Offspring measurements
Time of larviposition was recorded, including whether or not larvae produced were viable (Baldry et al. 1992). Production of non-viable offspring was considered a spontaneous abortion (‘abortion’ hereafter). It was not possible to record egg abortions as they were not visible due to their small size. For pupae collected on a Monday morning, we consulted the video recordings to determine the exact day of larviposition (S2 File). For aborted larvae, however, we could not determine the date for those aborted over the weekend, because early larval stages were also not visible on the camera and were therefore recorded as occurring on the Friday. For viable pupae, we measured the wet weight, to 0.1 mg. Of the pupae produced, we randomly selected 70% of samples to track emergence. The remaining 30% were destructively sampled for fat analysis (S2 File) in order to quantify how wet weight correlates with fat reserves.
Pupae assigned for emergence studies were placed singly into 50 ml Falcon tubes with a 3 mm hole drilled in the centre of the screw cap to allow air flow. Pupae were observed daily for date of emergence on working days and again video recordings were used to determine the date of emergence on weekends. Pupae were observed for a maximum of 50 days. Any pupae that had not emerged by 50 days, were recorded as a failed emergence (Hargrove 2004). Each emerging fly remained in the tube until it died of starvation. Sex of emerging flies and date of death were recorded.