibited the highest monthly densities of biting An. darlingi females (Figure 3; Table 2). These represent transitional periods between the end of the wet season (January) and the end of the dry season (May to July). The three collections performed in April averaged the smallest number (15) of An. darlingi adults. Overall, An. darlingi populations exhibited an exophagic biting behavior with a total of 923 collected indoors and 972 collected outside the experimental hut giving an outdoor to indoor ratio of 1.:.95 (Table 2). However, upon monthly examination, a strong endophagic response was seen in six (February, March, April, August, September, and October) of the nine months in which collections were conducted, with outdoor to indoor ratios ranging from 1.:.61 in May to 1.:2.92 in August. Annual rainfall for the study year 22 totaled 3,237 mm, with 83% (2,697 mm) falling in the wet season (June to December) and 17% (54 mm) within the dry season (January to May). For the months in which the study was conducted, the highest precipitation level (15 mm) occurred during the month of July and the lowest (.7 mm) in April (Figure 3A). The average monthly river levels showed a peak of 2.12 m in August, and the lowest level (1.19 m) occurred in January (Figure 3B). The average high temperature during the collection months ranged from 29 C in January to 34 C in the month of May (National Meteorological Service) (Figure 3C). Average low temperatures ranged from 19 C in January to 23 C from May to September (Figure 3D). Nonparametric analyses of average environmental parameters with monthly average densities of An. darlingi indicated no significant associations. However, because there was a negative trend (r=-.22; p=.235) between the level of the Sibun River (i.e., the main source of An. darlingi detritus habitats at the present study site) and the total number of An. darlingi females collected, further data exploration was

6 5 Journal of Vector Ecology June 26 Average An. darlingi collected per hour Hour of Collection A Inside n=9,611 Outside n=9,267 Average An. albimanus collected per hour Average An. pseudopunctipennis collected per hour Hour of Collection Inside n=264 Outside n= Hour of Collection Inside n=242 Outside n=129 B Figure 2. The 12-h biting patterns of (A) An. darlingi; (B) An. albimanus and; (C) An. pseudopunctipennis, captured inside and outdoors of an experimental hut during 25 all-night collections performed from January 22 to May 23.

7 Journal of Vector Ecology Average An. darlingi / 2h collection A Oct 1 Nov 1 Dec 1 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT Nov 2 // Average Rainfall (mm) Month Average An. darlingi / 2h collection B Average River Level (m) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT // 1 Month Average An. darlingi / 2h collection C JAN FEB MAR APR MAY JUN JUL AUG SEP OCT // Average High Temperature (C ) Month Average An. darlingi / 2h collection D JAN FEB MAR APR MAY JUN JUL AUG SEP OCT Month Figure 3. Seasonal variation of An. darlingi population densities captured from a total of 31 2-h collections performed from January-May and July-October 22 overlaid with average monthly (A) rainfall (mm); (B) river level (m); (C) high temperature ( C) and; (D) low temperature ( C). // Average Low Temperature (C )

8 52 Journal of Vector Ecology June 26 Figure 4. The total number of An. darlingi captured per night during 31 2-h collections from January-October 22 (1=January; 2=February, etc.) with overlay of mean river level difference for the 14 days prior to the collection. A negative difference represents a decrease in river level. performed to determine if a stronger association could be described. The mean difference in river height for the 14 days prior to the date of a collection (i.e., oviposition to adult emergence with 24 h pre-host seeking) was calculated and correlated with the total number of An. darlingi captured for the individual night (Figure 4). Again, there was no indication of a significant association (r=-.279; p=.128). Seasonal trends of other anophelines collected at the study site were also examined (Figure 5). Anopheles albimanus had the highest population densities (57%; 93/162) during the months of July to October, and similarly, An. vestitipennis exhibited the highest catches (22/24) during August to October. The majority of An. pseudopunctipennis (65%; 11/ 17) were captured in the month of April but adults were also collected August to October. Anopheles punctimacula females exhibited two peak population densities, one in January (21/ 65) and another in August (21/65). Anopheles apicimacula populations were highest during the months of January to April (28/3). Natural sporozoite infections A total of 1,242 An. darlingi, 373 An. albimanus, 216 An. pseudopunctipennis, 88 An. punctimacula, 78 An. vestitipennis, 44 An. apicimacula, 14 An. gabaldoni, and 4 An. crucians specimens from both the all-night and 2-h seasonal collections were screened for malaria sporozoites using the VecTest. None of the total 1,134 pools assayed were found to be infected. DISCUSSION The following study incorporated the use of an experimental hut to describe the 12-h biting behavior and seasonal variation of An. darlingi adult populations in the Cayo District of Belize. An experimental hut was used because there is less control over the use of indigenous homes, which can lead to data confounding. For example, the increase in persons indoors (i.e., indoor:outdoor ratio assessment) and, most importantly, local village huts may have been treated with insecticide. However, results from the present study can be generalized to indigenous structures that have no spray history because the experimental hut was assembled based on the sleeping area, design, and construction materials of 1 homes within a local village adjacent to the study site. Results from 12-h collections show that An. darlingi populations exhibited a bimodal biting pattern. The strongest indoor and outdoor biting activity occurred 3 h post-sunset with relatively high numbers of females continuing to bite throughout the night at both indoor/outdoor stations. Because no other 12-h An. darlingi collection studies have been conducted in Belize, comparisons of same-species all-night biting activity within other regions of the country cannot be made. However, in studies from South America, An. darlingi has been shown to exhibit unimodal (Hudson 1984, Lourenco de Oliveira et al. 1989, Rozendaal 199), bimodal (Forattini 1987, Tadei et al. 1988) as well as trimodal (Pajot et al. 1977) biting activity patterns. A study from Honduras has also described a unimodal biting pattern for An. darlingi (Zimmerman and Rangel 199). The all-night activity pattern presented here is most similar to the activity pattern of An. darlingi reported from Brazil, where a major peak in biting occurred within 3 h post-sunset and another minor peak occurred during the last 2 h pre-sunrise with biting continuing throughout the night (Charlwood and Wilkes 1979, Charlwood and Alecrim 1988, Roberts et al. 1987, Klein and Lima 199). It should be noted that no attempt was made in the present study to age-grade captured females. All-night collections were performed during periods when An. darlingi adult

9 Journal of Vector Ecology 53 // Figure 5. Seasonal variation of other anopheline populations captured at the study site during 31 2-h collections performed during January-May and July-October 22. densities were highest (i.e., dry season). In addition, the study site was selected based on its proximity to major breeding sites (i.e., the Sibun River). This ensured sufficient numbers of females would be captured to discern nightly biting patterns. However, this study design may have introduced bias in defining the overall 12-h biting pattern because more nulliparous females were present and may have been collected than parous females during a particular time of night. Previous studies defining the age-composition of An. darlingi in Brazil have described variations in the time of peak biting between nulliparous and parous adults (Charlwood and Wilkes 1979). The majority of nulliparous females were captured during crepuscular hours while older females were most often collected later at night. Such behavioral differences have implications in disease transmission and should be assessed in future studies. Throughout South America, An. darlingi has been described as an endophagic vector (Giglioli 1948, Elliott 1972, Hudson 1984, Charlwood 198, Fleming 1986, Roberts et al.1987). The present study also indicates an endophagic biting behavior of An. darlingi. The outdoor to indoor biting ratios of all-night (O:I=1.:1.4) and 2 h seasonal (O:I=1.:.95) collections, however, define a stronger endophagic response than that previously described for other An. darlingi populations located within Belize (O:I=1.:.6; Roberts et al. 22, O:I=1.:.53; Grieco 4 ). Collections within both of these previous studies were performed during the first 2 h post-sunset at local houses. The discrepancy between the degree of endophagic behavior reported here and previous studies can most likely be contributed to the previous spray history of the indigenous homes that could result in lower numbers of entering mosquitoes. Differences between the overall endophagic biting response of An. darlingi in 2-h seasonal (O:I=1.:.95) and 12-h all-night (O:I=1.:1.4) collections performed in the present study are most likely a reflection of natural movement patterns undertaken by females toward a blood meal source. No systematic surveys have been conducted in Belize regarding the resting behavior of host-seeking An. darlingi; however, studies from South America have described behavior patterns of outdoor resting during pre-sunset hours before entering homes (Roberts et al. 1987). Data from 12-h and 2- h collections reported in the present study show a portion of these An. darlingi that congregate outside during the early evening will commence feeding on an out-of-doors host while another portion will continue to move inside to feed. Data suggest the number of females entering a home to bite will be greater than those feeding outdoors only after a resting period and/or population threshold is reached. At the present research site, this time period began during the second hour post-sunset and peaked at 3 h post-sunset. Because the seasonal collections were conducted for only 2 h post-sunset, the time of peak indoor feeding for An. darlingi was not captured as reflected in the weaker overall endophagic biting ratio. Grieco 4 emphasized the need to conduct all-night collections for describing biting behaviors when it was shown that An. vestitipennis populations in southern Belize also exhibited a much stronger indoor biting activity from 12-h (O:I=1.:.9) versus 2-h (O:I=1.:.53) collections. Similar results have been shown in biting studies of An. darlingi in South America (Voorham 22). However, seasonal biting ratios presented in the current study may be biased due to the study design (i.e., experimental hut located adjacent to breeding sites) for similar reasons stated earlier. More importantly, An. darlingi was found biting inside the experimental hut at the present research site during both types of collections confirming the tendency of this species to enter homes to feed. Because of their importance in malaria transmission in the coastal lowlands and foothills throughout the Central and South American region (Forattini 1962, Rodriguez and Loyola 1989, Ramsey et al. 1986, PAHO 1991), general descriptions

10 54 Journal of Vector Ecology June 26 of biting behaviors for both An. albimanus and An. pseudopunctipennis from the present study have also been conducted. Anopheles albimanus is considered to be exophagic throughout its geographic distribution (Brown et al. 1991, Rodriguez et al. 1992, Roberts et al. 1993). Other all-night studies conducted in both the southern Toledo District (Grieco et al. 2) and northern Orange Walk District of Belize (Bangs 6 ) have also described An. albimanus as having a strong exophagic biting behavior. Interestingly, Anopheles albimanus biting populations collected over a 12-h period in the present research indicate an endophagic response (O:I=1.:1.). This discrepancy is most likely due to the low numbers (527) of specimens captured. The all-night biting pattern and low numbers of females captured throughout the night described here, however, is similar to the 12-h behavior patterns described for An. albimanus in other regions of Belize (Bangs 6, Grieco 4 ). A more exophagic response of An. albimanus is defined (i.e., O:I=1.:.71) when 2-h seasonal collection data from the present research is examined. These results are similar to previous collections conducted during 2 h post-sunset at indigenous homes in the same region as the present study site (Roberts et al. 1993, Roberts et al. 22). The biting behavior of An. pseudopunctipennis has been defined in limited areas throughout its geographic distribution (Fernandez-Salas 1992, Zimmerman 1992). Currently, no information is available for all-night biting patterns in Belize. As such, this study offers further insight into the behavior of this important vector. In the Tapachula Foothills of southern Mexico, An. pseudopunctipennis was shown to be exophagic with a 12-h bimodal activity pattern (Fernandez-Salas 7 ). Although the total number (371) of An. pseudopunctipennis captured in 12-h collections at the Belize study site was low, data also indicate An. pseudopunctipennis to exhibit a bimodal biting pattern. However, a more prominent endophagic behavior was exhibited (O:I=1.:1.88) than either An. darlingi (O:I=1.:1.4) or An. albimanus (O:I=1.:1.). In addition, An. pseudopunctipennis exhibited a longer period of peak biting, with the cumulative majority not being captured until 1 h later than the other two vector species. Other studies by our group were dedicated to An. pseudopunctipennis in earlier years (Manguin et al. 1996b, Roberts et al. 1996). In subsequent studies this species was de-emphasized as an important vector due to its relative scarcity in most regions. However, results from the present study continue to support a role for An. pseudopunctipennis in malaria transmission in Belize and further studies focusing on this species are 6 Bangs, M.J The susceptibility and behavioral response of Anopheles albimanus Weidemann and Anopheles vestitpennis Dyar and Knab (Diptera: Culicidae) to insecticides in northern Belize. Doctoral Dissertation, Uniformed Services University of the Health Sciences. 448 pp. 7 Fernandez-Salas, I Bionomics of the primary malaria vector, Anopheles pseudopunctipennis, in the Tapachula foothill area of southern Mexico. Doctoral Dissertation, Uniformed Services University of the Health Sciences. 157 pp. warranted. Descriptions of An. darlingi adult seasonal population distributions presented here are the first report from the central Cayo District of Belize. Densities were highest during the months of January, May, and July, all corresponding to the transition periods between the wet and dry seasons, with May exhibiting the highest density. The only other seasonal study of An. darlingi in Belize is from the southern Toledo District where An. darlingi exhibited large adult populations during the dry season months of March and April, as well as the transition month of June, with the highest density (39/1,2) also occurring in May (Grieco 4 ). In South America, An. darlingi has shown high variability in seasonal distributions with some reports describing populations to be highest during the wet season (Rozendaal 199), dry season (Rozendaal 1992), or correlated with transitional months between seasons (Ferraroni and Hayes 1979, Charlwood 198). These differences reflect the geographical variation in larval breeding sites. The present study found no significant associations between changes in monthly An. darlingi adult densities and rainfall, river level, or temperature during the collection months. Examination of correlations, however, did prove insightful. Anopheles darlingi densities were higher in months during or following low river levels (i.e. January to May). As described in previous studies in South America (Rozendaal 1992), low river levels most likely lead to the persistence of detritus mats (i.e., preferred An. darlingi larval habitats) within the river and a low probability of flushing. Similar results were seen in the Toledo District of Belize, where An. darlingi populations were negatively associated with river levels (Grieco 4 ). Although, there was not a significant association, a trend was seen between negative mean differences (i.e., decreasing river levels) and high An. darlingi collection nights. Seasonal distributions of other adult anopheline species at the research site were similar to those described for other regions of Belize. Anopheles albimanus was present during each of the collection months but was at highest densities during the wet season (i.e., July to October). These data confirm other studies that have found highest An. albimanus populations during the wet season months in both the southern Toledo (Grieco 4 ) and northern Orange Walk Districts (Grieco et al. 25b) of Belize. The preferred breeding habitat of this vector in Belize are marshes containing cyanobacterial mats (Rejmankova et al. 1993); however, larvae have also been found breeding in rivers (Manguin et al. 1996a, Achee 8 ), roadside ditches, and flooded fields (Grieco 4 ). This flexibility in oviposition site selection will maintain populations throughout the year and will allow densities to increase following periods of heavy rain due to increased habitat availability. Anopheles vestitipennis, another important vector, was also captured in highest numbers during the wet season months of July to October in the present study. Again, these results are similar 8 Achee, N. 24. A study on the bionomics of Anopheles darlingi Root (Diptera: Culicidae) in Belize, Central America. Doctoral Dissertation, Uniformed Services University of the Health Sciences. 3 pp.

11 Journal of Vector Ecology 55 to those described from previous research (Grieco 4, Grieco et al. 25b). Larvae of this species are found predominately in tall dense macrophyte marshes (Rejmankova et al. 1998, Grieco et al. 25b), but have also been collected from flooded forests during the rainy season (Grieco 4 ). Seasonal collections from the present research indicated that An. pseudopunctipennis adult populations were at their highest density in the month of April. This species has been previously characterized to breed in sunlit pools containing filamentous algae along waterways (Manguin et al. 1996b, Rejmankova et al. 1993). The population increase described at the current study site corresponds to the period of lowest recorded monthly precipitation (.7 mm). A decrease in precipitation causes river levels to fall, exposing rock pools, and resulting in more larval habitats. Extrapolating the association between An. darlingi adult populations at the present research site and monthly malaria distribution for the closest villages was not performed due to the low number of reported cases (5) during the study year of 22, the distance of the research site from these village (i.e., approx. 5 miles) and the inability to detect natural sporozoite infections in sample pools of An. darlingi. The fact that all mosquito pools screened were negative for Plasmodium spp. infections was expected. Biting behavior studies were conducted at this location specifically for its high population densities of the target species but just as importantly for its isolation from other competing human-hosts. Performing the VecTest was still appropriate, however, because randomly scattered family groups did inhabit the immediate area. There is a high-level of confidence in the negative VecTest results presented in the current study due to previous reports comparing the VecTest to the standard circumsporozoite enzyme-linked immunosorbent assay (ELISA) with pools of An. darlingi collected in Guatemala, Peru and Venezuela (Ryan et al. 22). In conclusion, this research is the first to define the allnight activity patterns of An. darlingi in Belize and to describe the changes in An. darlingi adult densities by season for the Cayo District. Because of the propensity of An. darlingi to bite indoors throughout the night in relatively high numbers, its role as an important malaria vector is further confirmed. Future efforts in Belize should focus on advanced adult behavioral studies of An. darlingi including defining differences in biting patterns with regard to parity and the time of house entering/exiting to determine the time period in which An. darlingi will remain in a house before and after searching for a bloodmeal. Also, the effect of insecticide spraying on adult biting and movement patterns should be evaluated. There is a need to conduct village based epidemiological studies involving active case detection of malaria infections with concurrent systematic vector surveys to provide detailed information on the microepidemiology of malaria transmission. Such information would describe the definitive role of An. darlingi in disease transmission throughout the country. Acknowledgments The authors express their gratitude to the families of both Mr. Ramon Galvez Sr. and Mr. Novelo for generously allowing the use of their property for the success of the study. This research was supported by the NIH-NSF Ecology of Infectious Diseases program, Grant # R1 AI Disclaimer: The opinions and assertions contained in this article are not to be considered as official or as reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences. REFERENCES CITED Achee, N.L., J.P. Grieco, R.G. Andre, E. Rejmankova, and D.R. Roberts. 25. A mark-release-recapture study utilizing a novel portable hut design to define the flight behavior of Anopheles darlingi Root in Belize, Central America. J. Am. Mosq. Contr. Assoc. (in press). Achee, N.L., C.T. Korves, M.L. Bangs, E. Rejmankova, M. Lege, D. Curtin, H. Lenares, Y. Alonzo, R.G. Andre, and D.R. Roberts. 2. Plasmodium vivax polymorphs and Plasmodium falciparum circumsporozoite proteins in Anopheles (Diptera: Culicidae) from Belize, Central America. J. Vector Ecol. 25: Arruda, M., M.B. Carvalho, R.S. Nussenzweig, M. Maracic, A.W. Ferreira, and A.H. Cochrane Potential vectors of malaria and their different susceptibility to Plasmodium flaciparum and Plasmodium vivax in northern Brazil identified by immunoassay. Am. J. Trop. Med. Hyg. 35: Brown, D.N., M.H. Rodriguez, J.I. Arredondo, E.G. Loyola, and M.C. Rodriguez Age structure and abundance levels in the entomological evaluation of an insecticide used in the control of Anopheles albimanus Wiedemann in southern Mexico. J. Am. Mosq. Contr. Assoc. 7: Charlwood, J.D. and W.D. Alecrim Capture-recapture experiments with Anopheles darlingi in Rondonia, Brazil. Ann. Trop. Med. Parasitology 83: Charlwood, J.D Observations on the bionomics of Anopheles darlingi Root (Diptera: Culicidae) from Brazil. Bull. Entomol. Res. 7: Charlwood, J.D. and T.J. Wilkes Studies on the agecomposition of samples of Anopheles darlingi Root (Diptera: Culicidae) in Brazil. Bull. Entomol. Res. 69: Da Silva-Vasconcelos, A., M.Y.N. Kato, E.N. Mourao, R.T.L. de Souza, R.N. Lacerda, A. Sibajev, P. Tsouris, M.M. Povoa, H. Momen, and M.G. Rosa-Fritas. 22. Biting indices, host-seeking activity and natural infection rates of anopheline species in Boa Vista, Roraima, Brazil from 1996 to Mem. Inst. Oswaldo Cruz 97: Davis, N.C A note on the malaria-carrying anophelines in Belem, Para, and in Natal, Rio Grande do Norte, Brazil. Riv. Malar.1: Deane, L.M Malaria vectors in Brazil. Mem. Inst. Oswaldo Cruz 81 (Suppl.): 5-14.

12 56 Journal of Vector Ecology June 26 Deane, L.M., O.R. Causey, and M.P. Keane Notas sobre a distribuicao e a biologia dos anofelinos das regioes Nordestina e Amazonica do Brasil. Revista del Servicio Especial de Saude Publica 1: Elliot, R The influence of vector behavior on malaria transmission. Am. J. Trop. Med. Hyg. 21: Ferraroni, J.J. and J. Hayes Drug-resistant falciparum malaria among the Mayongong Indians in the Brazilian Amazon. Am. J. Trop. Med. Hyg. 28: Fleming, G Biology and Ecology of Malaria Vectors in the Americas. J. Pan Am Hlth. Org. PNSP/ Forattini, O.P Comportamento exofilo de Anopheles darlingi Root, em regiao meridional do Brasil. Rev. Saude Pub., Sao Paulo 21: Forattini, O.P Entomologia medica vol. I. Faculdade de Higiene e Saude Publica, Sao Paulo, Brazil. 662 pp. Giglioli, G An investigation of the house-frequenting habits of mosquitoes of the British Guiana coastland in relation to the use of DDT. Am. J. Trop. Med. Hyg. 28: Grieco, J.P., N.L. Achee, D.R. Roberts, and R.G. Andre. 25a. Comparative susceptibility of three species of Anopheles from Belize, Central America to Plasmodium falciparum (NF-54). J. Am. Mosq. Contr. Assoc. (In press). Grieco, J.P., R. Vogtsberger, N.L. Achee, P. Macek, I. Briceno, R. King, E. Vanzie, R. Andre, D. Roberts and E. Rejmankova. 25b. Evaluation of habitat management strategies for the reduction of malaria vectors in northern Belize. J. Am. Mosq. Contr. Assoc. (In press). Grieco, J.P., N.L. Achee, R.G. Andre, and D.R. Roberts. 2. A comparison study of house entering and exiting behavior of Anopheles vestitipennis (Diptera: Culicidae) using experimental huts sprayed with DDT or deltamethrin in the southern district of Toledo, Belize, C.A. J. Vector Ecol. 25: Harbach, R.E., D.R. Roberts, and S. Manguin Variation in the hindtarsal markings of Anopheles darlingi (Dipt.: Culicidae) in Belize. Mosq. Syst. 25: Herrera, S., M.F. Suarez, G.I. Sanchez, M.L. Quinones, and M. Herrera Uso de la tecnica immuno-radiometrica (IRMA) en Anopheles de Colombia para la identificacion de esporozoitos de Plasmodium. Col. Med. 18: 2-6. Hudson, J.E Anopheles darlingi Root (Diptera:Culicidae) in the Suriname rain forest. Bull. Entomol. Res. 74: Klein, T.A., M.S. Tada, and J.B.P. Lima Comparative susceptibility of anopheline mosquitoes to Plasmodium falciparum in Rodonia, Brazil. Am. J. Trop. Med. Hyg. 44: Klein, T.A. and J.B.P. Lima Seasonal distribution and biting patterns of Anopheles mosquitoes in Costa Marques, Rondonia, Brazil. J. Am. Mosq. Contr. Assoc. 6: Kumm, H.W. and L.M. Ram Observations on the Anopheles of British Honduras. Am. J. Trop. Med. 21: Lourenco-de-Oliveira, R., A.E. Guimaraes, M. Arle, T.F. da Silva, M.G. Castro, M.A. Motta, and L.M. Deane Anopheline species, some of their habits and relation to malaria in endemic areas of Rondonia State, Amazon region of Brazil. Rev. Bras. Biol. 49: Loyola, E.G., J.I. Arredondo, M.H. Rodriguez, D.N. Brown, and M.A. Vaca-Marin Anopheles vestitipennis, the probable vector of Plasmodium vivax in the Lacandon forest of Chiapas, Mexico. Trans. R. Soc. Trop. Med. Hyg. 85: Manguin, S., D.R. Roberts, R.G. Andre, E. Rejmankova, and S. Hakre. 1996a. Characterization of Anopheles darlingi (Diptera: Culicidae) larval habitats in Belize, Central America. J. Med. Entomol. 33: Manguin, S., D.R. Roberts, E.L. Peyton, E. Rejmankova, and J. Pecor. 1996b. Characterization of Anopheles pseudopunctipennis larval habitats. J. Am. Mosq. Contr. Assoc. 12: Oliveira-Ferreira, J., R. Lourenco-de-Oliveira, A. Teva, L.M. Deane, and C.T. Daniel-Riberio Natural malaria infections in anophelines in Rondonia State, Brazilian Amazon. Am. J. Trop. Med. Hyg. 43: 6-1. Padilla, N., P. Molina, J. Juarez, D. Brown, and C. Cordon- Rosales Potential malaria vectors in northern Guatemala. J. Am. Mosq. Contr. Assoc. 8: Pajot, F.X., F. LePont, J.F. Molez, and N. DeGaller Agressivite de Anopheles (Nyssorhynchus) darlingi Root, 1926 (Diptera, Culicidae) en Guyana Francaise. Cah O.R.S.T.O.M. Ser. Entomol. Med. Parasitol. 15: Pan American Health Organization (PAHO) Biologia y ecologia de Anopheles albimanus Wiedeman en America Central y Panama. Informe Fianl. 55 pp. Ramsey, J.M., E. Salinas, D.N. Brown, and M.H. Rodriguez Plasmodium vivax sporozoite rates from Anopheles albimanus in southern Chiapas, Mexico. J. Parasitol. 8: Ramsey, J.M., D.N. Brown, J.L. Aron, R.L. Beautdoin, and J.F. Mendez Field trials in Chiapas, Mexico, of a rapid detection method for malaria in anopheline vectors with low infection rates. Am. J. Trop. Med. Hyg. 35: Rejmankova, E., A. Harbin-Ireland, and M. Lege. 2. Bacterial abundance in larval habitats of four species of Anopheles (Diptera: Culicidae) in Belize, Central America. J. Vector Ecol. 25: Rejmankova, E., K.O. Pope, D.R. Roberts, M.G. Lege, R. Andre, J.P. Grieco, and Y. Alonzo Characterization and detection of Anopheles vestitipennis and Anopheles punctimacula (Diptera: Culicidae) larval habitats in Belize with field survey and SPOT satellite imagery. J. Vector Ecol. 23: Rejmankova, E., D.R. Roberts, R.E. Harbach, J. Pecor, E.L. Peyton, S. Manguin, R. Krieg, J. Polanco, and L. Legters Environmental and regional determinants of Anopheles (Diptera: Culicidae) larval distribution in Belize, Central America. Environ. Entomol. 22: Roberts, D.R., S. Manguin, E. Rejmankova, R. Andre, R.E. Harbach, E. Vanzie, S. Hakre, and J. Polanco. 22. Spatial distribution of adult Anopheles darlingi and Anopheles albimanus in relation to riparian habitats in

13 Journal of Vector Ecology 57 Belize, Central America. J. Vector Ecol. 27: Roberts, D.R., J.F. Paris, S. Manguin, R.E. Harbach, R. Woodruff, E. Rejmankova, J. Polanco, B. Wullschleger, and L.L. Legters Predictions of malaria vector distribution in Belize based on multispectral satellite data. Am. J. Trop. Med. Hyg. 54: Roberts, D.R., O. Chan, J. Pecor, E. Rejmankova, S. Manguin, J. Polanco, and L. Legters Preliminary observations on the changing roles of malaria vectors in southern Belize. J. Am. Mosq. Contr. Assoc. 9: Roberts, D.R., W.D. Alecrim, A.M. Tavares, and M.G. Radke The house-frequenting, host-seeking and resting behavior of Anopheles darlingi in southeastern Amazonas, Brazil. J. Am. Mosq. Contr. Assoc. 3: Rodriguez, M.H., D.N. Brown, C. Villarreal, J.I. Arredondo, and E.G. Loyola Gonotrophic cycle and survivorship of Anopheles albimanus (Diptera: Culicidae) in southern Mexico. J. Med. Entomol. 29: Rodriguez, M.H. and E.G. Loyola Situacion epidemilogia actual y perspectives de la investigacion entomolgicas del paludismo en Mexico. Mem. IV Simp Nac. Entomol. Med. Vet. pp Rozendaal, J.A Relations between Anopheles darlingi breeding habitats, rainfall, river level and malaria transmission rates in the rain forest of Suriname. Med. Vet. Entomol. 6: Rozendaal, J.A Observations on the distribution of anophelines in Suriname with particular reference to the malaria vector Anopheles darlingi. Mem. Inst. Oswaldo Cruz 85: Rozendaal, J.A., J.P.M. van Hoof, J. Voorham, and B.F.J Oostburg Behavioral responses of Anopheles darlingi in Suriname to DDT residues on house walls. J. Am. Mosq. Contr. Assoc. 5: Ryan, J.R., K. Dave, K.M. Collins, L. Hochberg, J. Sattabongkot, R.E. Coleman, R.F. Dunton, M.J. Bangs, C.M. Mbogo, R.D. Cooper, G.B. Schoeler, Y. Rubio-Palis, M. Margis, L.I. Romer, N. Padilla, I.A. Quakyi, J. Bigoga, R.G. Leke, O. Akinpelu, B. Evans, M. Walsey, P. Patterson, R.A. Wirtz, and A.S. Chan. 22. Extensive multiple test center evaluation of the VecTest malaria antigen panel assay. Med. Vet. Entomol. 16: Tadei, W.P. and B.D. Thatcher. 2. Malaria vectors in the Brazilian Amazon: Anopheles of the subgenus Nyssorhynchus. Rev. Inst. Med. Trop. S. Paulo. 42: Tadei, W.P., B.D. Thatcher, J.M. Santos, V.M. Scarpassa, I.B. Rodrigues, and M.S. Rafael Ecological observations on anopheline vectors of malaria in the Brazilian Amazon. Am. J. Trop. Med. Hyg. 59: Tadei, W.P., J.M.M. Santos, W.L.S. Costa, and V.M. Scarpassa Biologia de anofelinos amazonicos. XII. Ocorrencia de especies de Anopheles, dinamica de transmissao e controle de malaria na zona urbana de Ariquemes (Rondonia). Rev. Inst. Med. Trop. Sao Paulo. 3: Voorham, J. 22. Intra-population plasticity of Anopheles darlingi s (Diptera, Culicidae) biting activity patterns in the state of Amapa, Brazil. Rev. Saude Pub. 36: Wilkerson, R.C. and D. Strickman Illustrated key to the female anopheline mosquitoes of Central America and Mexico. J. Am. Mosq. Contr. Assoc. 6: Zimmerman, R.H Ecology of malaria vectors in the Americas and future direction. Mem. Inst. Oswaldo Cruz. Suppl. III 87: Zimmerman, R.H. and Y. Rangel Memoria del Simposio Latinoamericano sobre biologia y control de vectores de enfermedades tropicales. Bol. Dir. Malariol. Y San. Amb. Suppl 1: 1-97.