Outside Tapachula, Chiapas, Mexico—10 miles from Guatemala. To reach the cages, we follow the main highway out of town, driving past soy, cocoa, banana and lustrous dark-green mango plantations thriving in the rich volcanic soil. Past the tiny village of Rio Florido the road degenerates into an undulating dirt tract. We bump along on waves of baked mud until we reach a security checkpoint, guard at the ready. A sign posted on the barbed wire–enclosed compound pictures a mosquito flanked by a man and woman: Estos mosquitos genéticamente modificados requieren un manejo especial, it reads. We play by the rules.

Inside, cashew trees frame a cluster of gauzy mesh cages perched on a platform. The cages hold thousands of Aedes aegypti mosquitoes—the local species, smaller and quieter than the typical buzzing specimens found in the U.S. At 7 a.m., the scene looks ethereal: rays of sunlight filter through layers of mesh creating a glowing, yellow hue. Inside the cages, however, genetically modified mosquitoes are waging a death match against the locals, an attempted genocide-by-mating that has the potential to wipe out dengue fever, one of the world’s most troublesome, aggressive diseases.

Throughout a swath of subtropical and tropical countries, four closely related dengue viruses infect about 100 million people annually, causing a spectrum of illness—from flu-like aches to internal hemorrhaging, shock and death. No vaccine or cure exists. As with other mosquito-borne diseases, the primary public health strategy is to prevent people from being bitten. To that end, authorities attempt to rid neighborhoods of standing water where the insects breed, spray with insecticides, and distribute bed nets and other low-tech mosquito blockers. They pursue containment, not conquest.

Anthony James, however, is mounting an offensive. James, a molecular biologist at the University of California, Irvine, and his colleagues have added genes to A. aegypti that block the development of flight muscles in females. When a genetically modified male mosquito mates with a wild female, he passes his engineered genes to the offspring. The females—the biters—don’t survive long. When they emerge from the pupal stage, they sit motionless on the water. They won’t fly, mate or spread disease. The male progeny, in contrast, will live to spread their filicidal seed. In time, the absence of female offspring should lead to a population crash, which James’s collaborator has already demonstrated in the controlled environment of an indoor laboratory in Colorado. Now he has brought his bugs south.

The technology marks the first time scientists have genetically engineered an organism to specifically wipe out a native population to block disease transmission. If the modified mosquitoes triumph, then releasing them in dengue-endemic zones worldwide could prevent tens of millions of people from suffering. Yet opponents of the plan warn of unintended consequences—even if mosquitoes are the intended victims.

Aedes aegypti