Pursued from above, U-165 begins to crash dive, September 1942.

PHOTO: DEPARTMENT OF NATIONAL DEFENCE—PL12814

By the time Paul Hartwig and Eberhard Hoffmann in U-517 and U-165 began their rampage in the lower St. Lawrence River, the navy and the government had decided to close—as soon as possible—the river and gulf to oceanic shipping.

The loss of His Majesty’s Canadian ships Raccoon, Charlottetown and several other ships in early September confirmed the wisdom of that decision. Mackenzie King and Vice-Admiral Percy Nelles might have felt even more at ease with that decision if they had known how little luck Canadian forces would have had that month in sinking the intruders. In truth, the Canadians deserved better, but luck was on the U-boats’ side. The escape of Hartwig and U-517 from a watery grave was sheer chance: twice in September he and his sub were ‘claimed’ and in one instance certain death was avoided by a few feet of water.

The decision to close the St. Lawrence was confirmed by King’s government on Sept. 9, two days before the sinking of Charlottetown in daylight off Cap-de-la-Madeleine. The ostensible reason for doing so was an agreement to commit more Canadian escorts to support the Allies. In August, the Allies asked if the Royal Canadian Navy could spare escorts for the upcoming invasion of North Africa, Operation Torch, which was slated for November. By paring and scrapping, the navy found 17 corvettes (16 Canadian and one British) it could spare from its operational forces. Among them were several of the latest corvette types, with improved hull forms and bridges. The British undertook to help refit them with much heavier secondary armament—20-mm Oerlikon guns—which would be essential for air defence in the eastern Atlantic and Mediterranean. At a time when corvettes operating on the North Atlantic Run, Western Local Escort Force and St. Lawrence were still fitted primarily with .303 Lewis guns or .50-calibre Brownings, assigning Canada’s limited supply of powerful 20-mm cannons to Operation Torch was a selfless act. So, too, was sending 17 escorts, including some of the navy’s latest.

The work to prepare the Torch corvettes was well underway by the time the government formalized the transfer in early September. King was reluctant to see them go. “[T]he few that we could send there,” he wrote in his diary, “might be relatively unimportant for the purpose for which they were needed compared to the purpose which they would be serving while here…” In view of what happened over the next few weeks, it’s hard to fault King’s logic. Reflecting on the recent sinkings around the gulf, King opined that, “Letting corvettes go means routing freight by the river will have to stop for this season.”

As King well knew, the RCN was content with the prospect of closing the St. Lawrence. By September 1942 the crucial work had been done: most of next year’s oil had been delivered. The rest could come by pipeline from Portland, Maine. The navy and the government were also under pressure from the British Ministry of War Transport and from the Admiralty to close the St. Lawrence to oceanic shipping. The ports of Montreal, Trois-Rivières and Quebec were important to Canada, but the British felt no sentimental attachment to them. Moreover, from Montreal to the convoy assemble port of Halifax was 1,600 kilometres: a third the distance across the North Atlantic. Taking ships to Canada’s inland ports was, from the British perspective, a waste of time and shipping capacity—and escorts. They preferred, and the Canadian government and navy agreed—reluctantly and without public fanfare which would have been politically explosive—to route Canadian trade by rail to east coast ports and handle it there. As shipping dwindled sharply through 1942 the urgency to do this increased.

Then there was the problem of defending shipping in the St. Lawrence. It was not that the navy was poorly trained or ill equipped, although it was more than a little of both. But the complex hydrography of the river and the gulf could not be solved by technology, nor could the canalizing effect of the geography which forced convoys into predictable and easily intercepted tracks. The open area of the gulf allowed for some evasive routing, especially routing the convoys between Prince Edward Island and Iles-de-la-Madeleine where shallow waters virtually precluded submarine operations. But the western gulf and lower St. Lawrence River are deep, and room for manoeuvre constricted. The best way was to eliminate the targets. This would have the bonus effect of easing pressure on the government and the navy to do “something”—like kill U-boats.

His Majesty’s Ship Salisbury.

PHOTO: LIBRARY AND ARCHIVES CANADA—E01085922

The inability of the navy and air force to sink U-boats during the 1942 campaign was not for want of trying: in the end, the U-boats drew heavily on their stores of luck. HMCS Drummondville came within an ace of sinking U-132 in July and HMCS Weyburn was right on top of U-517 in August and “never got a sniff.” Air attacks, too, had come close. As the navy and government agreed to close the river, events bore out all these trends: the inability of Canadian sailors and airmen (and American flyers, too) to sink U-boats; the enduring problems of keeping the daring submariners away from the shipping; and the incredible luck of the U-boats.

Evidence of the wisdom of the Sept. 9 decision appeared almost at once. After sinking HMCS Charlottetown on Sept. 11, Hartwig and Hoffmann remained in the lower river hunting convoys. By Sept. 14, Hartwig was off Cap-des-Rosiers ready to intercept SQ-36, the largest St. Lawrence convoy of the season with 22 ships. It also had the largest escort yet. This included the British destroyer HMS Salisbury (an ex-American four stacker), and the Canadian corvette Arrowhead, both with modern radar. The new RCN official history states that a second corvette was also part of the screen but does not identify it. Other vessels rounded out the escort.

Air support for SQ-36 from the Royal Canadian Air Force was direct and continuous. In fact, by early September 1942 the air force had heavily reinforced the western Gulf of St. Lawrence and the lower river. The detachment of 119 Squadron Hudson bombers at Mont-Joli, Que., was joined by more from 11 Sqdn., bringing the total to six, and a special “striking force” of three Hudsons from the Yarmouth-based 113 Sqdn. were deployed to Chatham, N.B. These New Brunswick-based aircraft were among the first of the RCAF’s Eastern Air Command to be painted in the new all-white scheme adopted by the British, and to adopt the tactic of patrolling higher—at 5,000 feet—rather than the 1,500 feet normally used by maritime patrol aircraft.

The paint scheme and higher altitude made them harder to detect. It was this combination that allowed Pilot Officer R.S. Keetley, operating from Chatham, to surprise and attack Hoffmann on the surface south of Anticosti Island Sept. 9. His aircraft came in too high and too late; clearly the drills had not yet been fully assimilated. However, even had he placed his charges alongside U-165, it is likely that Hoffmann would have escaped—the RCAF had not yet adopted the necessary depth charge pistols to attack a U-boat on the surface—that too, would come.

On Sept. 15, Hartwig used the eastward flow of the river and the forward motion of convoy SQ-36 to drift—submerged and totally silent—down on his targets. The RCAF Catalina Canso of 117 Sqdn., out of Sydney, patrolling ahead of SQ-36 never saw him, nor did the escort screen as it passed over U-517 while Hartwig watched through his periscope (since he was travelling in the water mass of the river, his periscope would not make the telltale ‘feather’ of spray that often gave subs away in daylight). As Michael Hadley wrote, the approach required an exceptional degree of professionalism and steady nerves, especially on a bright summer afternoon. At one point Salisbury slipped past just 400 metres away and then turned back, searching to within 150 metres of U-517 before turning away again. When asked about this many years later Hartwig observed, “one develops a sixth sense, like a tiger in the jungle.”

Once Salisbury had cleared off, Hartwig opened his torpedo doors and fired a spread of four “eels” at two overlapping targets in the second column. Both ships were struck and Hartwig confirmed the strikes before gunfire at his periscope from nearby merchant ships forced U-517 down. The first ship hit was the 2,741-ton Dutch steamer Saturnus, travelling in ballast. The second, the 2,166-ton Norwegian Inger Elisabeth, was carrying coal. Four men died in the two vessels, both of which sank in about 15 minutes. Searches and counterattacks by the escort proved fruitless. The British captain of Salisbury later remarked on how poor the sonar conditions were: this was not news to his Canadian colleagues.

A Lockheed Hudson on patrol.

PHOTO: SHEARWATER AVIATION MUSEUM

Hartwig, meanwhile, kept Hoffmann informed of SQ-36’s progress, and the latter was ready when the convoy came into sight off Cap-Chat, Que., in the early hours of Sept. 16. Hoffmann, too, launched a daylight submerged attack, relying on his periscope to guide him into firing position. He got away with it, but was very lucky. Ships sighted U-165’s periscope and opened fire before Hoffmann completed his firing run. Regardless, he got off a spread of torpedoes that hit two ships: the 6,624-ton British steamer Essex Lance was later salvaged and repaired, but the 3,667-ton Greek ship Joannis sank in 10 minutes. Eighty survivors were brought ashore. The response of the escort was not sufficient for Hoffmann to break contact with the convoy, and he returned to hit the American ship Pan York, which was also salvaged. With all his torpedoes expended, Hoffmann headed home, but never made it: U-165 was sunk by a mine in the Bay of Biscay.

Hartwig, however, was not done. He headed northeast towards Newfoundland to get out of the traffic area and shift torpedoes stored outside his hull into his forward torpedo room. By midday on Sept. 21 he was back off Cap-des-Rosier, following SQ-38. Hartwig thought he was safely submerged, but the fickle water conditions left a portion of his conning tower exposed and HMCS Georgian (joining SQ-38 from astern) surprised him. Hartwig later admitted having trouble maintaining a steady depth that day: “Those water layers!” he later recalled. Georgian was about to ram U-517 when it crashed dived. The Bangor-class escort then stood off and conducted a deliberate depth charge attack, and was turning to regain contact when U-517 suddenly breeched the surface, rolled onto its side and sank.

Acting Lieutenant-Commander A.G. Stanley, Royal Canadian Naval Reserve, had every reason to be pleased, and assumed he sunk the U-boat. He kept Georgian at it for the rest of the day, until all his depth charges were expended. The staff in Halifax agreed with his assessment: U-boat sunk. But Hartwig got away—again. He moved north, to clear the area, repair the damage and resume hunting.

By mid-September, the RCAF had committed even more aircraft to the gulf, and Hartwig’s heydays were over. By the time he returned to action, 15 Hudsons and 20 Cansos, including the rest of 113 Sqdn. from Yarmouth, N.S., were in the area. The RCAF also adopted a practice of providing air cover at night, making it especially nerve-racking for Hartwig, and those who came later. In many ways, Hartwig was now the hunted. In one 24-hour period 113 Sqdn. Hudsons from Chatham caught him on the surface seven times, including at night, and attacked him three times. U-517 was shaken and exceptionally lucky. In one attack three of the four depth charges failed to release, the one that did shook U-517 badly: four that close would have sunk it. After another attack U-517 resurfaced to find a depth charge lodged in its deck. Once rolled into the sea the charge exploded at its set depth: fortuitously, Hartwig had kept his sub shallow following the attack. His luck held several days later when another Hudson from Chatham, flown by Flying Officer M.J. Belanger, bracketed U-517 with four well-placed depth charges. “The charges were seen to explode all around the hull,” Belanger reported, “slightly ahead of the conning tower. One large explosion occurred around the hull…” U-517’s motion was stopped, her bow came out of the water and then the sub settled by the stern. The initial assessment was that U-517 had been sunk—again.

Hartwig persisted in hunting convoys in the lower river until early October, joined by U-69, but neither had any luck. U-517 headed home on Oct. 6, ending the exploits of the most skilful—and lucky—submariner ever to attack shipping in Canadian waters. The navy and the RCAF had done well. They had nearly killed him several times and—just as importantly—suppressed him following the attack on SQ-36. Hartwig had also been lucky to escape with his life. Unfortunately, for the Canadians, the 1942 inshore campaign did not end with his departure.

Email the writer at: writer@legionmagazine.com

Email a letter to the editor at: letters@legionmagazine.com