-We studied the autumn morning flight of Neotropical migrants for two years at Cape May, New Jersey. A majority (86%) of the 24,378 identified Neotropical migrants were warblers, while thrushes (0.01%) rarely made morning flights. A disproportionately large number of migrants was seen on the day of and the day following cold fronts. Only 23% of all Neotropical migrants flew on days not associated with cold fronts, and nearly 12% flew in rain. Morning flights of Neotropical migrants commenced in late July, and 72% passed before 15 September. All migrants flew north. At Cape May, 87% of visible morning flight occurred within 2 h after sunrise. We examine several hypotheses advanced to explain morning flight and reverse migration, and conclude that the morning flight of Neotropical landbird migrants at Cape May probably consists of a combination of birds returning to land from the ocean, birds compensating for lateral drift incurred during the previous night's migration, and birds seeking habitat in which to rest and forage. Topography may dictate the direction of morning flight. We suggest that researchers in North America commence studies of autumn passerine migration in late July and that researchers studying habitat selection by migrant passerines wait 2 to 3 h after sunrise before conducting surveys to avoid sampling birds undertaking morning flight. Received 18 March 1991, accepted 13 January 1992. THE MORNING FLIGHT of nocturnally migrating songbirds occurs in several locations in North America (Gauthreaux 1978, Bingman 1980, Hall and Bell 1981) and Europe (Alerstam 1978, Lindstrom and Alerstam 1986). In the autumn of 1988 and 1989, we studied the morning flight of Neotropical migrant songbirds at Cape May Point, New Jersey (38?56'N, 74?58'W). We sought to determine which species undertook morning flight, study their flight behavior, and describe the seasonal and diel timing of their morning flight. In addition, we examined the influence of weather frontal systems on the number of birds observed flying. In 1988, we conducted a daily transect survey through representative vegetation types to compare the species composition of birds remaining in the habitat to that of birds migrating out. This comparison allowed us to determine which of the Neotropical migrants undertook morning flight and which did not. With our data, we examine seven of the hypotheses Richardson (1982) listed to explain northbound migration in the fall, as well as Bingman's (1980) onward-migration hypothesis for diurnal passerine migration. I Present address: Department of Ornithology, The Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania, 19103, USA. The fact that so many species and such large numbers of migrants participate in morning flight strongly suggests that this behavior is an integral part of migration. Our study examines the function of morning flight. In addition, morning flight has important implications for researchers studying both seasonal timing of passerine migration and use of stopover habitat by passerine migrants.