Evolution of prolonged development: a life table analysis for periodical cicadas.

According to conventional wisdom, natural selection should favor early reproduction. Prolonged development of 13 and 17 yr has been difficult to explain for periodical cicadas. Earlier, I hypothesized that development may be long for periodical cicadas because fecundity increases as a function of longer development with little increased risk of mortality. In this article, I tested whether the fecundity was greater for 17-yr cicadas than for 13-yr cicadas and estimated the shape of the survivorship curve. A cohort of 17-yr cicadas was followed from 1979 through 1996. Most mortality occurred during the first 2 yr; thereafter mortality was uncommon. Across 17 yr, adult densities increased at five out of seven sites, and in no case did the change exceed three times. Seventeen-year adults had heavier ovaries than did 13-yr adults; this effect was greater for Magicicada cassini (1.80 times) than for Magicicada decim (1.16 times). For M. cassini, the extra fecundity associated with 17-yr development swamped the potential advantages of more frequent reproduction of 13-yr forms under most conditions. For M. decim, realistically low rates of mortality roughly matched the small effect of development on fecundity. Prolonged development in periodical cicadas may be explained adequately by the demographic hypothesis, although it may also reduce predation risk to adults or result from strong selection for large body size.