On September 15, 1993, approximately 1400 hours Pacific daylight time (PDT), a Beech G35 "V" tail Bonanza, N4641D, registered to J. B. Phillips, and being flown by Robert L. Woolverton, a certificated instrument rated private pilot, sustained substantial damage when the left stabilizer main spar separated along a fracture line through several spar attach bolts within the empennage area. The pilot and passenger were uninjured, and the aircraft, which was entering the landing pattern at the Arlington Municipal Airport, Arlington, Washington, at the time of the occurrence, was successfully landed without further damage. Visual meteorological conditions prevailed at the time and no flight plan had been filed. The flight, which was personal in nature, was to have been operated in accordance with the requirements set forth in 14CFR91 and originated at Arlington. The Safety Board was first notified of the occurrence by the insurance adjustor on October 8, 1993. The pilot reported to a Beech representative that while descending into the downwind leg of the traffic pattern at about 170 mph he felt a "bump" similar to what one would expect when encountering their own wake turbulence after a 360 degree constant altitude turn. Immediately thereafter, the airplane began to vibrate with the control column oscillating fore and aft. The pilot reduced power to idle, increased propeller control to high RPM and began to decelerate the aircraft. The vibration ceased soon after and the aircraft was landed without difficulty. Inspection of the aircraft by the Beech representative revealed damage to the aft fuselage and empennage (refer to attached letter). Additionally, the left stabilizer spar was found to be fractured through the upper and lower outboard attach bolt holes. The balance of both the left and right ruddervators was found to be within acceptable limits. Additionally, Beech Service Bulletin 2188 had been applied to the aircraft several years previous to the occurrence. The two sections of spar were sent to the Board's Materials Laboratory Division for examination. There was no evidence of any preexisting fracture and no bending deformation, and the fracture surface was described as "typical of an overstress separation" (refer to attached Metallurgical report).