Although there is strong evidence that a few Neotropical migrant land birds have severely declining populations, we know of no evidence of general overall declines. One widely cited analysis of data from the Breeding Bird Survey (BBS), the only continental—scale monitoring program for land birds in the breeding season, indicates declining populations in eastern forests in the 1980s, but other analyses conclude that most species have had stable or increasing populations in the last 25 yr. These conflicting results call for a new perspective. Here, we propose two approaches to the analysis of BBS data that emphasize the detection of temporal and geographic variation in population trends. The first approach uses nonlinear regression to detect temporal changes in population trends, and permits the construction of graphs that compare nonlinear trends among regions. The second approach provides statistical tests of multispecies patterns of geographic variation in population trends. We report analyses of BBS data for 1966 to 1992 for the 26 species of wood warblers (Parulinae) for which the most data are available and the chance of obtaining reliable estimates is best. Using pooled data for all 26 species for eastern and central North America, nonlinear regression indicated declining populations in the Appalachian Mountains and the Eastern Foothills in the 1970s, and stability elsewhere. As examples of the use of this first approach in single—species analyses, we give maps, graphs, and statistical tests to illustrate variation in population trends by physiographic strata in the Blue—winged Warbler (Vermivora pinus) and the Ovenbird (Seiurus aurocapillus). As an example of the second approach, we applied a probit—normal model to a matrix of increases and decreases in the 26 species in 37 geographic strata. The analysis identified specific highland areas as having unusually high proportions of declining populations. These results suggest that correlates of elevation should be considered as factors possibly regulating populations of warblers and other land birds. Our perspective is that focus on studies of temporal and geographic variation in population trends of all birds, including Neotropical migrants, could help conservation biologists identify where species and groups of species are in the most trouble.