NHHC home page Return to Underwater Archaeology
Between 24 May and 2 June 1996, the Naval Historical Center's Underwater Archaeology Branch, the Consejo de Arqueologia Subacuatica de Puerto Rico (Puerto Rico Council of Underwater Archaeology), and the Institute of Nautical Archaeologya undertook a joint project to investigate three unidentified wreck sites in Puerto Rican Waters. All were suspected of being the wrecks of American military craft. As part of this cooperative effort, Rich Wills, one of the Underwater Archaeology Branch's nautical archaeologists who was also a member of Institute of Nautical Archaelology's archaeological staff, was temporarily loaned to the Consejo at the request of Dr. Jeronimo Lynn Hall Jaeger, the Underwater Archaeologist for the Territory of Puerto Rico. In addition to Rich Wills and Dr. Hall, the Consejo's dive team consisted of archaeologists Richard Fontanez Aldea, Felix Cuevas Guzman, and Juan Vera Vega, and Department of Natural Resources marine biologist James Timber. Timber's goal was gathering data on the interaction of wrecks and their environment in their dual capacity as artifical reefs. Completing the field team was underwater archaeologist Jemison Beshears, an authority on antique arms and armor with knowledge of Second World War-era U.S. Army Air Force aircraft and aircrew equipment.
The archaeologists conducted the site assessments with seven general objectives in mind: (1) to locate and record the precise positions of these sites, (2) to visually examine and document the sites, (3) to identify the types of craft represented, if possible, (4) to advise the Consejo regarding how to go about the necessary further archival research required, (5) to identify the wrecks if possible, (6) to assess the sites for any potential presence of unrecovered weapons systems, unexploded ordnance, and/or human remains, and (7) to assess the sites in terms of which one would be the best candidate for nomination as a pilot Department of Natural Resources underwater preserve. In the course of this process, all of the sites were visually documented with still and video imaging equipment.
The Three Wrecks:
The first wreck assessed was a four engine aircraft with a wingspan of approximately 140 feet, located between approximately two and three miles off Aguadilla and resting at a depth of 120 feet. The first dive confirmed that the wreckage was of a Boeing B-29 or early B-29A Superfortress strategic heavy bomber. Only the center fuselage and wing assembly section of the aircraft was recognizable, sitting on an almost even keel on a sand and coral bottom and lying in the center of a large debris field. Both wings were intact along the entire length of their span, and all four engines were hanging on their mounts. The aircraft's four airscrews were located nearby, with the appearance of having been removed and laid face-down on the sea bottom subsequent to the wreck. The gun turrets also appeared to have been removed, perhaps after the crash. Although a search for diagnostic engine manufacturer's serial data plates was performed, none were uncovered.
Neither the exact model nor the identity of the individual aircraft could be determined during the dives. However, the archaeological team considered it a distinct possibility that the craft had been operating out of the nearby airfield at Borinquen, which had formerly been Borinquen Army Air Field during the Second World War, and later Ramey Air Force Base, an important Strategic Air Command installation, during the early Cold War. After the completion of the assessment project, Master Sergeant Mike Stowe, USAF undertook archival research and discovered a likely candidate for identification with the wreck. According to records examined by Stowe, the Superfortress wreck may be Army Air Force B-29 serial number (or "tail number") 42-65287, which like its sister aircraft Enola Gay was one of 204 Martin-built Superfortresses constructed in Omaha, Nebraska.
Although it was lost in Puerto Rican waters, 42-65287 was assigned to the Army's Second Air Force, 17th Bomber Operational Training Wing, 246th Army Air Force Base Unit, which was home-based at Pratt, Kansas; while all personnel aboard at the time of the loss were assigned to the Second Air Force, 346th Bomb Group (Very Heavy), 461st Bomb Squadron (Very Heavy).
On the evening of 31 May 1945, the nearly brand new aircraft lifted off from Borinquen Army Air Field, crossed the coastline, and headed over the ocean to initiate a bombing and gunnery mission as part of Gypsy Sub Task Force THREE. Shortly after takeoff, oil pressure began dropping in engines one and two, and the propeller blades on those engines jammed in their pitch settings and started to "run away," that is, run at high, uncontrollable speeds while providing no thrust. Due to the stuck blade pitch settings and the uncontrollable speeds of the two propellers, the pilot reported "two props running away" to the base control tower, declared an emergency landing, and, with the closest land astern of them, began to turn around. By this time both of the engines were throwing out a substantial amount of smoke. The tower closed the field to all other traffic at this time, and radioed that "287 was cleared to the field." When the ship was about halfway through its starboard turning leg for its new course back to base, the number three engine suddenly lost oil pressure, and that engine's prop jammed in its pitch settings, lost thrust, and jumped to an uncontrollable speed as well. The pilots, 2nd Lieutenant James B. Giacomo and 2nd Lieutenant James A. Barrett, and the flight engineer, Master Sergeant Edward Herrick, attempted to "feather" all three props to neutral pitch settings, but the props would not respond. Due to the loss of the third engine, Giacomo determined that they would not be able to hold sufficient enough altitude to allow them to clear the coastal ridgeline, make the field, and execute an emergency landing. After updating his aircraft's situation to the base control tower, he turned the aircraft back out to sea and into the wind, and prepared the crew of thirteen for an emergency wheels-up ditching at sea. The last interphone communication came from Barrett, who warned the crew to "prepare for immediate ditching." The last message received by the control tower from the Superfortress was "287 ditching."
The Superfortress hit the water flying into the wind and waves with the tail low, wheels up, and full flaps on, flaring out just before impact. Giacomo later related that "we flew the aircraft into the water at 110 M.P.H. and in a landing attitude." The ship met with rough seas, on which six to eight foot waves were running across strong swells. The Superfortress skimmed across the surface for about 400 feet before the nose ploughed into a wave and went under. The impact with the surface of the sea broke the plane's back, shearing it in half at the radar room. The tail section sank in about 45 seconds, while the forward section floated for between three and four minutes, during which time it rode with the waves, and toward the end of which it stood on its nose as it sank. The gunnery instructor on the flight observed that "the ship went under nose first, exposing the entire bottom torn off, and the tanks and bombs intact."
The crew of thirteen, some helping one another and others fighting their way out individually, forced their way through inrushing, shoulder-high water to escape from the rapidly filling and sinking aircraft. Some were knocked unconscious by the impact, and others suffered serious injuries. After escaping from the sinking aircraft, many crewmen had to kick free of debris or disentangle themselves from the loose rigging in which they had become tangled. One airman who became tangled in loose control cables was almost dragged under with a portion of the aircraft. In the water, the crew tried to assist each other as much as possible. Many crewmembers experienced trouble inflating their Mae West flotation vests due to the air chambers in them having been split by the hard impact. Giacomo, perhaps driven by the additional burden of responsibility for the crew charged to his leadership and care, managed to make his way out on the port wing as it was going under, where he successfully removed a five-man life raft from its compartment and inflated it for use by the most injured crewmembers.
Within about ten minutes, Army crash boats from the Eleventh Army Air Force Emergency Rescue Boat Squadron were directed to the crash scene by overhead aircraft. By the time the Army boats arrived, two sections of the bomber were already underwater. All that remained on the surface were survivors, some hanging to the sides of a life raft and some adrift, and, according to the crash boat crews, "a lot of equipment floating in the water...oxygen tanks, cushions, and pieces of metal." Despite the crew's efforts to help one another exit the sinking Superfortress and stay afloat, and the quick response time of the crash boats, three of the crew, the navigator, radar operator, and tail gunner, were lost at sea.
Based upon the similarities in location and physical characteristics of the wreck, the association between the loss of 42-65287 and the Aguadilla Superfortress wreck seems probable. Specifically, the rear portion of the Aguadilla wreck is missing, which supports the historical account of 42-65287 breaking in half. The nose of the wreck is extensively damaged, which may be consistent with the survivors' accounts of the wreck sinking "nose first." Rigging is still visible on the site trailing aft from the wreckage, supporting some survivors' experiences of entanglement. Finally, the fact that the propellers were removed and relocated in a nearby area, and that no serial plates or propeller control assemblies were present on the motors, may indicate a post-wreck investigation, which would appear reasonable in light of the B-29's emerging role near the end of World War II as the United States' premiere strategic bombing platform.
The next research step will be to examine Navy records for any post-crash salvage activities undertaken at the site to determine the cause of the 42-65287 wreck before a similar fate could befall other new Superfortresses. Hopefully, if such records are located, they will also reveal whether any unexploded ordnance was recovered from the wreck. Possibly, future remote sensing activities in the area surrounding the wreck may reveal the rear half of the aircraft.
Located in the waters just off the treacherous rocks of El Desecheo Island, the second wreck lay in approximately 30 feet of water, near an underwater cavern called Tornado Cave. The entire area surrounding the wreck is designated on some maps as a Navy "unexploded ordnance disposal area." The wreck site, while proving difficult to reach, provides a spectacular visual panorama. The vessel appeared to have been of wood hull construction but with steel engine beds for a pair of straight eight-cylinder internal combustion engines. At the end of each propeller shaft was a three-bladed screw. The configuration of the vessel and the character of wreck-related debris provides a possible indication that the vessel may have been some sort of local military patrol craft.
Following the completion of the survey, Naval Historical Center shipwreck database manager Barbara Voulgaris extracted from the computerized database all known Navy wrecks in Puerto Rican waters. From this list, one possible candidate emerged: the U.S. Coast Guard district patrol vessel EM Dow (WYP 353), formerly the Menhaden type fishing vessel Annie Dow, an emergency government wartime acquisition under charter by the Coast Guard (the prefix EM signified "emergency manning"). Dow measured 134 feet 3 inches in length, 21 feet 7 inches in beam, and 10 feet 8 inches in draft, and possessed a gross tonnage of 241 tons. Stationed out of San Juan, Dow was assigned to duty as part of the Caribbean Sea Frontier.
On 14 October 1943, near Mayaguez, approximately one mile south of the Point Higuero Lookout Station, the crew of Dow was forced to abandon their foundering vessel in a strong gale. Fortunately, the crew suffered no casualties, but Dow was driven onto the rocks by the storm. The wreck was reportedly sold to a local salvor on 20 August 1948 for its hull and fittings. Significantly, Point Higuero is located almost directly across the channel from El Desecheo Island. However, this potential identity for the El Desecheo vessel has not yet been confirmed.
Further research is necessary to confirm or eliminate a connection with Dow. Significantly, there is a discrepancy between the number of engines possessed by Dow and the number possessed by the wreck. According to the Office of the Coast Guard Historian, Dow was powered by a single engine and propeller, while the 1996 assessment of the wreck revealed a twin engine, shaft, and screw configuration. Further comparison is necessary in regard to the El Desecheo vessel's overall dimensions, powerplant configuration, and location relative to what is historically recorded for Dow. Additional research is necessary regarding whether the vessel might have been refloated and sunk near El Desecheo. If not Dow, perhaps the wreck is that of another USCG vessel lost under similar circumstances. Still-unresearched archival records may reveal the answers.
The third site is located on Motor Reef, near Turrumote Cay, off La Parguera. The wreckage lies approximately five miles offshore, at a depth of 55 feet. Dives on the immediate site area revealed only a solitary but largely intact radial aircraft engine, with a single loose component lying nearby. The engine is encased in its cowling, and the propeller is attached and largely undamaged. Interestingly, the propeller blades were set fully feathered into the wind (that is, the blade pitch settings were totally neutral, so that the edges of the blades pointed directly into the wind).
An examination of the engine revealed a manufacturer's serial data plate which, although completely encrusted, was successfully removed from the block. Cleaning of the plate at the Naval Historical Center Underwater Archeaology Branch's conservation facility revealed the engine to be a Pratt & Whitney R-1830-92. This number is significant because the "-92" suffix indicates that the engine was intended for use with a U.S. Navy aircraft. The only two Navy aircraft types which used this model of engine were the R4D Skytrain cargo aircraft and later models of the the PBY Catalina flying boat. Interviews conducted by the Consejo's Fontanez with local fishermen revealed one who remembered a flying boat landing with an engine afire sometime in the 1940s after which the fisherman rescued the crew of nine. Preliminary archival research has failed to substantiate this story. Another possibility is that this engine may have come from an Army OA-10, the Army's equivalent to the PBY, since some OA-10s may have reached the Army through Navy channels. Preliminary research in naval aviation accident records revealed one PBY-6A which disappeared with all hands during a passenger flight from Naval Air Station Roosevelt Roads, Puerto Rico to Naval Air Station Key West, Florida. However, due to differing locational factors, it is unlikely these two aircraft match. Further archival research is necessary to investigate R4D and OA-10 losses in the Caribbean theater during the Second World War.