Admiralty approves the convoy system

On 14 June 1917 the First Sea Lord, Admiral John Jellicoe, formally approved the report of the Convoy Committee, which the Admiralty had set up in the middle of May. The report, submitted on 6 June, provided for regular convoys (PL) of ships bound for Britain from four assembly points: Gibraltar, Dakar, Hampton Roads in Virginia and New York. Outward-bound convoys were to assemble at Milford Haven, Lamlash on the Scottish island of Arran and Falmouth. Across the ocean the ships were to be escorted by cruisers, and 300–400 miles from the coast destroyers were to take them over. The plan covered not only British ships but also Allied and neutral ones bound for British and French ports.

Trials and hesitation

The decision was preceded by two successful experiments. The first trial convoy from Gibraltar arrived without loss in May. The second, which left Hampton Roads on 24 May under the escort of the cruiser Roxburgh, arrived around 8 June, two days after the report was submitted. The convoy itself did not lose a single ship. Two steamers that could not keep up with the rest, however, dropped out on the way, and one of them was torpedoed. The captain of Roxburgh, Captain Whitehead, judged that the ships had kept station well and that a convoy could escort many more of them. On 8 June Jellicoe reported to the War Cabinet on the success of both trials and announced weekly convoys of tankers and food ships.

The Admiralty still believed, however, that for lack of escort vessels it could not protect the greater part of the traffic across the Atlantic. Lloyd George later wrote, citing a report by the Shipping Controller, that the consent of 14 June had been given “in the letter, not in the spirit”, because the necessary forces had not been allotted. According to the prime minister, it was only his firm intervention that induced the admirals to “put their own decision into effect”. Naval historians treat this account with caution, as they do his description of his earlier visit to the Admiralty. What is certain is that ships really were lacking. According to Lloyd George, of the 279 destroyers in British waters only 20–30 were initially allotted to convoys. About a hundred of those destroyers belonged to the Grand Fleet.

Organisation

The work of the system was directed by the Assistant Chief of the Naval Staff, Rear-Admiral Alexander Duff, and his assistant was Commander Reginald Henderson. It was he who in the spring had shown that far fewer ocean-going ships called at British ports than had been thought. On 25 June Fleet Paymaster H. W. E. Manisty, a member of the committee, was appointed organising manager of convoys. Lloyd George recalled that he was given neither a room nor staff and had to obtain them himself. Gradually, however, he built up a convoy section that drew up sailing schedules, allotted escorts and laid down routes. In the Ministry of Shipping a similar section was set up under Norman Leslie. It gathered information on where ships were, where they were going, how fast and with what cargo.

The need was urgent. In May the Germans sank less than in the disastrous April, but British shipping alone lost about 350,000 tons. In June losses rose again. According to the British official history they reached almost 420,000 tons, and since the start of unrestricted submarine warfare British ships had already lost nearly 2 million tons.

By the end of June four more trial convoys had left Hampton Roads. On 2 July a regular series of convoys began from there; on 10 July the first convoy left Sydney on the Canadian island of Cape Breton, which the report had not provided for at all; on 14 July from New York; on 26 July from Gibraltar; and on 11 August from Dakar. The convoys sailed at 8 knots, and those from Gibraltar at 7 knots. Ships faster than 12 knots and slower than 8 knots continued to sail independently. Outward-bound convoys from Britain did not begin until August. In the following months it turned out that ships sailing in convoy were lost far less often than those sailing alone. Overall losses, however, began to fall only when the system covered the greater part of shipping.