Evolutionarily stable seasonal timing of univoltine and bivoltine insects

Seasonality Maintains Alternative Life-History Phenotypes

“…The diapausing developmental stage did not affect the predicted qualitative differentiation between alternative life histories, but the magnitude of differentiation was smaller with egg diapause compared with pupal diapause, particularly in juvenile traits. Roff () predicted longer development time and larger body size under direct development than in the diapause pathway if egg is the overwintering developmental stage, whereas Iwasa et al () predicted the opposite body size difference between the developmental pathways when food quality peaks early in the season. We are unaware of any empirical data consistent with the models by Iwasa et al () and Roff (), but the qualitatively different predictions of their analyses, combined with the quantitative effect of the life cycle found in this methodologically different study, suggest that the type of life cycle may affect the evolution of alternative life histories.…”

“…Theoretical studies of insect life‐history evolution have generally assumed that individuals following the alternative developmental pathways have similar life histories. The few counter‐examples include models by Iwasa et al () and Roff (), predicting that development time may be longer in the direct development pathway than in the diapause pathway. The analysis by Abrams et al () and empirical evidence imply the opposite; directly developing individuals have shorter development times and higher growth rates than diapausing ones.…”

“…Production of an additional generation within a year is favored when the season becomes sufficiently long, although it inevitably results in a shorter time available for each generation (Roff ). Accordingly, populations along a gradient of changing season length are predicted to show phenological and life‐history adaptations to local conditions that results in complex saw‐tooth clines across the gradient (Roff ; Iwasa et al ; Kivelä et al ). The saw‐tooth pattern arises because the gradual change in life histories in relation to season length is interrupted by abrupt changes where there is a change in voltinism.…”

“…The saw‐tooth pattern arises because the gradual change in life histories in relation to season length is interrupted by abrupt changes where there is a change in voltinism. Development time and body size are predicted to increase with increasing season length, but sharp decreases are predicted for both traits when an additional generation emerges (Roff ; Iwasa et al ). Reproductive effort is predicted to decrease with increasing season length with a sharp increase at the transition from a univoltine (one generation per season) to a partially bivoltine phenology (a phenology where only the earliest offspring of overwintered individuals enter direct development; Kivelä et al ).…”

“…The discrepancies between theory and observations may arise because theoretical studies have focused on a single trait at a time (Roff ; Iwasa et al ; Gotthard et al ; Kivelä et al ). Although this approach may facilitate analytical solutions, it ignores the fundamental fact that the whole life history is under selection.…”