Control of alternate-year flight activities in high-alpine Ringlet butterflies (Erebia, Satyridae) and Burnet moths (Zygaena, Zygaenidae) from temperate environments

Abstract

Fourteen of 27 species of Ringlet butterflies of the high-alpine zone have a two-year developmental cycle with obligatory dormancy in young larval instars (L₁, L₂) in the first winter and in the penultimate instar (L₄) in the second winter. Long-term collection records show that all semivoltine species fly mainly in odd years. Oscillations in population density have been synchronous since 1918. They cannot be explained exclusively by the semivoltine cycle. The few specimens in the intervening years might originate from parallel broods. Samples of field-collected young larvae from semivoltine species were attacked by parasitoids (Ichneumonidae, Braconidae) mainly in odd years. It is hypothesized that parasitoids emerging from the larger population of alternate years control the smaller populations of the following years. Once established in an abundant semivoltine species, these density-dependent biennial fluctuations of parasitoid frequencies would control the biennial reproductive success of all semivoltine Ringlet species that exhibit two obligatory, density-independent dormancies.

Burnet moths from temperate environments are abundant every year whatever the weather. Larvae can repeatedly enter a facultative diapause in different instars (L₃—L₁₀), characterized by moulting to a decolourized morph of smaller size. Consequently some individuals skip one or more reproduction periods, so that the generation time may vary from one to four years, and generations overlap. Diapause processes with different diapause stages not only are under the control of different temperature-photoperiod combinations, but also are influenced by the number of photoperiodic cycles experienced during the temperature-dependent period spent in each larval instar. In experiments with photoperiods below the critical daylength, the developmental rate of pre-diapause feeding instars increased and individual differences in developmental velocity were pronounced. In extreme short-day conditions larvae with slow larval development perceive more short-day cycles from the L₁ stage onwards and hence enter diapause in earlier stages. Larvae with more rapid development enter diapause later in older instars (L₅, L₆). Larvae with slower development are predisposed to enter diapause in early instars and to repeat diapause in successive years in later instars. Specimens following this developmental pathway may have allowed successful colonization of more northern habitats with unstable and unpredictable weather conditions, where cold summers prevent reproduction in some seasons. Data from local populations reveal clinal variation in these diapause reactions.

Occurrence of repeated diapause is an apomorphic trait in the genus Zygaena and may have assisted the adaptive radiation of this species-rich genus in the Palaearctic. In Erebia, repeated diapause seems to have evolved separately in closely related species groups. A problem still unsolved is how variable growth rates evolved within Zygaena populations and among Erebia species.