One of the most efficient defences against obligate brood parasitism in birds is egg ejection, where a host recognises and removes the parasitic egg from the nest. This defence often selects for egg mimicry in parasitic species to reduce the likelihood of egg ejection. If a parasite uses multiple host species with distinctive egg types, this could lead to the evolution of egg gentes (host-specific egg types) in the parasite. There is observational evidence that the brood parasitic Brush Cuckoo (Cacomantis variolosus) might exhibit egg gentes, but there has been no objective study conducted to determine how closely eggs of this cuckoo species resemble those of its hosts from a bird’s visual perspective. Using objective measurements to quantify egg appearance, we found that Brush Cuckoos exhibit at least two egg morphs that closely match the eggs of two of its primary hosts in colour, luminance and volume. While the determination of actual egg gentes in the Brush Cuckoo was beyond the scope of our study, our results are a first and necessary step in determining whether egg gentes might exist in this species. We suggest at least a third egg morph matching another primary host (or at least the genus of that host) might exist, but more data would be necessary to confirm this. Additionally, we provide a mechanism researchers can use to help distinguish between Brush Cuckoo eggs that are closely matched to their host eggs for future studies in this system.

We are providing three types of data in this dataset: average spectral reflectance of eggs (the average for each egg measured), egg pattern energy (for each egg measured) and egg volume (for each egg meausred). These data could be used in further analyses by other researchers to calculate values such as JNDs, photon catch cone values, etc. Eggs were measured at the Australian National Wildlife Collection, CSIRO, ACT in 2019. Each egg is identified with the clutch ID from the museum. Abbreviatons of species names are as follows: BRCU = Brush Cuckoo; BRHE = Brown-backed Honeyeater; GRFA = Grey Fantail; LEFL = Leaden Flycatcher; RUFT = Rufous Fantail; RBFA = Red-backed Fairy-wren; REFL = Restless Flycatcher; BBHE = Bar-breasted Honeyeater; RBHE = Rufous-banded Honeyeater. Data of minor host eggs and cuckoo eggs from minor host nests were only included for the spectral reflectance dataset. Methods of how each dataset was collected are detailed below and can also be found in the article associated with this dataset.

We measured one cuckoo and one host egg from 32 parasitised clutches. We took objective measurements of egg colour and luminance by measuring their spectral reflectance from 300-700 nm using an Avantes AvaSpec-2048 fiber optic spectrometer, an AvaLight-DHc light source, an FCR-7UV200-2-ME reflection probe with a 6.35 mm diameter, and the AvaSoft 7.7 software (Avantes, the Netherlands). The light probe was held inside an RPH-1 probe holder at a 90° angle to the egg. Before taking a set of measurements, a WS-2 white reflectance tile was used to take a white reference and a dark reference was taken after the light source had been switched off. Nine measurements of each egg were taken in three regions around the egg (three measurements at the cap, three around the middle and three at the blunt end). These nine measurements were averaged together to obtain an average spectral reflectance measurement of each egg. Due to the small sizes of the eggs and their spots, we did not distinguish between spot colour and background colour. To reduce noise from ambient light, measurements were taken in a darkened room with the lights turned off to obtain the darkest lighting possible. What we provide in this dataset is the average spectral reflectance of each egg from 300-700 nm averaged to 1-nm intervals using the pavo package (Maia et al. 2019) in the R Statistical Package (R Core Team 2019).

To measure egg pattern, we took a single photograph of each clutch (one host egg and one cuckoo egg) on an 18% grey standard Mennon card using a Canon EOS M100 camera with a 15-45 mm lens. We always had the lens set to 15 mm, as this produced the clearest image of the egg. Objective image analysis was performed using the multispectral image calibration and analysis toolbox (Troscianko and Stevens 2015) for ImageJ (Rasband 1997-2014). The toolbox performs image calibration, ensuring linear reflectance images that control for lighting changes are used for image processing. Granularity pattern analysis was performed using standard bandpass methods on the camera’s green reflectance channel as this most closely approximates bird double cone peak sensitivities (Spottiswoode and Stevens 2010), which are thought to be important in the perception of pattern (Jones and Osorio 2004). We calculated pattern energy spectrums for each egg using a size range of 2-512 px with all photographs standardised to a 33 px/mm scale (Troscianko and Stevens 2015), which is the data we have provided in this dataset.

To determine if Brush Cuckoos match their hosts in egg volume, we used the volume of each egg estimated by the multispectral image calibration and analysis toolbox (Troscianko and Stevens 2015). This model used the same photograph we took for measuring egg pattern. We included the same scale bar in each egg photo and used this to standardise the measurement to obtain an estimate of volume (ml), which is the data we have provided in this dataset.

Please see associated README file for information on what this data is and how this data was collected and processed.