From late May through June, Brood X of the periodical cicadas will emerge from the ground, having spent the past 17 years as nymphs feeding off tree roots. After digging their way out and molting into adults, billions of the big, clumsy, red-eyed insects will sing their earsplitting love songs. Last seen in 1987, the brood will provide a prodigious if brief feast for birds, along with an incomparable opportunity for researchers. Fascinated naturalists have been writing about periodical cicadas for four centuries. But much remains unknown about the insects' periods or what triggers their synchronized appearances.

Brood X is perhaps the largest and best studied of the approximately 15 broods of periodical cicadas (researchers dispute the exact number). A brood emerges somewhere east of the Great Plains almost every spring. Worldwide, investigators have identified some 3,000 cicada species but know the life cycle for only a dozen or so. William Bradford, governor of the Plymouth Colony, first described periodical cicadas in 1633, although Native Americans probably knew of the creatures before then. The 17-year life cycle was firmly established less than a century later; by the mid-19th century, naturalists had recognized 13-year cicadas.

For more than 100 years, entranced mathematicians and biologists have tried to explain why periodical cicadas have evolved these prime-number cycles. One idea has been that the different cycles reduce competition for resources and interbreeding, because 13- and 17-year broods in the same locale will emerge together only once every 221 years. But in fact, different periodical cicada broods tend to be dispersed; little geographic overlap exists among most of them. And they do almost all their competitive eating during their long underground years, when they are sucking sap from tree roots.

Theorists have also argued that these oddball life cycles help cicadas to avoid predators and parasites with shorter, even-numbered life cycles. In 2001 researchers at the Max Planck Institute of Molecular Physiology in Dortmund, Germany, reported that prime-numbered life cycles emerged from their mathematical model of predator-prey relations.

Cicada researchers are deeply dubious about this explanation, however. The theory has not been falsified, notes evolutionary biologist Chris M. Simon of the University of Connecticut, because it cannot be tested. Her colleague David C. Marshall points out that true periodicity is rare in cicadas--separate groups of most species emerge every year. "If periodical cicadas evolved longer and longer life cycles to avoid a synchronizing parasitoid species," he notes, "then why has this apparently not happened in scores and scores of other cicada species that suffer predation and parasitism, not to mention in other kinds of insects and other animals?"

More curious to biologists such as Simon is the interaction among broods. As it does every spring, the University of Connecticut team will map cicada distributions, collect the insects for genetic analysis, and conduct small experiments on mating behavior. This year, Simon says, the researchers will scoop up samples from parts of Kentucky and Georgia where Brood X meets Broods XIX and XXIII of the 13-year cicadas and examine these specimens' DNA for evidence of past hybridization.

In addition, scientists are curious about developmental anomalies: broods sometimes drop or add a four-year stage called an instar. Entomologist Gene Kritsky of the College of Mount St. Joseph has reported accelerated development in Brood XIV, a 17-year cicada due out in 2008. He will be studying whether Brood XIV members come out this year, four years early, along with Brood X. In 2000 Kritsky also documented an early emergence of some of this year's Brood X cicadas. He hopes to be around to observe whether the eggs hatched in 2000 will stick to their new timetable and emerge in 2017--thus establishing a new brood--instead of reverting to the normal Brood X year, 2021.

Tabitha M. Powledge writes about biology and medicine from the greater Washington, D.C., area.