Evolution has shown living things a thousand ways to save themselves.
The leopard gecko's tail pulls off, leaving the cat clutching nature's version of the tear-away jersey. The female pea crab Pinnotheres, unsatisfied with her own shell, spends its life inside an oyster. The bacterium Thermotoga maritima grows in water just below boiling temperature, an environmental niche into which most organisms won't dip a toe.
Few strategies, however, are as strange and unlikely as the one periodical cicadas found.
These large, ungainly insects in the genus Magicicada spend either 13 or 17 years underground, then emerge nearly simultaneously in densities that can exceed 1 million per acre. Their few weeks of life in the open air are spent molting, calling for a mate (in the case of the buzzing males), copulating and depositing eggs in nests made in gashed twigs (in the case of the diligent females).
They do little to defend themselves. They fly poorly, don't fight and taste great. In the parlance of animal behavior, cicadas are "predator foolhardy" -- they are always available for lunch. Birds consume them in the greatest numbers, but many other animals get in on the act. Squirrels, dogs, cats, turtles, fish and spiders all eat cicadas, which for a few weeks are the protein equivalent of manna from heaven.
Eventually, though, everyone gets full -- and there are still billions of cicadas alive. This is the survival strategy known as "predator satiation." It is a passive strategy that depends almost entirely on timing. If too few cicadas emerge, or if they come out over an extended period of time, they are likely to be wiped out by predators. If this occurs before they find a mate and create a new generation to carry their genes forward, they will eventually disappear completely.
Most species of cicada have life cycles between two and eight years, with a fair amount of variability. If five years is the dominant length, for example, many members of a population may come out in four years -- or not until the sixth.
Of course, this isn't apparent to the casual observer. That's because in most places a person hears cicadas -- often more than one species -- every summer. A fraction of the population reaches maturity and emerges every year, making the insects annual, not periodical.
Unlike Magicicada, these other species survive by strategies more clever than simply waiting for predators to get sated. For example, the large dog-day cicadas of the Deep South "are very fast, powerful flyers -- that's how they get away from the birds," said David Marshall, an evolutionary biologist at the University of Connecticut. On the other hand, many small species survive by being so well-camouflaged that "you can't see them even when they're a foot from your head," he said.
The periodical cicadas about to emerge here once shared a common ancestor with these insects. But a lot has happened since then. They have lengthened their life cycles and evolved into geographically distinct "broods" in which all members are on the same developmental schedule. They have also settled on specific life-cycle lengths, either 13 or 17 years. Both of those are large prime numbers, which means they can be divided only by themselves and 1.
How and why did this happen?
As with many questions about natural selection, nobody can say for certain. The crucial events lie in the deep past; we have only the finished product. However, knowledge of the conditions in which the traits evolved, and logical reasoning, suggest a scenario.
Biologists believe that periodical cicadas evolved during the Pleistocene Epoch, which began about 1.8 million years ago. It was a time when glaciers repeatedly advanced and retreated, and the climate of eastern North America was alternately -- and somewhat unpredictably -- warm and cool.
Members of the genus Magicicada -- the periodical cicadas -- require prolonged temperatures above 68 degrees Fahrenheit to fly, copulate and lay eggs. The region where they evolved -- the southern edge of glaciation -- had many summers that simply weren't that warm. Cicadas that emerged in those summers would have died before they could produce offspring.
It turns out that for reasons of mathematical probability, a good strategy for avoiding randomly cold summers is to stay underground for as long as possible. The less often a brood comes up, the less often it encounters a killing summer.
The ancestors of periodical cicadas didn't somehow choose to stay underground for longer periods. However, in those ancient cicada populations were individuals that because of existing genes or new mutations were destined to develop more slowly and emerge a year or two later than their brethren. In an era of sporadically cold summers, those insects were more likely to survive -- and pass on their slow-development genes to their offspring.
Randel T. Cox of the University of Memphis and C.E. Carlton of the University of Arkansas calculated the odds of survival of populations of cicadas of different life-cycle lengths over a 1,500-year period in which 1 of every 50 summers was fatally cold. Cicadas with six-year life cycles had a 4 percent chance of surviving. Those with an 11-year cycle had a 51 percent chance. Those with a 17-year cycle had a 96 percent chance.
Each time the glaciers arrived, the ice wiped out the cicadas in the northernmost regions of eastern North America, where Magicicada was evolving. But the populations south of each "glacial maximum" would have survived and kept evolving. Over millions of years of advancing and retreating ice, genes leading to long life cycles would have been favored. They would have been "enriched" in the gene pool until ultimately they became the norm.
But why 13 or 17 years? Life cycles that long are mathematically more likely to have survived the Pleistocene era than shorter ones, but that doesn't explain the benefit of a prime number.
It turns out that if an area contains populations of cicadas with different life-cycle lengths, broods with long cycles that are high prime numbers will share summers less often with other broods. But why is that an advantage?
When broods with different cycle lengths emerge at the same time, some members will interbreed. Their offspring will be hybrids, carrying a mixture of genes. If the precise timing of a cicada's life cycle is produced by the interaction of several genes, then getting those genes from two different populations might change the interaction. That might affect the length of the life cycle.
If the offspring of 11-year and six-year cicadas are cicadas of a third cycle length -- say, nine years -- then the number of insects emerging on either parent's schedule in the next generation will decrease. That, in turn, will diminish the strength-in-numbers survival strategy on which all Magicicada species depend.
Cicadas with 13- and 17-year life cycles emerge less often with other broods than do the ones whose cycles are non-prime numbers. For example, when a brood with a 14-year life cycle emerges, it will share the summer with two-year and seven-year broods, with which it will interbreed and produce mongrel offspring. A 16-year brood will share a mating season with two-, four- and eight-year broods, and an even more diverse group of hybrids will result.
In contrast, when 13- and 17-year broods are out, they share the season only with broods having short life cycles (such as one, two or three years) -- and life cycles that short presumably couldn't survive the Pleistocene climate. The net result was that 13- or 17-year cicadas didn't have their genes "diluted" by hybridization -- except every 221 years, when they were out together in the few places where they shared the same turf.
Of course, periodical species didn't pick 13 and 17 as their magic numbers. As with all evolutionary processes, choice played no part.
What happened, instead, was that broods of other cycle lengths simply became extinct. They emerged in a cold summer and failed to reproduce, or they emerged in insufficient numbers and were eliminated by predators. What remained were the broods mathematically most likely to make it across the Pleistocene minefield -- 13 and 17. Furthermore, in those populations synchronicity was essential. Individuals whose timing was off just a little -- ones that emerged a year early or late -- were extirpated. But their brethren whose genes endowed them with perfect timing generation after generation, survived.
Various versions of this scheme of evolution have been proposed by several researchers, including Cox, Carlton and a Japanese researcher named Jin Yoshimura. But did it actually happen?
There is indirect evidence it did in today's geographic distribution of the two populations of periodical cicadas.
The 17-year broods today lie north of the 13-year broods. The border between them follows a well-known S-shaped line from the Atlantic Ocean to the Great Plains. It sweeps south of most of the Appalachian Mountains and then turns north as far as southern Illinois before turning again and passing south of the Ozarks. It divides eastern North America into two zones -- one with long, harsh winters to the north, and the other with mild, shorter winters to the south.
This line marks the northern and southern ranges of various plant and animal species, or the border between subspecies with different markings, size and behavior. The life-cycle lengths of cicadas follow this same climate contour -- and strongly suggest that temperature played a crucial role in determining the life-cycle length of the periodical cicadas.
Still, Earth has other places whose ecological history is much like that of eastern North America. Why did periodical cicadas evolve only here?
"This," said Randel Cox, "is a vexing question."
"I don't have an answer to that," said David Marshall.