Revisiting the evolution of population stability due to selection for rapid development and early reproduction in Drosophila: the role of generation length

Abstract

The ubiquity of stable populations in nature generated considerable interest in how population stability might evolve, especially after the realization that higher per capita population growth rates typically yield unstable dynamics. Yet, most empirical and theoretical treatments implicitly assume that population dynamics and stability are invariant to generation length, an assumption that remains largely untested. Theory suggested that population stability could evolve as a correlated response to life-history evolution, which was first experimentally demonstrated using D. melanogaster populations selected for rapid development and early reproduction (FEJs). Constancy stability of FEJs evolved to be higher than their ancestral controls (JBs), likely due to correlated reductions in fecundity and pre-adult survivorship. However, that study assessed stability on a 21-day generation length (matching the JBs), resulting in a considerable mismatch with the generation length of FEJs (10 days). This raises a broader question of whether the observed differences in stability reflect evolved life-history changes or are an artifact of the generation length at which the populations were assayed. To address this, we assessed the stability of FEJs, JBs, and relaxed-selection populations derived from the FEJs (CRFs and FRFs) across two generation lengths (12 and 18 days), tracking 320 small populations for 27 generations. Contrary to the earlier findings based on a 21-day generation length study, FEJs did not differ in constancy from JBs when assayed at a shorter generation length. Thus, even a modest difference in life-cycle length can significantly influence population stability, thereby underscoring the need to account for generation length when comparing stability across populations. Interestingly, constancy and persistence stability of selection regimes evolved in opposite directions, highlighting the need to assess stability along multiple axes. We discuss these results using an empirical framework, emphasizing its utility over simple population growth models for a richer understanding of population dynamics and stability.