SUMMARY This paper reviews qualitative results of study of 35 mm. moving pictures taken of PPI scopes of 23 cm. radar installations on the southeast coast of New England during spring and autumn migrations in 1959 and 1960. Gulls, waders and passerines are readily distinguished from each other; small, medium, and large movements are easily separated. New England weather is controlled by eastward flow of large air masses out of the continent of North America or the Gulf of Mexico. These take several days to cross New England, so that short-range weather forecasts are more reliable than in northwestern Europe. Winter is characterized by regular eastward passage of air masses. Wintering buntings increase their fat deposits in low temperatures and move their feeding grounds with storms. Unstable early spring weather marks the end of winter. Early spring migrants respond to 10° rise in temperature and usually move with overcast skies. The dispersal of flocks of wintering Tree Sparrows studied over five years and conspicuous first arrivals of spring migrants, including large flocks of day-migrating icterids, may be as early as 22 February or as late as 10 April and coincide with the first moderate movements on radar. Late spring is marked by orderly flow of maritime air up the east coast of North America, either with the movement of low-pressure systems or with circulation around the Bermuda high. Late spring migrants are more regular in their schedule of arrival and respond to clear skies, rising temperatures, and favourable winds. In mid- and late May, stalling of the warm air in a reentrant centred over coastal New England results in the grounding of large numbers of birds which winter in Central and South America. Week-long rain resulting from such stalled weather produces especially large movements when the skies clear. Passage of a sharp cold front stimulates reversed migration during spring months, more marked in March and April, but still evident in May. In early August, adults start to migrate and continue as a steady flow in small numbers, which increase with the passage of each cold front until early September. As the sun's declination falls, cold continental polar air masses cross New England, accompanied by increasingly evident cold fronts and falls of temperature during September and October. Warblers and thrushes move with each cold front in early September; buntings join them in late September and become predominant among night migrants in October. In September, tropical storms on the Atlantic coast bring southern stragglers with them. Wintering buntings appear and day-migrant icterids again are conspicuous in October and November, with continued cold fronts and stalled high-pressure systems of Indian Summer. The first freezing weather sends icterids, geese, White-throated and Fox Sparrows farther south in reversed order of spring arrivals. Unstable early spring weather marks the end of winter. Unstable stalled air masses and late autumn northeast storms forecast winter's arrival. Early spring migrations, especially of diurnal migrants, are reliably reported by ground observers. Large movements are similarly well reflected, especially by reports of a general exodus. Ground observer reports which indicate small movements every day are usually correct. Laboratory recordings of migratory unrest accurately show the span of the instinctive period and closely approximate the schedule of movements seen in the field; but recordings do not yet clearly show effects of weather stimuli. We suggest the hypothesis that specific changes of weather, for example in autumn, overcast and rain, followed by heavier rain and squalls, followed by clearing and cold weather, combine (including the proper sequence of changes) to make additive stimuli whereby a population reacts to fly in an air mass which will be most suitable for migration. Radar shows three regular offshore movements across eastern and coastal Massachusetts in autumn. Although adaptations exist which allow certain species to avoid the overwater crossing or to re-orient and regain land, most of the birds continue to cross the sea, presumably because they fly non-stop to their goal. The three offshore movements are as follows. Regular southeastward movements on radar are of waders, are not consistently weather-controlled, and are most frequent from mid-July to mid-October. Regular southwestward movements on radar are of passerines, primarily thrushes, wood warblers and buntings, and form the eastern limit of the general southwestward flow of these species. These movements follow cold fronts from early August through mid-November. The unexplained puzzle of heavy concentrations of immature songbirds on the coast is discussed. Regular southward movements on radar are of passerines, and follow cold fronts, occurring most frequently from early September to mid-October. We believe these movements to represent a previously unrecorded migration of wood warblers that winter in the Antilles and South America, whose track cuts directly across the western Atlantic from New England and the Maritime Provinces. Some estimates are made of the rate of energy consumption during these flights, based on the observed departure weights and the recognition that the flights are successful. Calculations based on weights of birds in New England and in Bermuda show that the birds possess abundant energy resources for this flight.