On August 19, 1995, at 1753 central daylight time, a DeHavilland DHC-1, N320WM, was destroyed as a result of impacting with the ground after performing aerobatic maneuvers. The air transport pilot and one passenger received serious injuries. The 14 CFR Part 91 flight departed a private airstrip near Baldwin, Wisconsin, on a local flight. Visual meteorological conditions prevailed and no flight plan was filed. The pilot reported that he had been doing aerobatic maneuvers for about ten minutes before returning to the private airstrip at Baldwin, Wisconsin. He started decending from 4,000 feet mean sea level at about 140 knots, and after leveling off at about 2,000 feet mean sea level (1,000 feet above ground level) the aircraft started to vibrate or shake violently. He reported pulling the power back to idle thinking he may have lost a propeller blade, and to slow the aircraft down. He had no idea what was wrong with the airplane, and does not remember if the control stick was moving or stuck in one place. He set up for an emergency landing in a cornfield. He reported that it was about one minute from the time the airplane started shaking violenly till the time he landed in the field. The aircraft landed in the cornfield and nosed over. A witness saw about the last one minute of flight of the airplane before it crashed. He reported seeing the airplane at about 1,000 feet above ground level going west in level flight, and heard the engine running smoothly. He reported that the airplane entered a left bank at about forty-five (45) degrees angle of bank. The airplane started to lose altitude, and the witness stated that he thought it was not a coordinated turn. The airplane wings dipped further into the bank to perhaps greater than fifty (50) degrees angle of bank before the airplane rolled out in a wings level descent, heading east. He reported the engine noise getting nearly "nonexistent." He observed the airplane in a steep descent angle of about twenty (20) degrees and he thought the pilot was either buzzing somebody or about to pull up for some aerobatic maneuver. As the airplane neared the horizon he heard the engine again for a second or two and saw the airplane accelerating. He saw the airplane descend below a hill and expected the airplane to crash. An investigation of the aircraft by Federal Aviation Administration officials revealed that the propeller had separated from the engine upon ground contact, and that all three blades were still attached to the prop hub. The propeller displayed evidence of being powered when it contacted the corn and ground. They reported that the airplane appeared to have a large amount of airspeed upon impact. They performed a control continuity check check and found nothing abnormal. In their conversation with the pilot, the pilot stated that he thought the aircraft may have had a structural failure of some sort, but that he could not determine its source during the flight. The experimental aircraft's horizontal stabilizer/elevator was attached to the empennage via a four (4) tube truss assembly. The left and right horizontal stabilizer support tube assemblies had been locally modified by the pilot/owner to be adjustable in length to control the angle of incidence of the horizontal stabilizer. The two adjustable tubes were sent to the Materials Laboratory of the National Transportation Safety Board for examination. The Materials Laboratory reported that optical examinations of both tubes' fracture areas found deformation, yielding and fracture features indicative of overstress separations of both assemblies. No indications of preexisting cracking were noted. The lab reported that the deformation patterns on the right support tube assembly indicated overstressing loads.