Stable-isotope analysis of feathers is an increasingly important source of information on diet and movement of birds. Feathers are typically cleaned with a solvent before analysis, but the effects of this cleaning on the resulting data have not been examined critically. We conducted an experiment to determine whether a cleaning regime affected hydrogen (δD), carbon (δ¹³C), and nitrogen (δ¹⁵N) stableisotope ratios in feathers of Japanese Quail (Coturnix japonica). A paired design was used to clean feathers with 2:1 chloroform: methanol or detergent. Results after initial cleaning indicated enrichment of hydrogen by ∼40‱ in feathers treated with 2:1 chloroform: methanol-treated compared with other treatments but no correlation between treatments. We found a similar pattern among treatments for carbon, but the effect was on the order of 0.2‱. Nitrogen values showed no discernible correlation, but both uncleaned and 2:1 chloroform: methanol treatments had enriched values, on average, compared with detergent treatments. Further, variance among samples was high for hydrogen and nitrogen measurements. After recleaning with alternate treatments, differences in mean hydrogen-isotope ratios were no longer evident, data for hydrogen and carbon became less variable, and carbon maintained its initial pattern. We suggest a standard method of cleaning feathers, first with a dilute detergent solution, then with 2:1 chloroform:methanol solvent. This study has implications for increasing repeatability of hydrogen, carbon, and nitrogen stable-isotope measurements, which would increase the validity of inter-laboratory comparisons and the utility of large-scale projects using compiled data sets.