Relative feather mass indices: are feather masses needed to estimate the percentage of new feather mass grown for moult regression models?

Abstract

During the moult of primary wing feathers in birds it is likely that new feather material is being produced at an approximately constant rate if the energetic requirements of the birds are met. In moult regression analyses it is, therefore, desirable to transform moult measurements into a variable based on the amount of new feather material produced because of its linear properties. This can be achieved if the mass of each primary feather is known, and used as a weighting factor in a moult index. These primary feather masses would have to be known for every species being analysed. As an alternative, it is here tested if feather mass indices may be sufficient replacements for species-specific feather masses. Thirty-five species of birds with known primary feather masses were divided into four wing-shape groups, and a feather mass index was built for each group. Within each group, comparisons were made between estimates of moult parameters using the moult index with those using the known primary feather masses. Within groups there were no significant differences between any of the moult estimates.