We evaluated 3 techniques (air jet, flat skin, grab sample) to assess molt in dead waterfowl (greenwinged teal [Anas crecca] and king eider [Somateria spectabilis]). The air jet technique was fast and required no specimen preparation. The flat skin technique required lengthy preparation time after which molt was assessed quickly. Both techniques underestimated the extent of molt (P < 0.01). The grab sample technique needed no preparation but involved time-consuming counting to obtain the molt assessment. The grab sample was preferred when study skins did not need to be preserved because it was easy to conduct, and it was repeatable and more accurate than the other techniques. J. WILDL. MANAGE. 54(2):219-222 Comparison of studies reporting molt is often difficult because different techniques are used to assess molt. The most frequently used techniques are (1) gross examination of plumage (Watson 1963, King 1972), (2) detailed examination of feather tracts (Greenwood et al. 1983) or various scores indicating proportions of molting feathers (Oring 1968, Morton et al. 1969, Owen and Krohn 1973), (3) measurement of new growing flight feathers (Weller 1957, Johnson and Johnson 1983), and (4) scoring of molt on prepared flat skins (Oring 1968, Billard and Humphrey 1972, Young and Boag 1981). Our objective was to compare different techniques of molt assessment in terms of information obtained for time spent. We were concerned with molt of body contour feathers (not down) of fully grown birds. Molt of remiges and rectrices was not considered because it varies little among individuals (Bailey 1982). We are grateful to M. O. Hammill, T. G. Smith, T. W. Aldrich, M. R. McLandress, and hunters visiting the California State Wildlife Areas for assistance in collecting ducks. D. G. Raveling hosted RDT at the University of California-Davis and provided help, as well as access to facilities in the Department of Wildlife and Fisheries Biology. H. A. George and R. B. Reno, California Department of Fish and Game, were helpful in granting access to wildlife areas. M. A. Fanous gave statistical advice. We are grateful to D. M. Bird, D. S. Gilmer, P. S. Humphrey, P. Stettenheim, and G. A. Baldassarre for reviews. E. A. Titman assisted with field work, and J. Viallet assisted in the laboratory. Financial support was provided by the Natural Sciences and Engineering Research Council of Canada. McGill University provided computer