Experimental setup
We performed a cross-factorial experiment in which parental flies (P) were raised in constant and variable thermal environments, and their offspring were split and maintained in either the parental environment or the opposite (Figure 1). To obtain this experimental design, we use more than 200 inseminated D. melanogaster collected in central Chile (33°26′S; 70°39′W at 500 m above sea level) during 2016 in a nearly 500 m2 habitat. After collection, twenty groups were established with approximately ten females each. Groups were reared in controlled conditions at a constant ambient temperature Ta = 24 °C and a Light:Dark 12:12 photoperiod. Flies were maintained for three generations in 250 mL glass vials with Burdick culture medium (Burdick, 1955). Third generation adult males and virgin females from this stock were randomly assigned to two thermal treatments that differ in the variance of temperature. Thermal treatments were 28 ± 0 ° C and 28 ± 4 ° C, a constant (C) and variable (V) thermal environment, respectively, and crossed under these conditions. Acclimation temperatures were chosen based on the well-known limits of fruit fly eggs viability (Hoffmann, 2010) (eggs-to adult viability is 80% at 28ºC and 0 to 5% at 32ºC, for details see (Hoffmann, 2010; Cavieres et al. , 2018)). In the variable thermal environment, during the day, temperature started to increase linearly at 7:00, reached the maximum at 11:00, then stayed constant, and began to decrease at 19:00 and reached 24°C at 23:00h, the heating/cooling rate between the minimum and maximum temperatures was 0.03°C min−1 (Figure 1).
The offspring, which corresponds to the parental generation P in our breeding setup, was maintained from eggs to adult in each thermal treatment. Subsequently, adult males and virgin females from both treatments were evenly divided into two breeding groups and transferred to constant and variable thermal conditions. Their offspring correspond to the F1 in our breeding setup, resulting in a factorial experiment with two P (C and V for constant and variable, respectively) and four F1 groups (CC, CV, VV and VC, which reflect both the parental and offspring thermal environment). As detailed below, we estimated for all P and F1 groups the lower and upper thermal critical limits (CTmin and CTmax , respectively), net reproductive rate (R 0) and generation time (Tg ).