Numerical Bimodality Among North American Gypsy Moth Populations

Recent studies led to the hypothesis that North American populations of the gypsy moth, Lymantria dispar (L.), are regulated by processes that tend to produce a numerically bimodal population system. We postulated that this system has 4 distinct features—2 major and relatively stable modes (outbreak and innocuous) and 2 transient phases (release and decline). First, the system can apparently be contained indefinitely within a sparse density range, though the processes that result in numerical stability at innocuous levels tend to be least effective near the advancing front of the generally infested area. Second, processes must exist through which sparse populations can expand numerically to outbreak levels. These processes are found most commonly within outbreak foci. Third, there are processes that can maintain the outbreak for several years (up to a decade), on an area-wide basis, as a quasi-stable feature. Such processes tend to be most effective were a mixture of dense and sparse subpopulations is represented. Fourth, there are processes that ultimately cause outbreaks to collapse; these include disease, starvation, and reduced fecundity. Our results support this hypothesis; we conclude that this population system is now, and has been, numerically bimodal in North America.