The Habitats of 17‐Year Periodical Cicadas (Homoptera: Cicadidae: Magicicada Spp.)

Abstract

Three species of 17—yr periodical cicadas–Magicicada septendecim (Linnaeus), M. cassini (Fisher), and M. septendecula Alexander and Moore–have ranges that are coextensive except around the edges. All the cicadas occurring together are the same age and emerge in the same 17th yr (hence they are "periodical"), yet populations in different parts of the range (called "broods") are out of phase with one another. Within the same brood, the three species are always perfectly synchronized, but they are separated microspatially by having different habitats within the same woodland. Magicicada septendecula prefers ovipositing in hickories and walnuts, and emerges in higher proportions under those trees than under comparable oaks. Both M. septendecim and M. septendecula occur together in upland woods, but septendecim exhibits much less host specificity than septendecula. The latter species is much rarer than septendecim; it can usually be heard chorusing in local patches within a woods occupied by septendecim. Magicicada cassini is a species of floodplain woods, and characteristically can be seen to replace septendecim and septendecula as one moves down a wooded slope leading to a stream. Over much of the eastern United States, however, the original forest has been extensively disturbed. Periodical cicadas survive and reproduce surprisingly well in cutover, scrubby second growth. Tree species characteristic of floodplains, like American elm, are often a component of upland second growth, and, especially in such situations, cassini, septendecim, and septendecula become intermixed though they remain reproductively isolated. The present lack of microspatial separation in many situations, then, is an artifact of human disturbance. Even so, on the edges of the range where only one species occurs, it does not invade the habitat of the other, even though its "competitor" is absent. These assertions are based on 4 yrs of summer field studies, from 1962—65 on Broods II, III, IV, and V primarily in Virginia, Iowa, Kansas, Oklahoma, and Ohio, and later, more cursory observations in 1968—70 on Broods VIII, IX, and X in Pennsylvania, West Virginia, and Indiana, as well as on previous studies by ourselves and others reported in the literature.