A gap in the top decking between the cockpit and the tail fin accommodated an inverted lifeboat. Test pilot Harry Hawker also insisted on a jettisonable undercarriage to save weight and drag and reduce fuel usage, the fuselage underside being reinforced with wooden skids for landing on grass.

Built in Sopwith’s factory in Kingston upon Thames, by 10th of February 1919 the Sopwith Atlantic was on engine tests at Brooklands aerodrome. The aircraft had fat tyres for rough field operations and a small retractable windmill driven electrical generator on the side of the fuselage for the experimental Marconi radio and radio direction finder. Test flights started on 21st of February, working up to a long practice flight.
    Meanwhile in America, Swedish pioneer aviator Hugo Sundstedt had this ‘Sunrise’ twin Liberty engined floatplane built but after just one mile on its first flight it suffered an engine failure. The US Navy was determined to be first to fly the Atlantic. However, the first of their huge US Navy designed and Curtiss built NC flying boats failed to take off with the required fuel load and they needed to fit a fourth engine, like the other three Ncs under construction.
    By early March Harry Hawker and his navigator, Kenneth Mackenzie-Grieve, had completed a 900 mile 9 hour endurance test flight and on the 18th of March left Liverpool for Newfoundland on the SS Digby with the aircraft and a small team including a Rolls-Royce engine mechanic.
    Sopwith’s Brooklands rivals, Martinsyde, had built the Raymor, named after pilot Fred Raynham and navigator Fax Morgan. It was similar to the Sopwith Atlantic but without the jettisonable undercarriage and lifeboat. Formal entries for the race so far included Whitehead Aircraft of Richmond on Thames, the Aero Club of America with an Italian Caproni flying boat and Handley Page with a four engined V/1500 bomber, whilst more secretly also preparing machines were Fairey Aviation, Short Brothers and Vickers. On 27th of March, after a successful 10 hour non-stop flight back and forth between Brooklands and Southampton, the Martinsyde Raymor was tied down in a hangar with the crew aboard for a 24-hour simulated Atlantic flight, after which the engine was found to be in excellent condition.
    The next day the Sopwith team arrived off Newfoundland to find that St. John's Harbour was still ice-bound so they had to go round to Placentia Bay where the two large crates and many other smaller crates had to be trans-shipped to a smaller steamer capable of getting into Placentia harbour, and then onto railway wagons at the quayside for an uncomfortable 60 mile narrow gauge train journey to St. Johns. All this took two extra days so Harry’s plan to build and test the machine in two weeks, then making the Atlantic flight around full moon in mid April, was already looking very optimistic. Despite being on the same latitude as northern France, Newfoundland is sub-Arctic, not warmed by the Gulf Stream, so there was still snow on the ground, and the terrain was much more rugged than Harry had expected. From the station the crated aircraft, mounted on timber platforms, was slowly drawn 4 miles uphill by five pairs of horses through deep mud and slush to the team’s chosen field at Mount Pearl. Harry’s concerns increased when he saw this best available field. The ground was sodden with melting snow and 60 local labourers were already employed attempting to fill the worst soft patches. The field was L-shaped around a steep-sided hill, the east-west leg being just 400 yards long and the sloping, north-south leg only 200 yards.
    On the 2nd of April the Handley Page V/1500, modified with an extra 2,000 gallon fuel tank and floatation bags in the fuselage, started extensive flight trials at Cricklewood, north London. On the 3rd April Sunstedt’s American built floatplane is tested by a Russian pilot but stalls, spins into the sea and is wrecked.
    It took a week to assemble the Atlantic in the substantial wooden hangar prepared for the Sopwith team whilst snow and rainstorms continued outside. One morning, with the mud-bath airfield hardened by frost, Harry managed a successful, lightly loaded 70 minute flight despite damaging the rudder on take-off, but sank into the softening ground on landing.
    Unknown to the Sopwith team, rivals Short Brothers, were close to stealing a march on them. The Short Shamrock made its first flight on the 8th of April and was nearly ready for an east-west flight from Ireland. A development of the Short Shirl torpedo bomber with its range dramatically extended by a 435 gallon external fuel tank suspended in place of the torpedo. Other competition is also catching up. Three days later Fred Raynham and the Martinsyde team arrived in Newfoundland and secured the use of a narrow strip of land alongside a lake at Quidi Vidi, just outside St John's. On 13th of April a Vickers Vimy bomber with two Rolls-Royce engines made a first test flight at Brooklands. Just demobbed from the Royal Naval Air Service, Cpt Jack Alcock had personally convinced the sceptical Vickers management that they should compete. They were impressed by his experience flying a Handley Page bomber over the Aegean and by the thought he had given to making a successful Trans-Atlantic flight whilst a prisoner of war in Turkey. He had personally been overseeing the building of this, the13th Vickers Vimy.
    Waiting for better conditions, Hawker and Mackenzie Grieve tested their ditching survival kit by removing the three ply lifeboat from the aircraft and taking it to a lake. The bow contained a sea anchor, emergency rations, paddles and smoke flares plus parachute flares for night signalling. Inflated, the stern floatation bag also supported the collapsible canvas top decking. In their American Navy pattern immersion suits they replaced the very heavy Kapok insulation with air bags which could be inflated if needed by mouth.
    On 16th of April, in a lull in the storms, the Martinsyde Raymor was dragged across the road from its tent hangar to their narrow airstrip on the water's edge at Quidi Vidi. Witnessed by a large crowd the lightly loaded machine made a 3-hour test flight and a perfect landing. Both the Martinside and Sopwith teams were staying at the Cochrane House Hotel watching each other in a cat and mouse game. The following day Raynham made another short trial flight whilst the Sopwith Atlantic was loaded with its 350 gallons of fuel, and oil, ready to get away, as always planned, around 4 p.m. local time. It would then be navigated by the stars through the night and reach Brooklands at 4 p.m. the next day. However, weather conditions deteriorated and another day was lost.
    Meanwhile, at Eastchurch, the Short Shamrock took off to fly to Ireland in preparation for its non-stop east-west flight across the Atlantic. Flying direct over England and North Wales, it was 12 miles out to sea off Anglesey when the engine stopped and could not be restarted. Gliding back towards land it ditched a mile off-shore. The crew was rescued and the aircraft floated but was seriously damaged by 22 hours in the sea. The problem was found to be an airlock in the fuel transfer system. Next to be tested was the Rolls-Royce Eagle-engined, Fairy Threelantic, developed from the Fairey 3C. It was the only remaining floatplane entry and had the luxury of covered cockpits.
    On 20th of April, and then again on the 22nd, both teams at St John’s readied their aircraft until forecast storms in mid-Atlantic aborted their departures. During April there had been two days completely fogbound, many days with gales, 17 of them with rain and 5 with snow. The airfields were still very muddy. At long last old friends and rivals, Harry Hawker and Fred Raynham, shook hands after agreeing to give each other two hours notice of their intention to leave to avoid the unnecessary risks of a rushed departure. Although the best time in May should have been the 12th or the 19th, with a full moon to aid visibility and navigation through the night, both teams would now prepare their aircraft each day for an earlier weather window, whilst their navigators honed their wireless skills. It was not just about weather conditions for the take off; from a patchy pattern of ship’s sea level weather reports, they could only best-guess the likely weather conditions high above the Atlantic.

To be continued