Diet
The diet of black wheatears was very diverse, with 337 prey items of 96 families and 29 orders (Table S3). Arthropods were detected in all samples and belonged to 5 classes and 22 orders, of which 17 orders were Insecta. The main prey belonged to the order Hymenoptera (Frequency of Occurrence: 83%), mainly ants (family Formicidae; 75%). Frequent animal orders that were detected in more than half of the samples included Lepidoptera (67%), mainly belonging to families Noctuidae (30%), Pterophoridae (25%) and Geometridae (15%); Coleoptera (62%), mainly Tenebrionidae (28%) and Carabidae (13%); Orthoptera (54%), mainly Acrididae (42%); and Diptera (51%), with 10 families identified but none detected in more than 10% of droppings. There were also other important arthropods as Hemiptera (40%), mainly from the family Pentatomidae (16%); and Araneae (34%), mainly Salticidae (11%). The only vertebrates found were lizards (Squamata) detected in two droppings. The vegetal component of the diet was less diverse, but also very common (60% of the droppings), with Solanum nigrum (order Solanales, family Solanaceae) being the most frequently detected (35%) (Figure 1; Table S3).
We found no differences between sexes in the average number of prey items detected per sample, irrespective of taxonomic resolution: highest prey item resolution ( = 8.344; LR Chisq = 0.232, df = 1, p = 0.630), families ( = 5.739; LR Chisq = 0.130, df = 1, p = 0.718) or orders ( = 5.226; LR Chisq = 0.083, df = 1, p = 0.773). However, the overall prey richness was higher for males than females for the analysis carried out at the highest prey item resolution (even if a 95% confidence interval was considered), while no significant differences between sexes were detected for analysis based on identifications at the family or order levels (Figure 2).
Regarding diet composition, we found a significant difference between sexes at the highest prey item resolution (Res. Df = 91, Deviance = 427, p = 0.006), family level (Res. Df = 90, Deviance = 139.9, p = 0.021), but not at the order level (Res. Df = 91, Deviance = 44.52, p = 0.054). The univariate tests showed that the differences found were due to 11 prey items and 6 families (Table S3). The prey item most important for compositional differences was one unidentified Myrmicinae species, that was also the prey most often detected in black wheatear droppings. This ant species was detected in 58% of females’ droppings, while in males it was only detected in 29% of droppings (Table S3). At the highest resolution level, all other prey had differences in frequency of occurrence between sexes smaller than 10% (Table S3). At the family level, the differences were mainly due to the families Pentatomidae, Formicidae, Tettigoniidae, that were preyed 24%, 21% and 11%, respectively, more often by females, while males preyed 23% more often on Tenebrionidae (Figure 1; Table S3). There were also 2 orders that differed between sexes (Hymenoptera and Santales), despite the overall effect of sex being non-significant when analysing prey composition at the order level (Table S3).