A total of three negative rocket-triggered lightning flashes without return strokes (two from 1997 and one from 1993) are analyzed in this paper in order to study the processes associated with the disintegration of the triggering wire and its replacement by an air-plasma channel. It appears that the gap resulting from the vaporization of the triggering wire by the upward-positive leader current is bridged by a leader/return-stroke type process. Electric fields at distances of 50, 110, and 500 m, the corresponding magnetic fields at 500 m, and the currents to ground are examined for the two 1997 flashes. The electric field prior to the triggering wire's vaporization in these flashes exhibits a positive (atmospheric electricity sign convention) millisecond-scale ramp due to the upward-extending positive leader. The electric field changes observed at the three distances just prior to wire vaporization are consistent with an equivalent point charge of about 0.3 C at a height of 1.2 to 1.5 km, suggesting that the charge density distribution at that time is strongly skewed toward the upward positive leader tip. The length of the triggering wire at the time of its vaporization was estimated from still photographs to be about 210--220 m. Following the ramp, a microsecond-scale V-shaped negative pulse, which resembles the close electric field signature of a small dart-leader/return-stroke sequence, is observed. The corresponding magnetic field decreases abruptly, simultaneously with the onset of the leading edge of the V-shaped pulse, to values near zero and remains there for tens of microseconds, indicating the attempted interruption (cutoff) of the upward positive leader current flow to ground through the triggering wire as it is vaporized by this current. Following the abrupt decrease, the magnetic field exhibits a rapid increase at a time corresponding to the trailing edge of the V-shaped electric field pulse, suggesting that the vaporized triggering wire is replaced by an air-plasma channel that becomes part of the upward positive leader channel when electrical connection to ground is restored. For the third triggered lightning flash, from 1993, similar inferences regarding the processes involved in the replacement of the triggering wire by an air-plasma channel are made from high-speed (streak) photography and from measurements of the current to ground and electric field at 30 m. In this flash, the upward positive leader exhibited very pronounced stepping: step current pulses had peaks, as measured at the ground, that were up to a few kiloamperes, and step charges that were up to 100 mC. Characterization of the attempted interruption and the following reestablishment of current to ground in rocket-triggered lightning may have important implications for the understanding of channel current cutoff in natural lightning flashes and may provide new insights into the formation of strokes observed to occur in the same channel within a millisecond or less.