Comparison of Mosquito (Diptera: Culicidae) Fauna Characteristics of Forested Wetlands in Sweden

CONSERVATION BIOLOGY AND BIODIVERSITY Comparison of Mosquito (Diptera: Culicidae) Fauna Characteristics of Forested Wetlands in Sweden MARTINA SCHÄFER AND JAN O. LUNDSTRÖM Department of Population Biology, Evolutionary Biology Center, Uppsala University, Norbyvägen 18D, SE-75236 Uppsala, Sweden Ann. Entomol. Soc. Am. 94(4): 576Ð582 (2001) ABSTRACT We studied mosquito faunas of four wetlands from northern to southern Sweden by trapping female mosquitoes in June, July, and August. A total of 52,298 individuals comprising 32 species in Þve genera were identiþed. The number of species increased from 10 and 12 in the two northern wetlands, to 16 in the central Sweden study area, to 24 in the wetland in southern Sweden. For a further characterization of mosquito fauna diversity, we organized all species recorded from Sweden into 14 functional groups based on biological and life history characteristics. The number of groups increased from three in the two northern study areas, to eight in central Sweden, and 13 in the southernmost study area. All functional groups present at one site were also present at the sites located farther south. Most successful species were univoltine, respiring from the water surface, laying their eggs on soil, overwintering in the egg stage, preferring forested or partly forested habitats, and having mammals as for blood meals. The mosquito faunas of the two northern study areas were similar and lacked several of the functional groups occurring further south. The mosquito fauna of the study area in central Sweden included species feeding on birds and with overwintering larvae. In the southernmost study area, 13 out of 14 functional groups were found, indicating a large variety of habitats. Our results demonstrated a southward increase in the number of both mosquito species and functional groups in forested wetlands. KEY WORDS mosquitoes, wetlands, diversity, functional groups IN THE NORTHERN region, a mixture of forest, forested wetlands, mires, and lakes is a characteristic landscape mosaic (Sjöberg and Ericson 1997). Mosquitoes can be found in almost all kinds of environments associated with water, and are most common in wetlands and humid forested areas. In Sweden, 48 species and Þve genera of mosquitoes have been recorded (Dahl 1997). They differ in requirements for larval (Laird 1988) and adult (Jaenson 1988, Lundström et al. 1996) habitats, overwintering strategy, (Kettle 1995), preference for blood-meal (Service 1971, Tempelis 1975, Jaenson and Niklasson 1986), and number of generations per year (Gutsevich et al. 1974). A detailed knowledge on the biology and life history is available for most mosquito species. Ecological characteristics of species can be linked with characteristics of habitats and vice versa. The feeding behavior of mosquitoes provides information on the presence of preferred in an area, such as mammals, birds, or amphibians. Habitat requirements include the presence of large deciduous trees for mosquito species that use tree holes as larval habitats, or rich submerged vegetation for the species with larvae and pupae taking oxygen from submerged plant stems. Other less specialized species breed in common forest ponds formed during the spring throughout Sweden. Diversity and conservation studies are often focused on species lists, but a broader approach sometimes seems more appropriate (Gaston 1996, New 1999). In this article, we compared mosquito diversity in four study areas from northern to southern Sweden, based on the presence and number of functional groups. The aim was to look for characteristics of mosquito faunas of different wetlands, and for patterns over the latitudinal gradient. Materials and Methods Study Areas. Mosquitoes were collected in four study areas in Sweden (Fig. 1), located in the northern boreal zone (Abisko and Allavaara), the southern boreal zone (Bergsåker) and the temperate zone (Egeside Sjö), respectively (Ahti et al. 1968). The Abisko study area (68 21 N, 18 49 E) is located 200 km north of the Arctic Circle on the south shore of Lake Torneträsk, ranging from 385 to 600 m asl (tree line). The forest consists mainly of mountain birch (Betula pubescens Ehrhart subsp. tortuosa Nyman) and dwarf birch (Betula nana L.), as well as different willow species (Salix sp.), interrupted by heaths of mountain avens (Dryas octopetala L.). The whole area is ßooded in spring by melting snow. The Allavaara study area (67 15 N, 20 05 E) is situated 65 km north of the Arctic Circle, 2 km south of the Allavaara village, in the meanders of the River Allajåkka. This area has a ßat topography and is dominated by extensive mires and small lakes and rivers. The forest consists mainly of mountain birch and wil- 0013-8746/01/0576Ð0582$02.00/0 2001 Entomological Society of America

July 2001 SCHÄFER AND LUNDSTRÖM: MOSQUITO FAUNAS OF FORESTED WETLANDS 577 Fig. 1. Location of the four study areas for mosquito collections in Sweden lows, together with stands of pine (Pinus sylvestris L.) and Norwegian spruce [Picea abies (L.) Karsten]. The ground is covered mainly with tufts of sedges (Carex spp.), the density of other herbs being intermediate. Uncontrolled and extensive annual ßooding in early summer by melting snow, as well as runoff from nearby mountains, turns the whole area into a wetland. Data from the Bergsåker study area (62 23 N, 17 17 E) are based on the study by Francy et al. (1989), in which one of the authors of the current study (J.O.L.) identiþed a major portion of the mosquitoes. Mosquitoes were sampled on a peninsula between the Selångersån River and Lake Selångersfjärden. The wetland area, surrounded by agricultural land, has dense stands of alder (Alnus glutinosa L.), willows, birch, some aspen (Populus tremula L.), European bird cherry (Prunus padus L.), and occasionally pine and spruce. Herbs are abundant, and stands of sedges and reed (Phragmites australis Trinius) occur as well. The river ßoods annually, usually in spring, but ßooding is limited by an existing channel between the lake and the Baltic Sea. The Egeside Sjö study area (55 52 N, 14 12 E) is situated 17 km south of Kristianstad, next to the river Helge å. Manipulation of the river ßow at the end of the 18th century turned the former lake into a swampy forest area. The study area is surrounded by agricultural land, and is characterized by stands of forests consisting mainly of alder, willow, birch, aspen, and ash (Fraxinus excelsior L.); European bird cherry, spruce, and pine trees are less common. The herb layer is dense. Forested areas alternate with open stands of reed and sedges. The whole area is ßooded in springtime. Sampling Methods. Centers for Disease Control (CDC) miniature light traps, with carbon dioxide as an attractant, were used to collect adult female mosquitoes. This method has the advantage of sampling a large variety of mosquito species (Service 1993). The traps were hung in trees at 1.5 m above ground. They were activated early in the evening for a duration of 12Ð14 h until the next morning. Collected mosquitoes were anesthetized with carbon dioxide, dispensed into plastic ampoules and immediately killed by freezing on dry ice. Mosquitoes were stored on dry ice and at 70 C until identiþcation. For identiþcation to species, keys by Mohrig (1969), Gutsevich et al. (1974), and Wood et al. (1979) were used. At Abisko, sampling was conducted in 1996 and 1997, using six traps for one night on three occasions each year. In 1996, mosquitoes were collected on 8Ð9, 17Ð18, and 22Ð23 July (results from two traps only); in 1997, on 21Ð22 and 28Ð29 July and 4Ð5 August. At Allavaara, collections from three traps run over a series of three nights in a row were used for each sampling period. The sampling periods in Allavaara were on 16Ð19 June, 3Ð6 July, and 27Ð30 July 1991. We identiþed the entire catch from one trap during the whole study period, and included randomly chosen subsamples from the other two traps. At Bergsåker, data from each sampling period consist of the number of mosquitoes collected by six traps run over a period of two nights in a row, with the exception of the Þrst sampling period in June when there was an interruption of 1 d between trap nights. Mosquitoes were sampled on 25Ð26 and 27Ð28 June, 23Ð25 July, and 12Ð14 August 1985. At Egeside Sjö, three traps were run over a series of three nights in a row per sampling period. Collections were made on 2Ð5 June, 29 JuneÐ 2 July and 13Ð16 August 1998. Analysis. The sampling effort was not consistent for the study areas. From Bergsåker, there were 36 samples (6 traps 2 nights 3 sampling periods) whereas from Allavaara and Egeside Sjö, there were 27 samples (3 traps 3 nights 3 sampling periods). From Abisko, during a 2-yr period, we had 32 samples (2 traps/night 6 traps/night 5 sampling periods). All collections were from different years. Nevertheless, we focused on comparing patterns of biological and life history characteristics of different faunas, with less emphasis on absolute species composition and abundance. Because the collections were made during different years, and usually for a 1-yr period per study area, our results represent a characterization of the wetlandsõ mosquito faunas at each time. We grouped all mosquito species recorded from Sweden and identiþable as females into 14 functional groups based on six biological and life-history char-

578 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 94, no. 4 Table 1. The mosquito species of Sweden organized into 14 functional groups based on six biological and life history characteristics Functional group Characteristics Species 1 Larvae respire from water surface An. maculipennis s.l. Cs. alaskaensis Females overwinter Cs. annulata Cs. bergrothi Cs. subochrea Variable habitats Cx. pipiens molestus Cx. pipiens pipiens 2 Larvae respire from water surface Cx. torrentium Females overwinter Birds as preferred blood meal Variable habitats 3 Larvae respire from water surface Cx. territans Females overwinter Amphibians as preferred blood meal Clean waters with rich vegetation as Preferred habitat 4 Larvae respire from water surface An. claviger Larvae overwinter Open/edges as preferred habitats 5 Larvae respire from water surface Ae. rusticus Larvae/eggs overwinter Forest as preferred habitat 6 Larvae respire from water surface Cs. fumipennis Cs. morsitans Larvae/females overwinter Cs. ochroptera Birds as preferred blood meal Forest as preferred habitat 7 Larvae respire from water surface Ae. cantans Ae. cataphylla Ae. communis Ae. diantaeus Ae. hexodontus Forest as preferred habitat Ae. impiger Ae. intrudens Ae. nigripes Ae. pionips Ae. pullatus Ae. punctor Ae. refiki 8 Larvae respire from water surface Ae. annulipes Ae. excrucians Forest/edges as preferred habitat acteristics (Table 1), including oxygen source for larvae, egg laying sites, overwintering life stages, preferred for blood meal, preferred larval habitat, and number of generations per year. The mosquito Table 1. Continued Functional group Characteristics Species 9 Larvae respire from water surface Ae. cinereus Forest/edges as preferred habitat 10 Larvae respire from water surface Ae. cyprius Ae. flavescens Ae. leucomelas Ae. riparius Open area as preferred habitat 11 Larvae respire from water surface Ae. caspius Ae. detritus Ae. dorsalis Open area as preferred habitat, often saline Water 12 Larvae respire from water surface Ae. rossicus Ae. sticticus Ae. vexans River inundation areas as preferred habitat 13 Larvae respire from water surface Ae. geniculatus Eggs are laid in tree holes Eggs/larvae overwinter Deciduous trees as preferred habitat 14 Larvae take oxygen from submerged plants Larvae overwinter Permanent waters with rich vegetation as Preferred habitat An. plumbeus Cq. richiardii species present at a site were sorted according to these functional groups for characterization and comparison of mosquito faunas. The groups present at each study area were ranked, with one being the most abundant. For a few species, information on certain characteristics is scarce or there is more than one strategy. Of the species in group 5, Aedes rusticus (Rossi) can switch from overwintering in the larval stage to overwintering in the egg stage, in response to climatic conditions. Of the species in group 6, life-history data on Culiseta fumipennis (Stephens) and Culiseta ochroptera Peus are rather scarce. It is assumed that these species feed preferably on birds. Besides, Cs. ochroptera is reported to overwinter as females in the eastern Ukraine (Gutsevich et al. 1974), and probably also in Sweden (Jaenson et al. 1984). In group 7, Aedes

July 2001 SCHÄFER AND LUNDSTRÖM: MOSQUITO FAUNAS OF FORESTED WETLANDS 579 cantans Meigen, Aedes communis (DeGeer) and Aedes punctor (Kirby), can sometimes have a small second generation after summer rains (Mohrig 1969, Gutsevich et al. 1974). Preferred habitat for all species means that they mainly occur there, but could also be found occasionally in other habitats. Some Anopheles and Culex species cannot be distinguished morphologically as females. In group 1, Anopheles maculipennis sensu lato is a complex of four species in Sweden (Jaenson et al. 1986a) that cannot be safely identiþed morphologically, but all belong to the same functional group. Culex pipiens/torrentium is a complex as well, with Cx. torrentium Martini and Cx. pipiens pipiens L. as bird-feeding species (group 2), and Cx. pipiens molestus Forskål as an anthropogenous species (group 1). Cx. pipiens and Cx. torrentium cannot be distinguished morphologically as adult females and larvae, whereas male hypopygia are different (Service 1968). Cx. pipiens pipiens and Cx. pipiens molestus can only be separated by testing autogeny and stenogamy. In Sweden, there is only a single record of Cx. pipiens molestus (Natvig 1948), whereas Cx. pipiens pipiens and Cx. torrentium are reported regularly (Dahl 1977). Therefore, unspeciþed females of Cx. pipiens/torrentium were sorted in group 2. In addition to the analysis based on functional groups, we also used several diversity measurements. Alpha diversity was measured using three indices: Shannon index (HÕ, using natural log), placing most weight on rare species; SimpsonÕs index (1-D), placing most weight on common species; and Q-statistic, with no weighting toward very abundant or very rare species (Magurran 1988, Krebs 1999). Beta diversity was measured with the Sørenson index, based on species presence/absence data. The rarefaction method was used to estimate the number of species for the maximum number of individuals common to all study sites. Diversity measurements and rarefaction were computed with the aid of BIODIV 5.1 (Baev and Penev 1995) and Ecological Methodology 5.1 (Krebs 1998). Results A total of 52,298 individual mosquitoes of 32 species in Þve genera was included in our analysis (Appendix 1). At Abisko we found 10 mosquito species, at Allavaara 12, at Bergsåker 16, and at Egeside Sjö 24 species. Sorting the mosquito species according to functional groups presented in Table 1 showed that three groups were represented at Abisko and Allavaara, eight groups at Bergsåker, and 13 groups at Egeside Sjö (Table 2). There were more groups from north to south, showing a nested pattern with all functional groups present at one site also being present at sites located further south. Groups 1, 7, and 8 were present at all study sites. Three species out of these groups, Ae. communis, Ae. punctor, and Aedes excrucians (Walker) (groups 7 and 8 only), were found in all four areas. At Abisko and Allavaara the same functional groups ranked in the same order were present (Table 2), representing species that take oxygen from the water Table 2. Number of mosquitoes collected per functional group for each of the four study areas in Sweden Functional group Abisko Allavaara Bergsåker Egeside Sjö 1 3 (3) 23 (3) 85 (6) 8 (9) 2 0 0 299 (4) 160 (5) 3 0 0 0 0 4 0 0 0 907 (4) 5 0 0 0 2 (11) 6 0 0 5 (8) 118 (6) 7 14,351 (1) 9,382 (1) 1,177 (2) 7,421 (1) 8 602 (2) 107 (2) 468 (3) 2,324 (2) 9 0 0 13,476 (1) 1,096 (3) 10 0 0 145 (5) 1 (12) 11 0 0 0 3 (10) 12 0 0 0 106 (7) 13 0 0 0 1 (12) 14 0 0 6 (7) 22 (8) 3 3 8 13 Total no. of groups present Total number of collected mosquitoes (abundance ranking with 1 being the most abundant group). surface, lay their eggs on soil, overwinter in the egg stage and prefer forested (most abundant group) and partly forested habitats (second most abundant group), as well as species that lay their eggs on the water surface and overwinter as females. All species in groups 1, 7, and 8 prefer mammals as blood-meal. In these northern localities no species with overwintering larvae were found. At Bergsåker, Þve functional groups were added, including species with overwintering larvae, and feeding on birds. Groups 9 and 10 prefer habitats at the forest edge and in the open. Group 14 has the single Coquillettidia species in Sweden. The larvae and pupae of this species get oxygen by piercing submerged plant stems, mostly Phragmites. Most abundant at this location was a single species (Aedes cinereus Meigen), of functional group 9. At Egeside Sjö, 13 of 14 groups were present, missing only group 3, which contains a single species preferring amphibians as blood meal (Culex territans Howard, Dyar & Knab). In this area, more groups with overwintering larvae were added, as well as species typical of inundation areas, and a species using tree holes as larval habitats. All alpha diversity measurements showed highest diversity for the Egeside Sjö collections (Table 3). The Q-statistic displayed an increasing diversity from north to south, but the Shannon and Simpson indices both revealed a deviation from this trend at the Berg- Table 3. Alpha diversity measurements for mosquitoes collected in four Swedish forested wetlands Abisko Allavaara Bergsåker Egeside Sjö Shannon index (H ) 1.14 1.12 0.633 1.82 SimpsonÕs index (1-D) 0.52 0.589 0.254 0.783 Q-statistics 0.955 1.03 1.60 1.80 No. of species 10.0 12 16 24 Rarefaction for 9.5 12 14.6 22.8 n 9,500 Variance 0.347 0.00378 0.908 0.949

580 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 94, no. 4 såker study area, having their lowest value there. This reßected the strong prevalence of a single species at this site. Both the crude number of species and the rarefaction for n 9500 individuals demonstrated a southward increase of mosquito diversity, like the Q- statistic. Comparison of the mosquito faunas of the study areas with the Sørenson index showed most similarity between the faunas of Abisko and Allavaara (0.82). All sites had highest values with their neighboring sites (Allavaara-Bergsåker: 0.50; Bergsåker-Egeside Sjö: 0.55), and values between non-neighboring sites varied between 0.31 and 0.39. Discussion The comparison of mosquito faunas of four forested wetlands in Sweden showed distinct differences among northern, central, and southern study areas. The number of species as well as the number of functional groups increased from north to south. Most prevalent at all four study sites were species taking oxygen from the water surface, laying their eggs on soil, overwintering in the egg stage, preferring forested or partly forested habitats, and using mammals as blood-meal (groups 7 and 8). The second most successful functional group consisted of species laying their eggs on water surface, overwintering as females, and preferring mammals as host for blood meals (group 1). However, the high number of individuals from these groups could also simply reßect high probability of being caught with the traps we used. Jaenson et al. (1986b) sampled mosquitoes in the Sässman area east of Edsbyn (61 23 N, 15 51 E) during 3 yr, with a variety of collection methods including suction traps, animal-baited traps, human bait, and resting stations. The pooled results showed that mosquito species of groups 7Ð9 and 1 were most prevalent. This is in accordance with our results, so that we exclude a major inßuence of collection method. However, Cx. territans, preferring amphibians as bloodmeal, was found in the Sässman collections, but not in our samples. Therefore, this species was either missing from our study areas, or less attracted to CDC light traps with dry ice. The two northern localities Abisko and Allavaara lacked species with larvae as the overwintering stage, species feeding on birds, as well as most of the species preferring open habitats. The long and cold winters may not allow larval overwintering, and strong winds in open areas might prevent establishment of most open habitat mosquito species. Just two species, Aedes impiger Walker and Aedes nigripes Zetterstedt, are typical for arctic localities including tundra (Vockeroth 1954, Gutsevich et al. 1974, Wood et al. 1979). Absence of bird-feeding species could be explained by the fact that most of these species also overwinter in the larval stage (group 6). However, one species feeding on birds and overwintering as females, Cx. torrentium (group 2), occurs in west Siberia (Gutsevich et al. 1974), but Cx. pipiens/torrentium has not been recorded from localities in Scandinavia north of the Arctic Circle (Dahl 1977, Utrio 1979, Mehl et al. 1983). Previous records of geographic distribution of mosquitoes (Mehl et al. 1983, Andersson 1991) would add four species to our list for northernmost Sweden, and one functional group (group 9, Ae. cinereus). At Bergsåker, a single species (Ae. cinereus) was particularly prevalent. This species differs from others in similar functional groups 7 and 8, in the number of generations per year. Ae. cinereus can have several generations per year, and this could explain the extremely high numbers recorded at Bergsåker. A second ßood during the summer could lead to additional hatching of this species, but not to hatching of the species of groups 7 and 8. Species typical for river inundation areas like Aedes vexans Meigen, Aedes sticticus Meigen and Aedes rossicus Dolbeshkin, Gorickaja & Mitrofanova (group 12) were missing. Ae. vexans extends its range up to 60Ð62 N in Europe (Gutsevich et al. 1974), so it could well occur in the geographic area of Bergsåker. All three species were recorded from the Dalälven area at a latitude of 60Ð61 N (Jaenson 1986). Absence of group 12 mosquito species might be due to rare events of summer ßooding. These species would not hatch during spring ßoods, whereas Ae. cinereus is able to hatch both during spring and summer ßooding. Therefore, this species can take advantage of all opportunities and become dominant, especially if there is an occasional summer ßood. Three more species were recorded from the nearby Sässman area by Jaenson et al. (1986b), adding one more functional group (group 3). Most species and functional groups were recorded from Egeside Sjö in southern Sweden. This site has the most variable environment of the four sites studied, with mixed forests containing large decidious trees, and extensive meadows of Salix bush and sedges and reed beds. This was reßected in the presence of many different functional groups, including those breeding in tree holes (group 13), and typical for inundation areas (group 12). Six more species have been recorded for southern Sweden (Dahl 1977), which add the only functional group missing in our study, representing the species feeding on amphibians. The diversity measurements showed, on the one hand, an increasing diversity from north to south (species numbers, rarefaction, Q-statistic), but a deviation from this trend for the data from the Bergsåker area (Shannon and SimpsonÕs indices). Because the Shannon and the SimpsonÕs indices are inßuenced by either very rare or very common species, respectively, they are lower for the Bergsåker mosquito fauna due to the prevalence of a single species there. This shows that a single index is not sufþcient to describe the diversity of a site (May 1975). Comparing presence/absence data with the Sørenson index showed highest similarity between mosquito faunas at sites located next to each other on the north-south gradient. This shift in species presence is in accordance with the observed presence of functional groups at the different study areas. For most organisms, a gradient of increasing diversity with decreasing latitude is a well-known pattern

July 2001 SCHÄFER AND LUNDSTRÖM: MOSQUITO FAUNAS OF FORESTED WETLANDS 581 (Taylor 1978, Begon et al. 1986, Cushman et al. 1993, Väisänen and Heliövaara 1994, Gaston and Williams 1996), although a few taxa show the opposite pattern (Heie 1994, Kouki et al. 1994). Our data support the existence of a latitudinal gradient of mosquito diversity in Sweden, both with regard to species richness and functional groups. Sorting mosquito species into functional groups also provided explanations for this gradient, as the biology of some species does not Þt climatic conditions at northern localities. At more southern regions, environmental variability increased, providing a large variety of habitats so that the number of species and functional groups present increased as well. Sorting of mosquito species into functional groups made it possible to reveal characteristic biological features of the mosquito faunas, linked with study site environmental features. Mosquitoes form an obvious faunistic component of wetlands and humid forested areas, and our study provides a basis for further characterization of wetlands by their mosquito fauna. Acknowledgments We thank Christine Dahl, Norbert Becker, Anders Folke, Lars Wikars, Stefan Ås and Minoo B. Madon for practical support. This study was Þnanced by the Swedish Foundation for Strategic Environmental Research (MISTRA) through the Swedish Water Management Research Program (VAS- TRA), Stiftelsen Olle Engkvist Byggmästare, Stockholm, Sweden (J.O.L), KABS/GFS, Ludwigshafen, Germany (M.S.), and by scholarships from the Abisko ScientiÞc Research Station, Abisko, Sweden (M.S.). References Cited Ahti, T., L. Hämet-Ahti, and J. Jalas. 1968. Vegetation zones and their sections in northwestern Europe. Ann. Bot. Fenn. 5: 169Ð211. Andersson, I. H. 1991. Nectar feeding behaviour and the signiþcance of sugar meals in mosquitoes (Diptera: Culicidae). Ph.D. dissertation, Uppsala University, Uppsala, Sweden. Baev, P. V., and L. D. Penev. 1995. BIODIV. Exeter Software, New York. Begon, M., J. L. Harper, and C. R. Townsend. 1986. Ecology. Individuals, populations and communities. Blackwell, Oxford, UK. Cushman, J. H., J. H. Lawton, and B.F.J. Manly. 1993. Latitudinal patterns in European ant assemblages: variation in species richness and body size. Oecologia 95: 30Ð37. Dahl, C. 1977. Taxonomy and geographic distribution of Swedish Culicidae (Diptera, Nematocera). Entomol. Scand. 8: 59Ð69. Dahl, C. 1997. Diptera Culicidae, mosquitoes, pp. 163Ð186. In A. N. Nilsson [ed.], Aquatic insects of North EuropeÑa taxonomic handbook. Apollo Books, Stenstrup, Denmark. Francy, D. B., T.G.T. Jaenson, J. O. Lundström, E.-B. Schildt, Å. Espmark, B. Henriksson, and B. Niklasson. 1989. Ecological studies of mosquitoes and birds as of Ockelbo virus in Sweden and isolation of Inkoo and Batai viruses from mosquitoes. Am. J. Trop. Med. Hyg. 41: 355Ð363. Gaston, K. J. 1996. Species richness: measure and measurement, pp. 77Ð113. In K. J. Gaston [ed.], Biodiversity. A biology of numbers and differences. Blackwell, Oxford, UK. Gaston, K. J., and P. H. Williams. 1996. Spatial patterns in taxonomic diversity, pp. 202Ð229. In K. J. Gaston (ed.), Biodiversity. A biology of numbers and difference. Blackwell, Oxford, UK. Gutsevich, A. V., A. S. Monchadskii, and A. A. Shtakel berg. 1974. Diptera. Mosquitoes, Family Culicidae. Keter, Jerusalem, Israel. Heie, O. E. 1994. Why are there so few aphid species in the temperate areas of the southern hemisphere? Eur. J. Entomol 91: 127Ð133. Jaenson, T.G.T. 1986. Massförekomst av Aedes rossicus och andra stickmyggor vid Dalälven hösten 1985. Entomol. Tidskr. 107: 51Ð52. Jaenson, T.G.T. 1988. Diel activity patterns of blood-seeking anthropophilic mosquitoes in central Sweden. Med Vet. Entomol. 2: 177Ð187. Jaenson, T.G.T., and B. Niklasson. 1986. Feeding patterns of mosquitoes (Diptera: Culicidae) in relation to the transmission of Ockelbo disease in Sweden. Bull. Entomol. Res. 76: 375Ð383. Jaenson, T.G.T., B. Henriksson, and Å. Espmark. 1984. Ockelbosjukans ekologi: Culiseta ochropterañen ornitoþl stickmygga funnen i Sverige. Entomol. Tidskr. 105: 70Ð74. Jaenson, T.G.T., J. Lokki, and A. Saura. 1986a. Anopheles (Diptera: Culicidae) and malaria in northern Europe, with special reference to Sweden. J. Med. Entomol. 23: 68Ð75. Jaenson, T.G.T., B. Niklasson, and B. Henriksson. 1986b. Seasonal activity of mosquitoes in an Ockelbo disease endemic area in central Sweden. J. Am. Mosq. Control Assoc. 2: 18Ð28. Kettle, D. S. 1995. Medical and veterinary entomology. University Press, Cambridge, UK. Kouki, J., P. Niemelä, and M. Viitasaari. 1994. Reversed latitudinal gradient in species richness of sawßies (Hymenoptera, Symphyta). Ann. Zool. Fennici 31: 83Ð88. Krebs, C. J. 1998. Ecological methodology. Exeter Software, New York. Krebs, C. J. 1999. Ecological methodology. Benjamin/Cummings, Menlo Park, CA. Laird, M. 1988. The natural history of larval mosquito habitats. Academic, London, UK. Lundström, J. O., J. Chirico, A. Folke, and C. Dahl. 1996. Vertical distribution of adult mosquitoes (Diptera: Culicidae) in southern and central Sweden. J. Vector Ecol. 21: 159Ð166. Magurran, A. E. 1988. Ecological diversity and its measurement. University Press, Cambridge, UK. May, R. M. 1975. Patterns of species abundance and diversity, pp. 81Ð120. In M. L. Cody and J. M. Diamond (eds.), Ecology and evolution of communities. Belknap Press of Harvard University, Cambridge, MA. Mehl, R., T. Traavik, and R. Wiger. 1983. The composition of the mosquito fauna in selected biotopes for arbovirus studies in Norway. Fauna Norv. Ser. 30: 14Ð24. Mohrig, W. 1969. Die Culiciden Deutschlands. Untersuchungen zur Taxonomie, Biologie und Ökologie der einheimischen Stechmücken. Fischer, Jena, Germany. Natvig, L. R. 1948. Contributions to the knowledge of the Danish and Fennoscandian mosquitoes, Culicini. Norsk. Entomol. Tidskr. Suppl. I: 1Ð567. New, T. R. 1999. Limits to species focusing in insect conservation. Ann. Entomol. Soc. Am. 92: 853Ð860.

582 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 94, no. 4 Service, M. W. 1968. The taxonomy and biology of two sympatric sibling species of Culex, C. pipiens and C. torrentium (Diptera: Culicidae). J. Zool. Lond. 156: 313Ð323. Service, M. W. 1971. Feeding behaviour and host preferences of British mosquitoes. Bull. Entomol. Res. 60: 653Ð 661. Service, M. W. 1993. Mosquito ecology: Þeld sampling methods. Halsted, New York. Sjöberg, K., and L. Ericson. 1997. Mosaic boreal landscapes with open and forested wetlands. Ecol. Bull. 46: 48Ð60. Taylor, L. R. 1978. Bates, Williams, HutchinsonÑa variety of diversities, pp. 1Ð18. In L. A. Mound and N. Waloff (eds.), Diversity of insect faunas. Blackwell, Oxford, UK. Tempelis, C. H. 1975. Host-feeding patterns of mosquitoes, with a review of advances in analysis of blood meals by serology. J. Med. Entomol. 11: 635Ð653. Utrio, P. 1979. Geographic distribution of mosquitoes (Diptera, Culicidae) in eastern Fennoscandia. Not. Entomol. 59: 105Ð123. Väisänen, R., and K. Heliövaara. 1994. Hot-spots of insect diversity in northern Europe. Ann. Zool. Fenn. 31: 71Ð81. Vockeroth, J. R. 1954. Notes on the identities and distributions of Aedes species of Northern Canada, with a key to the females (Diptera: Culicidae). Can. Entomol. 86: 241Ð 255. Wood, D. M., P. T. Dang, and R. A. Ellis. 1979. The mosquitoes of Canada. Diptera: Culicidae. Canada Department of Agriculture, Ottawa, Canada. Received for publication 18 August 2000; accepted 21 February 2001. Appendix 1. Mosquito species and number of individuals identified from four study sites in Sweden Species Study site Abisko Allavaara Bergsåker Egeside Sjö Total Aedes annulipes (Meigen) 0 0 0 2,306 2,306 Aedes cantans (Meigen) 0 0 150 4,606 4,756 Aedes caspius (Pallas) 0 0 0 2 2 Aedes cataphylla Dyar 0 0 20 142 162 Aedes cinereus Meigen 0 0 13,476 1,096 14,572 Aedes communis (DeGeer) 10,092 2,870 957 1,405 15,324 Aedes diantaeus Howard 0 1 18 0 19 Dyar & Knab Aedes dorsalis (Meigen) 0 0 0 1 1 Aedes excrucians (Walker) 602 107 468 18 1,195 Aedes flavescens (Mueller) 0 0 144 0 144 Aedes geniculatus (Oliver) 0 0 0 1 1 Aedes hexodontus Dyar 1,020 893 0 0 1,913 Aedes impiger (Walker) 55 8 0 0 63 Aedes intrudens Dyar 0 217 1 0 218 Aedes leucomelas (Meigen) 0 0 1 1 2 Aedes nigripes (Zetterstedt) 8 0 0 0 8 Aedes pionips Dyar 411 65 0 5 481 Aedes pullatus (Coquillett) 1,499 29 0 12 1,540 Aedes punctor (Kirby) 1,266 5,299 31 1,251 7,847 Aedes rusticus (Rossi) 0 0 0 2 2 Aedes sticticus (Meigen) 0 0 0 13 13 Aedes vexans (Meigen) 0 0 0 93 93 Subtotal Aedes 14,953 9,489 15,266 10,954 50,662 Anopheles claviger (Meigen) 0 0 0 907 907 Anopheles maculipennis s.l. 0 1 84 1 86 Subtotal Anopheles 0 1 84 908 993 Culiseta alaskaensis (Ludlow) 1 1 0 1 3 Culiseta annulata (Schrank) 0 0 0 6 6 Culiseta bergrothi (Edwards) 2 21 1 0 24 Culiseta fumipennis (Stephens) 0 0 0 5 5 Culiseta morsitans (Theobald) 0 0 2 113 115 Culiseta ochroptera Peus 0 0 3 0 3 Subtotal Culiseta 3 22 6 125 156 Culex pipiens/torrentium 0 0 299 160 459 Coquillettidia richiardii (Ficalbi) 0 0 6 22 28 Total 14,956 9,512 15,661 12,169 52,298