Statistical analysis
Data were analyzed using SAS v. 9.4 (SAS Institute, Cary, NC). Samples that comprised fewer than 10 fruits were excluded from the analysis. Week represents time after the first sample was collected and held constant for both years. Cultivation type denotes whether the berry was wild-grown or cultivated. Eggs per berry are the mean number of eggs per fruit while eggs per gram were calculated using the number of eggs per berry divided by the average per berry weight of each sample group. The eggs per gram value was log transformed to adjust for assumptions of normality. Unless noted otherwise, we used a generalized mixed model (GLIMMIX) with a log normal distribution. Adjusted means were compared using the Tukey-Kramer adjustment.
In order to examine average infestation by cultivation type and ripeness stage, ripeness stage and cultivation type were considered fixed effects, and year, elevation nested within year, location, and week nested within year as random effects. For weekly infestation rates, eggs per gram was used as the dependent variable, week, ripeness stage and cultivation type were considered fixed effects, and year, location, and berry nested in location by week were random effects.
The effect of elevation was assessed for JKWA samples from 2018 with elevation, ripeness stage and week as fixed effects. The proportion of infested fruit at each sampling point (ripeness stage/location/week) was calculated as the number of berries with one or more eggs divided by the total number of berries in that sample group. Data were fitted to a normal distribution via Proc GLIMMIX with cultivation type, ripeness stage, and their interaction as fixed effects and location, year, and elevation nested within location as random effects.
Oviposition preference was calculated as the proportion of eggs laid in either the wild or cultivated berries divided by the total number of eggs laid in each replicate. These proportion data were then evaluated with a paired Student t-test; replicates with non-responding flies (those which did not lay eggs during the experimental period) were removed from analysis.