Mosquitoes are among the most despised of insects, with remarkably few redeeming qualities. While they are certainly annoying intruders at backyard cookouts, mosquitoes are also responsible for at least 3 percent of all deaths worldwide, primarily through the transmission of malaria in the tropics. Americans face periodic (and less lethal) epidemics of such mosquito-borne pathogens as West Nile virus, St. Louis encephalitis virus and others.
So knowing that bats eat mosquitoes should make bat conservation a lot easier. It turns out, however, that it's not quite so simple.
The importance of bats as predators of mosquitoes and their potential to reduce mosquito-related disease have been suggested often in the past, but many scientists dismiss such claims. The skepticism stems mainly from studies of bat diets. While research shows that many bats will indeed eat mosquitoes, it also suggests that mosquitoes constitute only a small proportion of a bat's diet.
This debate over bats' impact on mosquito populations has been conducted for years with little empirical evidence on either side. Our experimental research now provides initial estimates of bats' potential impact on mosquito vectors of disease – and should, we hope, give new ammunition to those working to conserve the world's bats.
Female mosquitoes are the villains here, since only females spread disease by feeding on blood – and they are ideal disease vectors. Female mosquitoes emerge from an aquatic larval stage, mate and then feed on blood to support their egg production. This initial blood meal is generally their first opportunity to pick up a disease-causing pathogen. Next, they lay eggs in an appropriate aquatic habitat, then feed on blood again to develop another clutch of eggs. They repeat the blood meals and egg laying until their death after a few weeks. Pathogens are transmitted during the second and subsequent blood feedings, which suggests that attacking mosquito populations at the point of egg laying would be an efficient means of limiting disease transmission.
Previous studies of predation on adult mosquitoes have mostly been limited to "sit-and-wait predators," such as spiders, geckos and frogs. Active predators, such as bats, have not been well studied, despite the popular assumption that bats limit mosquito populations. A lack of systematic investigation means that we simply do not know if bats can actually fill this role.
We decided to explore this issue as graduate students at the University of Michigan, where one of us (Reiskind) studied mosquito ecology, while the other (Wund) investigated bat behavior. Our task required a balance between field-condition realism and experimental control. Our results provide the first measurement of how bats can, in fact, limit mosquito populations by preying on egg-laying females. While the question of whether bats are an important part of natural mosquito control remains open, we view our work as an important first step in determining how bats might limit the transmission of mosquito-borne diseases.
Our previous research had suggested that the mosquito Culex restuans, a carrier of the West Nile virus, could be readily attracted to a specific artificial aquatic habitat – a plastic tub filled with water and decaying hay – to lay their eggs. These mosquitoes always lay "clutches" of between 75 and 200 eggs glued into a raft that floats on the water surface. Our artificial habitats attracted as many as 520 egg clutches in a single evening. We tapped this behavior to bring large numbers of naturally occurring mosquitoes into contact with bats inside outdoor flight cages.
We trapped northern myotis (Myotis septentrionalis) from a colony in northern Michigan and built field enclosures using PVC pipe and plastic landscape netting with quarter-inch (6.35 mm) mesh. This allowed mosquitoes and other flying insects to enter the enclosures, but prevented bats from escaping. Each enclosure was approximately 105 cubic feet (3 cubic meters). At each of three sites in a nature preserve in southeastern Michigan, we placed two enclosures about 165 feet (50 meters) apart.
Over two months (June and July 2004), we randomly designated one enclosure at each site the "bat" enclosure and the other the "control" and placed one of the artificial egg-laying habitats in each one. A third, identical habitat was placed equidistant between the two enclosures. This unenclosed container allowed us to measure egg-laying rates at each site, independent of the enclosure treatments.
During June, we also placed artificial habitats directly outside the cages (as well as inside). If egg laying was reduced in these containers, it would suggest that mosquitoes were avoiding areas with bats, rather than just disappearing because they were being eaten.
Every morning for three weeks in June and three in July, we counted the number of egg clutches laid in the containers by naturally occurring female mosquitoes. We released bats into the bat enclosure on 18 separate nights.
We found that significantly fewer egg clutches were laid in enclosures with bats than in enclosures without them. There were, in fact, 32 percent fewer egg clutches in the bat enclosures than in controls. This reduction occurred only within flight cages and not in containers just outside the enclosures, confirming that female mosquitoes were present and that the bats were eating many of those that entered the cages. This observation also suggests that mosquitoes do not alter their egg-laying behavior in the presence of bats.
The number of egg clutches in the same enclosures was reduced by 31 percent on nights when bats were present compared to when they were not released into the flight cage. A final comparison found no difference in egg clutches between the control and bat enclosures on nights when bats were not released. These two comparisons demonstrate that the presence of bats is the only factor that explains the reduction of egg clutches.
We offer this first empirical study of the important question: Do bats help limit mosquitoes? We hope it will not be the last.
Our research design obviously was not a perfect reflection of reality. The use of enclosed bats presents an artificially high bat density, although we chose a species that forages intensively in dense habitats. Likewise, the use of a highly attractive habitat for egg-laying mosquitoes is probably somewhat outside the normal realm. What we have demonstrated, however, is that given a high density of bats and a high density of mosquitoes, the conventional wisdom may very well be accurate: bats can reduce mosquito populations.
We hope our results stimulate other researchers to creatively tackle the question of bats' role in controlling mosquitoes and reducing disease transmission by them. Just as importantly, we hope that our work gives new power to the argument that bats are essential allies in controlling pests and maintaining healthy ecosystems.
MICHAEL H. REISKIND is an Assistant Professor at the Department of Entomology and Plant Pathology at Oklahoma State University at Stillwater. His research focuses on mosquitoes as disease vectors. MATTHEW A. WUND is an evolutionary biologist and Assistant Professor at The College of New Jersey.
This research was originally published as "Experimental assessment of the impact of the northern long-eared bat on ovipositing Culex (Ditpera: Culicidae) mosquitoes" in the Journal of Medical Entomology, September 2009, 46(5): 1037-1044.