Capture
To assess circulating corticosterone (n=43) and pituitary gene expression for a subset of subjects (n=13), we captured experimentally parasitized females 2 hours after the addition of a model egg. The 2 h time point was chosen because plasma corticosterone in birds typically increases rapidly in the first 30 min to an hour following an exposure to an acute stressor (such as handling or nest disturbance), after which plasma corticosterone titer remains near its maximum if the stressor persists (Abolins-Abols et al. 2016; Romero et al. 1998; Meddle et al. 2003; Breuner et al. 2006). For logistical reasons we were not able to remain near the nest to determine the return time of each female, but in this population, females typically arrive back to their nests on average 12 min after being flushed (pers. obs. n=31, standard deviation (SD) = 7.37, range = 2 to 28). Most of the females (86%: 38 out of 43) were flushed from their nests prior the insertion of the model egg (the proportion of flushed females did not differ between the treatments (Fisher’s exact test, p=1.0)). The 2-hour time window therefore provided most of the females with an opportunity for a prolonged interaction with the model egg before capture, allowing any corticosterone increase in response to the model eggs to reach its maximum.
Previous research also showed that most female robins’ decisions to reject or accept a model egg are not immediate but occur on the timescale of hours rather than minutes (Hauber et al. 2019; Scharf et al. 2019). Importantly, the only other study we are aware of in Turdus thrushes investigating the effect of experimental parasitism on plasma corticosterone detected elevated corticosterone 24+ hours after the addition of the model egg (Ruiz-Rayaet al. 2018). However, sampling robins at 24 hours following the experimental parasitism without tethering the model eggs would have meant that nearly all the females would have rejected the non-mimetic parasitic eggs (e.g. Hauber et al., 2020b) and that they would therefore have been unstimulated for several hours before capture. The 2-hour time window thus allowed us to catch females during or closely following stimulation by model eggs, and, according to our estimate, allowed sufficient time for any changes in corticosterone in most females reach or remain near peak levels.
We captured females with a 6 m mist net, arranged in a V-shape around the nest tree. The poles and a rolled-up closed mist net were set up immediately before the addition of the model eggs. Two hours after the initial net setup and the addition of the model egg, we unfurled the mist net and captured the female either by flushing it into the net or passively capturing it as it attempted to land on the nest. Setting up the net prior to experimental parasitism enabled rapid set up and capture following the 2-hour mark (n = 42, mean capture time = 15.33 minutes, min = 0, max = 36, SD = 9.19) and allowed us to minimize any changes to corticosterone levels due to nest disturbance. Importantly, we found no association between baseline plasma corticosterone concentration (see below for corticosterone assay methods) and the time it took to catch the female (Pearson’s r =-0.16, df=40, p=0.30) or whether the female was flushed from the nest (two-tailed t-test, t=0.15, df=4.64, p=0.88). Therefore, we did not account for these metrics in the statistical models testing the effect of egg type on corticosterone levels (see section below). At the time of capture, we also noted whether the model egg was rejected or still present (“accepted”).