On 18–19 February 1979 a major east coast cyclone deposited a record-breaking snowfall on the Middle Atlantic States. The storm is noteworthy because of the failure of the operational prediction models to signal the intensity of the event. The life cycle of the cyclone is reviewed with emphasis on the synoptic and mesoscale features and their possible linkage. Prior to cyclogenesis the synoptic pattern features a massive cold anticyclone near the Great Lakes with a broad baroclinic zone extending from Texas eastward to the Atlantic coast. A region of enhanced lower tropospheric baroclinicity develops along the Carolina coastal strip in response to significant oceanic sensible and latent heat fluxes which warm, moisten and destabilize the boundary layer. Cyclogenesis is initiated along the coastal front as the result of lower tropospheric warm advection. The importance of the coastal front is that it effectively steers the cyclone north-northeastward parallel to the coast such that it eventually acquires a favorable phase relationship for deepening with respect to a vigorous short-wave trough moving eastward from the Ohio Valley by 1200 GMT 19 February. Explosive deepening takes place in the ensuing 6 h coincident with the outbreak of convection near the storm center. By 1800 GMT, satellite pictures reveal a closed, clear storm eye while surface ship and drilling rig data disclose the presence of minimal hurricane force winds, primarily in the northern semicircle of the storm. Unlike a hurricane, however, the convection is asymmetric with respect to the vortex, being concentrated in the region of strongest surface winds. The major operational model errors stem from poor sea level pressure and quantitative precipitation prognoses. Evidence is presented that initial analysis deficiencies coupled with inadequate boundary-layer and convective precipitation physics precluded a successful model forecast in this cam.