Efforts to tailor waterfowl hunting regulations to conditions in the Atlantic Flyway have been hampered by lack of information on local breeding populations. The Atlantic Flyway Council's technical section voted at its 1987 winter meeting (Atlantic Flyway Council Technical Section, Toronto, Canada) to establish a regional waterfowl breeding survey. Consequently, an annual survey was started in 1989 and further refined in 1993 using results from 1989 to 1992. During 1993-1997, annual spring surveys of more than 1,450 randomly selected 1-km2 plots, stratified by physiographic strata, were conducted in the Atlantic Flyway from New Hampshire to Virginia to estimate breeding populations of mallards (Arias platyrhynchos), American black ducks (A. rubripes), wood ducks (Aix sponsa), and Canada geese (Branta canadensis). Ground crews systematically surveyed all potential waterfowl habitat for these species in each plot. The adjusted mean mallard pair estimate over the 5-year period was 375,962 (range 310,299-415,182, mean SE 25,761) for the region surveyed. The estimate for black duck pairs was 31,1 54 (range 27,164'37,521, mean SE 4,978), and for wood duck pairs it was 240,473 (range 218,959-281,916, mean SE 25,408). Total number of Canada geese increased from 526,663 in 1993 to 892,278 in 1997. Population estimates for other species had unacceptably large standard errors.