Just after sunrise in early January, a delivery van trundled along a suburban street in Queensland, Australia. Inside were tubs filled with a type of mosquito that carries dengue fever, the flulike illness that annually sickens 50 million to 100 million people worldwide. Workers inside the van stopped at every fourth house, took out what resembled a small Chinese food container and released 40 mosquitoes into the wild. After a week, they had filled the air with 6,000 insects. By early March they had launched 72,000.

What may sound like bioterrorism is, in fact, a novel form of biological insect control. Scott O’Neill of the University of Queensland and his colleagues are testing a new method of reducing the spread of dengue, which is a growing scourge in the tropics and has recently shown up in the U.S. Although the disease is usually not fatal, dengue can land patients in the hospital, and it has no cure or vaccine.

O’Neill’s approach vaccinates mosquitoes instead of patients. In his lab, under a microscope, workers inject the bacterium Wolbachia pipientis, which is harmless to humans and common among insects, into eggs of Aedes aegypti mosquitoes, a major carrier of the dengue microbe. O’Neill has found that Wolbachia makes A. aegypti resistant to—and unable to transmit—the disease. What is more, all progeny inherit immunity.

O’Neill’s method, which entails no genetic modification, stands in contrast to dengue-control efforts that made headlines this past winter. In December, British biotechnology company Oxitec released 6,000 genetically modified male mosquitoes in Malaysia to the alarm of some groups that expressed concern about the possible effects of GM insects on humans and ecosystems. Results are not yet available for Malaysia, but Luke Alphey, chief scientist and founder of Oxitec, says an earlier release of 3.3 million of the mosquitoes on Grand Cayman Island resulted in an 80 percent reduction in A. aegypti, presumably because many females ended up mating with GM partners, which were infertile, instead of wild males.

O’Neill’s results are also promising. Early testing has shown that about 25 percent of larvae in the wild population were Wolbachia-infected and dengue-immune. By late May he hopes to have achieved his trial’s goal: to show that Wolbachia can improve resistance to dengue in a wild population of A. aegypti. If so, he will launch a similar trial in Vietnam in early summer.