Figure 1.

Scheme of experimental evolution by hybridization of differentially thermally-adapted subpopulations of fruit fly. Prior to the application of thermal selection, we created a series of replicated experimental populations, by combining flies of isofemale lineages collected from the Melbourne (putatively cool-adapted, or “C”) subpopulation, denoted in blue, and the Townsville (putatively hot-adapted, “H”) subpopulation (red). This was achieved over two generations, via a process of admixture of the individual isofemale lineages. In the Admixture 1 step, we pooled 5 virgin females (♀) from each of 18 of the H isofemale lineages, with 5 virgin males (♂) from each of 18 C isofemale lineages into one bottle, denoted by HC = 18 × 5(♀H) + 18 × 5(♂C). In parallel in Admixture 1, we performed the reciprocal cross wherein $H<=>\mathrm{C}$ above, denoted by CH = 18 × 5(♀C) + 18 × 5((♂H). Each bottle contained 90 males and 90 females (180 flies). In the following generation, at Admixture step 2, we combined 25 virgin females and 25 virgin males from HC bottles together with 25 virgin females and 25 virgin males from CH bottles, 25(♂CH) + 25(♀HC) + 25(♀CH) + 25(♂HC), across 15 biological replicates (7 of which were descendants of flies treated by antibiotics, 8 of which were descendants of untreated flies). At this stage, all flies had been maintained in standard laboratory conditions (25 °C) for 16 generations (14 generations as isofemale lineages, 2 during the admixture process). We then divided each of these 15 biological replicates into 4 subpopulations, subjecting each subpopulation to one of four thermal treatments (19 °C, 25 °C, fluctuating cold, and fluctuating warm), with each experimental subpopulation containing around 500 individuals. On the left side of the figure, yellow text denotes sample sizes associated with each stage of the admixture process for flies whose ancestors had been exposed to antibiotic treatment (ATB), while grey text on the right corresponds with untreated flies (UTR).