An unusual, isolated hailstorm descended on Boulder, Colorado, on the evening of 24 June 2006. Starting with scattered large, flattened, disk-shaped hailstones and ending with a deluge of slushy hail that was over 4 cm deep on the ground, the storm lasted no more than 20 min and did surprisingly little damage except to vegetation. Part I of this two-part paper examines the meteorological conditions preceding the storm and the signatures it exhibited on Weather Surveillance Radar-1988 Doppler (WSR-88D) displays. There was no obvious upper-tropospheric forcing for this storm, vertical shear of the low-level wind was minimal, the boundary layer air feeding the storm was not very moist (maximum dewpoint 8.5°C), and convective available potential energy calculated from a modified air parcel was at most 1550 J kg1. Despite these handicaps, the hail-producing storm had low-level reflectivity exceeding 70 dBZ, produced copious lowdensity hail, exhibited strong rotation, and generated three extensive bounded weak-echo regions (BWERs) in succession. The earliest of these filled with high reflectivities as the second one to the south poked up through precipitation-filled air. This has implications for low-density hail growth, as discussed in Part II.