On 2 Sep 1996 Malcolm Watts (New Zealand) reports a weekend spent on two-coil systems in his living room. He replaced the spark gap with four paralleled MTP3055E MOSFETs fired from a pulse generator. The primary was 5.3 turns of copper tubing on two 560 nF capacitors in series, coupled at 0.1 to a compact secondary of 1900 turns of 0.2 mm wire resonant at 153 kHz. The supply went up to 30 V. He was after quench and dwell times and CW operation in a disruptive system rather than sparks. In 2004 he added that by the end of the experiments the coil was making half-inch sparks, and called it a precursor of one of the later topologies. So here are transistors switching a charged tank into the primary, six years before Fritz's Off-Line Tesla Coil.
Sources
This is the card from the driver tree, written from what the builder published. Send corrections through the forum.