Conspecific brood parasitism (CBP) is an alternative reproductive tactic in egg-laying species, where a female lays her eggs in the nest of a conspecific. In a population of spotless starlings (Sturnus unicolor), some eggs are ejected from the nest during CBP events. It is unclear whether this ejection is a parasitic strategy (a host egg is ejected) or an anti-parasitism strategy (the parasitic egg is ejected). To clarify this, we genotyped the eggs ejected on the ground and found that 100 % of them were parasitic. Egg discrimination might be based on tactile or visual cues, and we hypothesized that egg size could be used by hosts to eject parasitic eggs. We conducted experiments in the field using dummy eggs of varying sizes. The results showed that starlings were more likely to eject eggs if they were smaller than their own eggs. In contrast, no significant pattern of egg ejection was observed for larger eggs. Our results suggest that starlings use egg volume recognition as an anti-parasitism strategy to avoid the costs of parasitism. Whether this is a frequency-dependent strategy is worth further studies.

We simulated 63 parasitism events by adding one dummy egg (3D-printed plastic or plaster eggs) during first and replacement clutches in 2023. To make sure the effect of dummy egg addition, if any, was not due to the egg material, we compared the ejection rate between natural eggs (N = 74), plaster eggs (N = 14), and plastic eggs (N = 49). For the natural type, we used natural parasitism in the population and quantified the percentage of egg ejection. As most of the ejected eggs were broken, we do not have data on their size. All types of nests (with natural vs. artificial parasitism) were checked daily and suffered similar experimental disturbances.

To reproduce natural parasitism, we placed a dummy egg in 54 nests on day three of laying (nine nests were used twice: once during the first clutch and once during the replacement clutch). We used only nests that were presumed not to have been previously parasitized and that presented a regular laying pattern, i.e. nests with three natural eggs on day three. We measured the length and the width of these three eggs with digital calipers (Mitutoyo, Japan) to the nearest 0.01 mm. Egg volume was calculated by the formula: 0.51 × length × width² [39]. According to the mean volume, we chose a dummy egg to put in the nest. We had five categories: minus 2 cm³, minus 1 cm³, control, plus 1 cm³, plus 2 cm³ – with respect to the mean volume of eggs already in the nest. Nests were randomly assigned to each category (Table 3). To test a nest under a specific category, we selected the dummy egg that most closely matched the target volume defined by the assigned category (see Table 3 for details).

We then followed nests daily to detect ejection behavior. If the dummy egg was not in the nest box anymore, we considered it “ejected”. On the contrary, if on day six of laying, it was still inside the nest box, we considered it “accepted”, and we removed it from the nest box.