The article briefly touches on why not 7 or (almost) 11: "If we discount those cicadas with life cycles of ten years or fewer (as being too close to predator l

The article briefly touches on why not 7 or (almost) 11: "If we discount those cicadas with life cycles of ten years or fewer (as being too close to predator life cycles), we find that the most successful emergence rates for cyber cicadas are thirteen and seventeen years—precisely what we find in the wild." Why not beyond 17? The article does not say. The next prime would be 19 (only two years longer). Perhaps 17 provides a sufficiently large local optimum that there is little pressure to lengthen the cycle to 19 or beyond.