Sources of variation in reflectance spectrophotometric data: a quantitative analysis using avian eggshell colours

Abstract

SUMMARY: 1. The evolution of coloration has generated some of the most diverse and variable phenotypes, both within and between species, in animals and plants. The objective quantitative assessment of the physical, chemical and perceptual basis of coloration has been greatly improved by the application of portable full-spectrum spectrophotometers to measure reflectance data. Yet, the analysis and interpretation of physical measures of colour spectra must be conducted within the constraints of the uncertainty regarding the relative impacts of methodological vs. biological sources of variation. 2. Here, we characterise the components of variation in data on reflectance spectra, related to sample storage and measurement equipment, to characterise colourful pigmentation of eggshells (Class: Aves) of two Turdus species. We quantified longitudinal shifts in reflectance occurring over repeated measurements of the same sets of avian eggshells. Shells were sampled at the time of collection and again after 5 years of dark storage using the same equipment. These data were then compared with spectra obtained from the same eggs after the storage with a different model of spectrophotometer for three colour metrics [blue-green chroma (BGC),UVchroma and brightness]. 3. Blue-green chroma and brightness of the same eggs varied systematically between years at a similar magnitude to the biological variation among different eggs. This suggests the need for future research into the extent of chemical and physical deterioration of eggshell appearance even during relatively short-term storage. The variation introduced by using two different spectrophotometers also was significant but relatively small compared with biological levels of variation for UV and brightness. 4. Our results confirm quantitatively that museum eggshell specimens are suitable for interspecific comparative analyses, but also highlight the requirement to account explicitly for variation in storage duration and measurement equipment, when ‘objectively’ comparing biological variation in coloration across individuals in space and also in time.