On June 9, 1995, about 0732 eastern daylight time, a Piper PA-23-250, N949BW, registered to the pilot, operating as a 14 CFR Part 91 personal flight, experienced a reported partial loss of engine power of both engines on initial takeoff climb, and made a forced landing to an open field in the vicinity of the Hidden River Airport, Sarasota, Florida. Visual meteorological conditions prevailed and a visual rules flight plan was filed. The airplane sustained substantial damage. The private pilot and two passengers reported no injuries. The flight originated about 2 minutes before the accident. The pilot stated he purchased fuel in Arcadia, Florida, on June 7, 1995, and flew back to Sarasota, Florida, on the same day. He pulled the airplane out of his hangar the next morning, checked the oil, and filled the two outboard fuel tanks with fuel from two 5 gallon cans. The fuel sumps were drained and water was present. He continued to drain the sumps until no water was visible. A walk-around inspection was completed and the airplane was pushed back in the hangar. On June 9, 1995, the airplane was pushed outside the hangar, and a walk-around inspection was completed to include draining the fuel sumps. No water was found and everyone boarded the airplane. The engines were started, the airplane was taxied to the runup area, and an engine run-up was completed with no deficiencies noted. They departed a short later and became airborne just under 70 knots using about 2/3 of the runway. He maintained the airplane in ground effect at about 3 to 5 feet while accelerating almost to 80 knots. As he was reaching for the landing gear handle the left engine experienced a reduction in power, and the airplane started to settle but did not touch the runway. The left engine seemed to experience an increase in engine power and right rudder was required to keep the airplane straight and level. The left engine experienced another reduction in power and the airplane drifted to the left. He pulled back on the control yoke in order for the airplane to clear a fence, ditch, trees, and allowed the airplane to settle on the ground. On landing rollout the main gear collapsed, and the airplane collided with trees before coming to a stop. The Vice President for Air-Cadia Inc., Arcadia, Florida, confirmed the pilot purchased fuel on June 7, 1995. He further stated the refueling system is inspected daily, and fuel samples taken before and after the accident did not indicate any form of contamination. Recovery personnel and a party representative to the NTSB investigation stated the pilot informed them that he found water while sumping the fuel sumps during the preflight inspection of the airplane on the day of the accident, and he drained the sumps until the fuel was clear. Examination of the airplane was conducted by the FAA and parties to the NTSB investigation on June 16, 1995, at 3711 Hidden River Road, Sarasota, Florida. Examination of the left engine revealed no evidence of a precrash mechanical failure or malfunction. The propeller blades were bent aft with evidence of rotation. The top sparkplugs were removed. Their electrodes, color, wear, and deposits were normal when compared with the manufacturer's chart, with the exception of the No. 2, No. 3, and No. 6 cylinder sparkplugs and piston domes. They appeared white in color. All other piston domes on the left engine appeared light brown in color. Fuel samples were taken from the left wing inboard and outboard nacelle tank drains. Water and grit were present. Water was present in the gascolator sump/bowl, internal baffle assembly, and congealed contaminates was present on the fuel screen. (For additional information see FAA Aviation Safety Inspectors Statement, and Field Notes, Textron Lycoming). Examination of the right engine revealed no evidence of a precrash failure or malfunction. The propeller blades were bent forward and torsional twisting was present. The top sparkplugs were removed. Their electrodes, color, wear, and deposits were normal when compared with the manufacturer's charts, with the exception of the No. 3 cylinder sparkplug and piston dome. They both appeared white in color. All other piston domes on the right engine appeared medium brown in color. Fuel samples were not taken. (For additional information see FAA Aviation Safety Inspectors Statement, and Field Notes, Textron Lycoming).