On Wednesday, March 11, 1998, approximately 1957 hours Pacific standard time, a "Rocket Engineering" modified Mooney M20K, N3556H, registered to and being flown by a commercial pilot, sustained minor damage during a forced landing slightly south of the Klamath Falls International airport, Klamath Falls, Oregon. The pilot was uninjured. Visual meteorological conditions prevailed and an IFR flight plan had been filed. The flight, which was personal, was to have been operated under 14CFR91, and departed Sacramento, California, shortly after 1800. The pilot reported that the aircraft, which was parked on the ramp (outdoors) at Boeing Field, Seattle, Washington, during the winter months, was not flown much. Prior to departing for Santa Monica, California, on March 8th, he pre-flighted the aircraft and "observed a puddle of water (smell & taste tested) under the aircraft low point - unusual because the tarmac was dry." He "drained the fuel lines using the gascolator without testing, but tested all four tanks." He stated that the right tanks tested fine but he found "an estimated total of a gallon of water from the two left tanks." The flight to Santa Monica, with a fuel stop at Sacramento was uneventful. On March 11th, he pre-flighted the aircraft at Santa Monica and "drained about 1/2 cup of water from the left side" before departing back to Seattle. The first leg of the trip, which was flown at 12,000 feet MSL, again stopped at Sacramento for fuel. The aircraft departed Sacramento at 1800 and was cleared to climb to flight level 240. He reported the outside air temperature as -27 degrees Centigrade after passing Red Bluff, California. Approximately 1925 the pilot "noticed a little engine roughness." Several minutes later he decided to proceed to Klamath Falls, Oregon, as he had "a rough-running engine." Descending through FL180 engine performance had become increasingly degraded with EGT/CHT readouts from the three right cylinders consistently lower than the readouts for the left three cylinders. The pilot executed a number of trouble shooting procedures attempting to correct the problem without success. He reported that the engine was now backfiring. Approximately 5-7 miles from the Klamath Falls airport the engine ceased developing power. The pilot then executed a forced landing under dark night conditions to an unlighted landing site. He reported lowering the landing gear prior to ground impact but did not feather the propeller. He stated that "in the end, I just descended into total blackness." Klamath County Sheriff's Department personnel responded and reported that the aircraft came to rest in an open field (refer to DIAGRAM I which shows the crash site location). The right main wheel was found approximately 90 feet from its initial ground impact scar and the ground slide from initial touchdown to final resting place was approximately 684 feet. Both the right main landing gear and the nose landing gear were observed to have collapsed shortly after the initial ground impact. The aircraft was modified in accordance with a Supplemental Type Certificate (STC) to a Rocket Engineering Corporation model Rocket 305. This modification included the replacement of the original 210 horsepower Continental TSIO-360-GB engine with a 305 horsepower TSIO-520-NB engine. The modifications included full feathering capability for the propeller. Post-crash examination of the aircraft at the site by an inspector from the FAA's Hillsboro Flight Standards District Office was conducted. The inspector reported that "fuel was drained from the sumps" and that "although the aircraft was positioned in a right wing low attitude several ounces of water were drained from the left wing tank." He also reported that "it appeared that N3556H touched down at (a) high rate of descent but continued to roll for approximately 200 yards before the right main gear departed the aircraft and the nose gear collapsed" (refer to attached statement). Post-crash examination of the aircraft by a mechanic from Klamath Aircraft, Inc., was conducted. The mechanic reported that the fuel caps, adapters and associated retaining chains looked good with no rust or corrosion noted. He also reported that the "O" rings in the fuel caps were in "fair" condition and that their "fit could be tighter" (refer to attached statement). Post-crash testing revealed no mechanical malfunction with the engine.