On September 29, 1998, about 0705 central daylight time, a Bell 47D1, N755E, listed as "registration pending", collided with trees during a forced landing near Oxford, Mississippi. Visual meteorological conditions prevailed at the time and no flight plan was filed for the 14 CFR Part 137 aerial application flight. The helicopter was substantially damaged and the commercial-rated pilot, the sole occupant, sustained minor injuries. The flight originated about 1 minute earlier from the University-Oxford Airport, Oxford, Mississippi. The pilot stated that he flew the helicopter for 8.1 hours the day before with no discrepancies noted. The morning of the accident, he performed a preflight to the helicopter which included greasing the tail rotor drive shaft bearings. He then performed a second preflight to the helicopter with no discrepancies noted. The flight departed with full fuel tanks and 1.5 gallons of chemical and while flying at 400-500 feet agl at an unknown airspeed, he felt what he thought was a bird strike. The flight continued and he felt a short minor vibration in the cyclic. He turned to return to the airport for a precautionary landing and during the turn, the helicopter began shaking and vibrating, and the cyclic required both his hands to control. He attained level flight and cleared a house that was ahead, then lowered the collective which "...allowed me enough time to regain some control..." then with school buses ahead, he maneuvered the helicopter towards trees that were located at the edge of a clearing. He intentionally maneuvered the helicopter towards the base of the trees to avoid colliding with the school bus. After the collision with the trees and coming to rest, he exited the helicopter. Additionally, more than 6 months after the date of the accident, the pilot first mentioned the fact that while performing a preflight to the helicopter some weeks before the accident date, he noted a crack in the laminate on the top of the "blue" main rotor blade. He contacted a Mr. Bob Pritchett who holds an airframe and powerplant certificate. The pilot who does not have an airframe or powerplant certificate was told by Mr. Pritchett to purchase a kind of glue that required to be mixed and to apply the mixed glue to the crack to prevent water from getting into the crack. The pilot complied with the instructions but the crack appeared the next day. The pilot again contacted the mechanic and also contacted the owner of the helicopter and after consulting both, knowing that the crack was only in the laminate, continued to fly the helicopter. The helicopter was then flown to Tennessee for spraying then flown to Holly Springs, Mississippi. While in Holly Springs, during a preflight to the helicopter, the accident pilot noted that the crack now extended completely through to the bottom of the blade; about 3 feet from the blade tip. The crack started at the trailing edge of the blade and extended chordwise approximately 1.25 inches then extended at a 45 degree angle inboard for about .75 inch to 1.0 inch. He noted that the crack was into the balsa wood and he was able to move both of the fracture surfaces opposite one another. He contacted the owner, advised him of the crack and he refused to fly the helicopter. He then contacted representatives from Bell Helicopter but wasn't sure what state or country he called. A representative of Bell Helicopter advised that the helicopter was unsafe to fly. The pilot advised the owner that he had contacted Bell Helicopter and was temporarily fired by the owner. The pilot further stated that Mr. Bob Pritchett came to Holly Springs, repaired the blade, tracked it, and there was no more problem with the main rotor blade. According to Mr. James O. Hopper, holder of an airframe and powerplant certificate, he was asked by the accident pilot to inspect the crack in the blade. The mechanic reported that he couldn't recall the registration number of the helicopter, but did observe a crack at the trailing edge of the blade extending chordwise then inboard. He noted water in the crack. Due to his limited helicopter experience, he suggested that the pilot contact someone with recent helicopter experience. The mechanic reported that the accident pilot contacted someone from Bell Helicopter and the mechanic described the crack to the technical representative (tech rep). The tech rep stated that the helicopter was safe to fly to its base of operations and recommended the blade be repaired before the helicopter was placed back in service. According to Mr. Mike Fisher, the chief pilot for the company, he believed that the helicopter was purchased near June 1, 1998. He recalled a crack in one of the main rotor blades at that time and numerous phone calls were made to operators, mechanics, blade shops, and the consensus was to inspect the blade daily to see if the crack became worse or if it went laterally. The crack was approximately 4-6 feet inboard from the blade tip on one of the blades on the bottom of the blade running chordwise, and also slightly on the upper surface at the trailing edge. He checked for opposite movement on both sides of the crack and deemed the blade safe and felt safe flying the helicopter. He further stated that the accident pilot was concerned about the crack in the main rotor blade and Mr. Bob Pritchett was requested to inspect, repair, and track the main rotor blades. Mr. Bob Pritchett stated that on or about August 28, 1998, he went to Holly Springs, Mississippi, to inspect the main rotor blades on the accident helicopter. He reported purchasing a two-part epoxy resin at Rotor Blades, Inc., which is a main rotor blade repair shop. On examination of the main rotor blades, he observed a "small scratch like tear in one Main Rotor Blade approximately 2 inches long and approximately 4 feet from the Main Rotor Blade tip. The scratch like tear was through the cloth." He inspected the blade in the scratch area using a 14 power magnifying glass and noted that the wood around the scratch was not damaged. He applied the epoxy resin to the scratch area approximately one inch around the scratch, and approximately .020 inch thick. He considered the repair temporary and balanced the blades; after which, the accident pilot test flew the helicopter. Examination of the accident site by an FAA inspector revealed that the helicopter came to rest inverted with the tail boom elevated at about a 45-degree angle. The tail boom was in place but was observed to be displaced to the left. Damage to a chain link fence was noted and damage to tree limbs and tree truck about 30 feet agl was noted. The "yellow" main rotor blade container, ballast; was found near the wreckage. The "blue" main rotor blade container, ballast; was not located. The helicopter was recovered for further examination. Examination of the helicopter by a representative of Bell Helicopter Textron, Inc., and a FAA airworthiness inspector revealed that both main rotor blades were fractured at the root end and the steel weight retention straps for both blades was distorted. The leading edges of both blades exhibited evidence of impact damage. The "blue" main rotor blade clevis, ballast retention; was pinned to the retention strap but the AN4-16 bolt was noted to be failed. Both stabilizer bars were installed but were bent approximately 90 degrees. Cyclic control continuity was confirmed from the cockpit control to the swashplate assembly. Collective control continuity was confirmed from the cockpit control to the bellcrank, collective pitch control; which was observed to be failed. Visual examination of the fracture surface revealed no evidence of preimpact failure or malfunction. The collective sleeve was moved manually and no discrepancies were noted. The tail boom aft of station 200 was separated. Examination of the tail rotor assembly, tail rotor drive shaft, and transmission revealed no evidence or preimpact failure or malfunction. The failed "blue" main rotor blade bolt with clevis assembly was retained for metallurgical examination. Examination of the failed AN4-16 bolt with clevis from the "blue" main rotor blade by the NTSB Metallurgical Laboratory located in Washington, D.C., revealed that it failed in shear. No evidence of preexisting cracks or corrosion was noted. The bolt was also examined by a metallurgist from Bell Helicopter who reported the same findings. A copy of the report from Bell Helicopter is an attachment to this report. The "blue" main rotor blade sections as identified by the epoxy material placed by the mechanic on the upper skin surface was recovered and submitted to the Engineering Laboratories of Bell Helicopter for further examination. The submitted fractured sections were between Blade Stations 144, where the blade was saw cut, and Blade Station 183, where the blade was fractured. A chordwise fracture of the 3 spanwise oriented spruce or pine wood pieces was noted between Blade Station 157-159; the entire leading edge segment was missing between the submitted segments. The balsa wood portion of the blade body which is by design located aft of the three spanwise spruce or pine wood segments was missing entirely between Blade Stations 159-164. Epoxy material was observed on the top skin at Blade Station 159.8, and improper repairs to the trailing edge of the blade were noted in 3 locations. The upper skin was noted to be separated from the balsa wood oriented chordwise at Blade Station 161.0 and the bottom skin was noted to be separated from the balsa wood oriented chordwise at Blade Station 161.8. Wood fibers were noted adhering to the skins that had separated from the balsa wood section. Examination of the fracture surfaces revealed weathering post accident which precluded determination of any preexisting failure or malfunction. Review of the manufacturers Maintenance and Overhaul Instructions manual revealed that pertaining to Section 2-33, Titled Repairing Fiberglas By Insertion, "Any damage to the fiberglas cloth, including all repairs of a temporary nature accomplished by patching, will be repaired by insertion as follows:" There is no mention that epoxy type material can be applied to the surfaces of the fiberglass cloth that is damaged. The helicopter minus the retained components was released to Mr. Harry Brooks representing Universal Loss Management on June 8, 1999. The retained components were also released to Mr. Harry Brooks on July 23, 1999.