Variation in butterfly diapause duration in relation to voltinism suggests adaptation to autumn warmth, not winter cold

1. The life cycles of animals vary in relation to local climate, as a result of both direct environmental effects and population-level variation in plastic responses. Insects often respond to the approach of winter by entering diapause, a hormonally programmed resting state where development is suspended and metabolism suppressed. Populations often differ in the duration of diapause, but the adaptive reasons for this are unclear. 2. We performed a common-garden overwintering experiment with respirometric measurements in order to investigate the progression of diapause in the butterfly Pararge aegeria. Both the duration of diapause and the depth of metabolic suppression were shown to vary between populations. 3. In contrast to previous results from various insects, diapause duration did not correspond to the local length of winter. Instead, the observed pattern was consistent with a scenario in which diapause duration is primarily a product of selection for suppressed metabolism during warm autumn conditions. The relationship between optimal diapause duration and the length of the warm season is complicated by variation in the number of yearly generations (voltinism). 4. These results shed new light on variation in diapause ecophysiology, and highlight voltinism as an integrated product of selection at multiple points in the seasonal cycle.A ReadMe file has been provided, with detailed descriptions of the dataset. The dataset includes three R scripts that provide the necessary data parsing and processing for the results to be interpreted.Three-part dataset: 1) Main experiment: diapausing butterfly pupae (Pararge aegeria) were subjected to different durations of cold exposure. The date and success rate of adult eclosion was noted, and weight was measured at the beginning and end of the cold treatment. Metabolic measurements were carried out on a subset of individuals at regular intervals. 2) An additional set of metabolic measurements were carried out on nondiapausing pupae during their development into adults, and measurements were made at matching timepoints for diapausing pupae. 3) The seasonal cycle of each population was reconstructed to provide adaptive context for the laboratory results, using local temperature data, citizen-science observations of adult P. aegeria, and a mechanistic life cycle model previously described by Lindestad et al. 2019 (Ecology)