HM Submarine Triumph's last patrol - December 1941

HM Submarine Triumph's last patrol - December 1941

On Boxing Day 1941 the T class submarine Triumph quietly left Alexandria on a patrol into the Aegean Sea.  With a ships company of 56, Triumph had spent a year in the Mediterranean, where she had mixed a useful clutch of sinkings with a large number of raiding party landings.  By November 1941 Triumph had received orders to head for home.  Her ship's company would have been busy ashore buying presents and luxuries to take home with them.  Her veteran CO, Sammy Woods, had already been relieved and replaced by Tommy Huddart, who had spent much of the war acting as "spare CO" for a miscellany of submarines.  Triumph was just about to head home when she was ordered to undertake one final patrol, centering on the landing and picking up of a party of agents in Greece.  She sailed on 26 December 1941.


She was lost somewhere in the Aegean Sea in January 1942.  No-one knows her position.  Two years ago I thought I had narrowed her resting place down to somewhere on a narrow 27 mile long track, running northwest from Cape Sounio in Greece (the location of the famous temple of Poseidon) to Piraeus.  I’ve spent the past four years visiting Greece in search of a sheaf of official permissions to search for her with a sidescan sonar.  This article brings the story up to date so far. 


On the introduction of our Defence Attache in Athens, Colonel Tony Morphet, I met Dr Angelli Simosi, who runs the underwater archaeology authority in Athens.  Greek law defines “archaeology” as anything older than 50 years (which makes me, and probably a few of the NR’s readers, archaeological relics), so Triumph lies firmly in her domain.  Dr Simosi’s permission is essential.


At our first meeting Dr Simosi was sympathetic (if a little puzzled by this eccentric Scot in her office), but told me quite firmly that Triumph was not where I thought she was, but instead near an island called Antiparos, bang in the centre of the Aegean, and sixty miles in the opposite direction to Piraeus.  A stilted semi-bilingual discussion followed, in which it became clear that Greek naval historians were convinced that Triumph had been sunk following some events on Antiparos. 


I was nonplussed.  Nothing in any RN records, including the Confidential Staff History of the Mediterranean submarine campaign, made any reference to Antiparos.  What was this story, and where had it come from?  As it was clear that Dr Simosi was pretty reluctant to give me permission to search in the wrong place, I set out to chase the Antiparos story down to its roots.  It turned out that I had much to learn.


At first I found myself at a dead end.  There were no references to Antiparos anywhere in any record or history (including Kew) other than Francis Jones’ book “Escape to Nowhere”, the memoir of a soldier’s journeys through Greece after escaping the surrender at Kalamata in April 1941.  Jones tells how, during his wanderings, he ended up in an Italian prison with 22 Australian escapers who had been captured at Antiparos, and who told him that they had been captured while waiting to be picked up by Triumph.  There were no details or dates in Jones’ book, and his report was in any case hearsay, so didn’t give me a very strong lead, but, my “Triumph” luck was in.  A chance meeting with a prominent Greek wreck-hunting diver[1] gave me a lead to Antiparos, in the form of a reference to a file at Kew with the rather intriguing name of “the Atkinson Grammatikakis affair”[2].  I made a date, and headed to Kew.


Bullseye.  The Atkinson Grammatikakis file contained a sheaf of papers and signals about an intelligence disaster that was triggered on Antiparos in January 1942.  Space doesn’t allow me to go into the report in detail, but in outline the file amounted to a blazing row and back-covering war between SOE (tasked with sabotage and sedition) and MI9 (tasked with recovering escapers), over a man called George Atkinson (Lt RASC) who had been landed at Antiparos by Triumph on 30th December 1941, and then captured.  As will be seen shortly, Atkinson’s capture had acute and embarrassing consequences.

This is his story.

Summer 1941 in Cairo.  8,000 Commonwealth soldiers trapped in southern Greece had surrendered at Kalamata.  Perhaps another 1,000 evaded capture (or escaped soon after) and disappeared into the hills.  In Cairo the local head of the Special Operations Executive (SOE) was trying to establish what SOE should be doing about sabotage and sedition in Greece.  The problem was acute.  SOE had no “stay behind” people in Greece.  Some weapons and explosives had been left behind, but no-one knew where they had ended up.  MI6’s people in Greece had all been blown or captured.  Greece’s political scene was highly fractured and contentious.  Royalists supported the return of an absolute monarch (at that time licking his wounds in Cairo).  Liberal democratic Venizelists wanted a moderately right wing constitutional monarchy and democratic government.  Republicans wanted a democracy without the king.  Two varieties of communists were working towards a dictatorship of the proletariat.  Metaxans supported the deposed dictator Metaxas.  Each group hated the others, and personal loyalties were by no means clear.  In Athens and elsewhere resistance cells were springing up (Ali Baba, Prometheus, 5-16-5, the Kyrou Group, the Zannas Group.  In January 1942 an SOE memo referred to 22 known cells).  People were sometimes members of more than one cell.  Envoys and messengers were turning up in Cairo (having travelled across the Aegean to Turkey) bringing appeals for cash, weapons, radios and training.  The Royalists in Cairo were scheming to get HMG to endorse the return of the king after the war, while the Foreign Office was trying to work out what it wanted and who it should therefore support.  And all the while SOE in London was demanding action.

Atkinson had been captured at Kalamata and was being held in a transit camp near Piraeus.  The camp had been thrown up in a hurry, and was not secure.  Atkinson escaped one night, and headed to Athens to try and find the Underground.  By a series of chances he happened on the Bouboura sisters, who took him to the Ioannou family, living in Patission Street.  The Iouannous were prosperous tobacco merchants with an import business in London, so were good English speakers, and had the space to hide Atkinson and the New Zealander Ted Cooper.  The daughter of the household was Efi Ioannou, aged 15, who now lives in London, and who remembers Atkinson well.  "we called him Mac, for some reason.  The house was crowded, with our parents, us three children, our English nanny Miss Morris, two servants and two Allied soldiers.  Food was in short supply, and we had to live by buying food on the black market".

Atkinson spent a couple of months in Patission Street.  He occasionally went out - on one memorable occasion sitting on a park bench next to a German soldier - but eventually arrangements were made to take him and other escapers to Alexandria, where he turned up in October 1941.

Allied soldiers who had managed to evade capture, or escape, were trickling into Athens in and other Greek cities in dozens looking for shelter, and help in getting back to Egypt.  This was willingly offered by Greek patriots, but in turn led to heartfelt pleas to Cairo for money and transport. 

Efi Ioannou, now Kulukundis, in London 

In Cairo MI9, the escape and evasion organisation, was struggling with the same problems as SOE – who to trust, how to communicate, how to get resources into Greece, and how to get people out.  SOE and MI9 were in essence competing for the same resources, and were far from friends.

A key problem was transport – how could people and supplies be got into and out of Greece?  The Greeks themselves supplied the early answer – small coasting vessels called caiques were capable of making the 500 mile crossing from Alexandria to the Aegean, and the much easier crossing from the mainland of Greece to the coast of neutral Turkey only a hundred miles to the East.  Almost as soon as Greece fell a trickle of caiques and travellers began to arrive in Alexandria by these two routes.  Return to Greece was more problematic.  The German and Italian authorities kept track of caique movements, and an absence of more than three days was enough to brand a caique and its captain as Allied collaborators.  Hence, once a caique arrived in Alexandria it could not return openly to Greece. 

A caique could pick people up in Turkey, if they could be got there covertly (the Turks were neutral, but basically hostile to the Allies), but inbound travellers would first have to travel by land from Cairo, and then take their chances on the arrival of a suitable smuggler.  In theory supplies could be parachuted into Greece, but in practice this wasn’t attempted until a year or more later, and problems with aerial navigation by night made parachute drops a more miss than hit affair.  Which left submarine landings.  The techniques of beach landings by submarine had been well established and practiced during 1941, and a boat could land three or four people in relative safety, along with supplies, radios, money and even weapons.  The problem was that submarines were in short supply, and landing agents in Greece was a low priority compared with sinking Rommel’s supplies.  A landing mission consumed about one tenth of the patrol time available to a boat.  Some landings were being made in 1941, but only those with the highest priority.

It was in this troubled environment that, on the 18th October 1941, a caique appeared in Alexandria, under the command of a Greek rogue and peacetime smuggler called Harry Grammatikakis.  Among the soldiers aboard was Lt George Atkinson of the RASC. 

Atkinson was swiftly enfolded in the embrace of both MI9 and SOE, as a source of recent intelligence on what was happening in Greece, and as a possible volunteer to return and work for both SOE and MI9 in Athens.  Both organisations quite reasonably surmised that Atkinson’s contacts would be useful, and that he would be seen as a trusted emissary.  Atkinson was an enthusiastic volunteer, and took up residence in the flat of a senior SOE officer in Cairo while he was trained and prepared for his new job. 

He arranged for the BBC to transmit a message on the evening broadcast - to Efi Ioannou in Athens who, he knew, would be listening intently to her secret and highly illegal wireless in the attic of Patission Street.

MI9’s chief in Cairo, Major (later Colonel) Tony Simonds, had a problem.  Bringing evaders out by caique meant “burning” the caique and its crew every time, because the caique would have to breach its three day allowance.  Apart from the inconvenience of losing caiques, MI9 also had to compensate their owners in hard cash, and cash was in short supply[3].  Evaders were dribbling into Smyrna in Turkey (one pair rowed across the Aegean), but there were many more hiding in and around Athens looking for a way to get home.  The problem with using a submarine to pick up evaders was in coordinating evaders and a boat into the same place and time without any safe means of communicating with the evaders in Greece.  Simonds’ problem was how to get enough evaders together to meet a boat to justify tasking a valuable submarine, while avoiding the risk that the enemy would discover the rendezvous and sink the boat, or worse, capture her. 

The problem was compounded by the fact that a boat could only pick up reasonable numbers of people at the end of her patrol.  There were three reasons for this.  First, the length of a patrol was already defined by the amount of food and water a boat could take to sea.  Add 20 or 30 mouths to feed for more than a few days and the boat’s patrol endurance would be slashed.  Second, the presence of many additional people aboard would endanger a boat because they would quite simply get in the way during an attack or an evasion.  Finally, and probably crucially, a dived submarine had just enough breathable air to support her crew (57 for a T class) for about fifteen hours – allowing her to stay submerged during daylight with a margin at each end for evasion if she was attacked.  Add 20 or 30 sets of lungs and her dived endurance would fall to about nine hours – forcing her to spend some time each day on the surface in daylight. 

The air situation in a submarine is graphically described by Antoni Banach, quoted in "The Fighting Tenth":

"At dawn we would submerge, , and after the first four hours the air inside the submarine would gradually deteriorate and become noticeably stale.  After about ten to twelve hours the temperature would be 27 Celsius,  and breathing became more difficult..  At the time I was 22 and found it uncomfortable, but the older men found it more difficult.  By the time eighteen hours had passed since submerging, some of the crew were having a hard time...the last few hours before surfacing were the worst for the older crew and the inexperienced, who would by now be gasping and struggling for every breath and perspiring heavily."

Adding a couple of dozen people to the 57 already aboard would cut Trumph's dived endurance from 18 hours to 12.  In the closely patrolled waters of the Aegean this would be tantamount to suicide.  So, if a large party of evaders was to be picked up, it would have to be done at the very end of a patrol, at night, allowing the boat to make a good hundred miles of surfaced passage before diving in open water with almost no chance of detection.  But how could 20 or 30 evaders be reliably and secretly assembled in one place for a submarine rendezvous?

Grammatikakis seems to have suggested the answer - a small barely populated island in the centre of the Aegean called Antiparos.  A quick look at Google Earth will show that Antiparos is almost designed for Simonds’ problem.  The island could be reached in less than a day by caique from pretty much any part of mainland Greece, allowing the caique to return home in less than three days, and so avoid suspicion from the enemy.  At the southwest end of Antiparos there is a perfectly sheltered sandy bay with a fine shelving bottom, and with good clear access to open water.  Finally, the island has some houses, for shelter, but no Italian garrison.  Best of all (and I suspect this is where the suggestion came from) Grammatikakis was having a relationship with a girl who lived on the island, so was a familiar face to the locals.

By the end of October the decision was made.  Antiparos would be established as a secret base in which to assemble and hide evaders.  With evaders collected and waiting, a submarine could simply arrive whenever convenient, knowing that there would be a good number of evaders within 20 minutes’ walk to collect and take home.  No advance rendezvous would be needed, so reducing the risk to the boat of being bounced by the enemy in mid-pickup.  The base could be provisioned with food, fuel, clothing and a radio, and the time delay between an evader arriving and being picked up would give a chance that any German "plants” could be detected and dealt with[4].  The plan was set, and christened CONEY ISLAND, while the base itself was codenamed HARLEM.  The first provisioning trip was made by HMS Thorn, leaving on 10 November 1941.  A few days later Thorn dropped Atkinson, Grammatikakis and a Sergeant Redpath (a New Zealander) plus a radio on Antiparos.  An arrangement was made for Thorn to return a fortnight later, and meanwhile Atkinson and Grammatikakis travelled to Athens by caique to round up evaders.  Redpath made himself comfortable in the house of an islander called Tzavellas.  The plan worked like clockwork.  Around 28th  November Thorn returned to Antiparos, and picked up Atkinson and 20 evaders (codenamed rather unkindly ELKS), returning to Alexandria on 30th November.  Redpath and Grammatikakis stayed behind, the former to maintain the base, and the latter to return to Athens to round up more evaders.

Back in Cairo SOE was anxious to get in on the act.  A picture of the competing and confusing resistance cells was beginning to emerge.  The most efficient and active seemed to be the two communist operations (which later merged), and these had been sent a radio set by an earlier landing party from Triumph.  The set and its operator were codenamed 333.  SOE knew it had to get sets and operators to each of the main parties in Greece, if it was to avoid the resistance being completely co-opted  by  the communists.  It also needed to get money to its various cells, and to transmit its instructions for sabotage, subversion and assassination.  At the same time it needed to clarify just who was working for and with who, and to understand what resources of people and equipment each cell could command.  All of these tasks demanded an envoy, and of course a perfect candidate was sitting in the SOE flat in Cairo, in the form of Lt Atkinson.  SOE knew, from chats over an evening whisky, that MI9 was planning a second supply and pickup mission to Antiparos/HARLEM for December.  Simonds decided to ask Atkinson if he would undertake an SOE mission on the same trip – namely go to Athens, make contact with two of the cells (the Kanellopoulos Cell, led by Professor Panayiottis Kanellopoulos, whose code name was CUTHBERT, and the Zannas cell, led by Alexander Zannas), and hand over cash and bring a wireless operator with a set.  Atkinson readily agreed, but with a suspicion that his MI9 employers might not be delighted at his moonlighting, asked SOE not to mention his SOE tasks to his MI9 colleagues.  SOE nominated a Greek called Diamantes who they had been training in wireless operation.  He was given the codename DIAMOND[5], And was ordered to act as wireless operator for CUTHBERT.  A joint meeting with Captain Raw, commander of the First Submarine Flotilla, led to the mission being designated a joint MI9/SOE landing, and in SOE’s files the mission was codenamed ISINGLASS.

Atkinson’s mission had become rather complicated.  On his to-do list were the landing of three tonnes of supplies and two tonnes of fuel on Antiparos, a trip to Athens escorting DIAMOND, meetings with Zannas and Kanellopoulos, passing over of instructions and money, negotiations with Zannas and Kanellopoulos to try and persuade them to cooperate, rounding up of evaders in Athens, and then their covert transport back to Antiparos.

While ISINGLASS was being planned SOE made a change in the way it ran its agents, and responsibility for Atkinson was passed to a new hand (codenamed DSO – I haven’t been able to find his actual name yet.  DSO probably stands for Director of Special Operations).  DSO was new to the game of agent handling, and did not notice the accumulation of risk that was building up in Atkinson’s orders.  Compounding his error, DSO decided that Atkinson had so much to do that he needed a written operation order.  This was duly prepared, and among other items contained a complete en clair list of all fourteen members of the Kanellopoulos cell known to SOE.  Even though this op-order was prominently headed “NOT TO BE TAKEN ASHORE”, it was sadly to prove Atkinson’s death warrant, as we shall shortly see.

While preparations were being made in Alexandria and Cairo, Grammatikakis had travelled to Athens and by 10th December 1941 had managed to get 18 evaders to HARLEM (by the skin of his teeth, as the Germans got wind of the move and narrowly missed capturing the whole party south of Athens).  He returned to Athens to pick up food (no easy task given that the Germans were now actively starving Greece to death.  Greek history narrates that 400,000 Greeks died of starvation during the occupation, some tenth of the population), and fuel, in case he needed to take the evaders to Alexandria himself.

In Alexandria, Triumph had come back from patrol on 11 December, to the welcome news that this was her last patrol.  Her next trip would be across the Mediterranean to Gibraltar, and then home for leave and a refit.  The news was premature.  ISINGLASS needed a boat, and Triumph was the only operational boat available.  She was given the job.

Triumph spent Christmas day alongside in Alexandria, doubtless a kindness to a tired ships company from Captain Raw, and slipped her moorings on the afternoon of boxing day, with Atkinson, DIAMOND, and an experienced liaison officer, Captain Craig, aboard.  Craig was another New Zealander who had made a number of trips back into occupied Greexe.  Also aboard were five tonnes of stores for HARLEM[6].

 It is here that the Staff History and our history diverge.  Followers of this story will recall that all UK sources agree that Triumph had been ordered to land a party near Piraeus, and reported completion of the landing by signal dated 30th December 1941.  However, nowhere in any of the hundreds of pages of original documents that I have seen, including the SOE’s and MI9’s own internal postwar narrative reports, is there any reference to a landing on the Greek mainland, or to a second landing.  Triumph certainly had one mission that December – land the ISINGLASS party on Antiparos.  The SOE record has a passing reference to the possibility of landing Atkinson on Attica (the region in which Piraeus sits), but this reference is early in the narrative, and it seems likely that SOE fitted in with the MI9 plan (it began as MI9’s mission, after all) to land on Antiparos, knowing that covert travel between Antiparos and Attica was quick and relatively safe.

So, if ISINGLASS was always a mission to Antiparos, then why did Piraeus appear in the official record?  There are hints of a reason in Triumph's casualty list.  The CWGC records show that, in addition to her crew, two soldiers died in Triumph - Corporal Clive Severn and Bombardier Alfred Child.  These two men were Commandos.  There is only one piece of evidence to show that Severn and Child were aboard Triumph when she sailed on Dec 26th - a letter in the record at Kew from Captain Raw (SM1) confirming that they were aboard, and were lost with her.  Were these men to be landed on Attica, and then to be picked up again before heading to Antiparos?  It is certainly possible.  Given that neither men were officers it is unlikely that their mission would have been liaison with the Greek underground.  Perhaps they were landed simply to carry out sabotage, with a quick in-and-out objective.  It is possible that these were the agents to be landed "near Piraeus".

We will come back to Severn and Child after describing the fate of ISINGLASS, but for now must return to a dark cold December night in the central Aegean, and Lt Johnny Huddart on the bridge of Triumph as she noses quietly towards Antiparos on 29th December 1941.

Here the British record goes very quiet.  Triumph was lost, and the subsequent embarassment led key people in Athens to allow the story to be buried.  All we have is a cryptic reference in the Staff History – “Triumph signalled that the landing had been completed on 30th December”.  However, as my research in Greece wound its slow and painful way forward I discovered that the Greek record does contain the Antiparos landing, and its aftermath, in high relief, so we will move from the cold dusty files at Kew to the more colourful narrative as told to me by my diver friend and others.  We now also have, by the greatest good fortune, a detailed account of what happened after the landing from Jim Craig himself, in the form of a detaiiled operational report he wrote after the war.  His daugher Robyn kept a copy, which she has kindly sent on to us.

Sergeant Redpath, hiding out in Antiparos with a radio set, knew of Triumph’s coming, for the Greek narrative records that signal fires were set on the Antiparos hillside to indicate to Triumph that the island was free of Italian patrols.  Everyone on the island was in high spirits.  The evaders, 18 mostly Australian squaddies, had been in hiding there for nearly three weeks, and on the run for several months, and were looking forward to getting home to the fleshpots of Cairo, to friends, and perhaps some leave.  Antiparos’ villagers were also in high spirits (it was Christmas after all) and there is one report that all the HARLEMites had been making free with the supplies of rum landed by Thorn. 

Triumph, though, had urgent business, namely the landing of five thousand kilos of supplies by Folboat.  In the dark, even in calm weather, this was a daunting task.  You can see a Folboat in the submarine museum at Gosport, and I doubt that it would be possible to load one up with much more than 50 kilos of supplies at a time. 



The Folboats would have to make about a hundred return trips the two hundred or so yards to the beach.  Triumph probably had two boats, so 50 round trips per boat, and allowing for the fact that each boat would have only one rower, and with loading and unloading time, it must have been a huge challenge getting the stores up from below and ashore before dawn.  The Greek narrative suggests that “the whole village” turned to in aid, and presumably the 18 evaders were also conscripted as a carrying party.  I found one report that the villagers used their fishing boats, which would have speeded the business considerably.  The noise and illumination would have been tremendous – forty or fifty men bumping around in the dark with torches and lanterns, tripping, swearing, and encouraging each other.  It is easy to imagine Johnny Huddart biting his nails on the bridge, guns crew closed up, anxiously scanning the horizon to seaward for the shadow of an E-boat.  If Triumph was caught on the surface in shallow water she would have to fight it out with her deck gun, and would probably lose.

The stores were successfully landed, for Triumph signalled completion at 9pm on 30th December.  Having started probably no earlier than 2100 on the 29th, she might have spent seven or eight hours unloading before thankfully slipping away to sea.  She must have crept away just before dawn, dived to clear the area, then surfaced around 2000 that night to send her signal at 2121.

 At this point we need to focus the story on the evaders collected at Antiparos.  As far as they were aware, as soon as the stores were unloaded it would be their turn to board Triumph and be carried home in relative comfort.  Whether Redpath got his wires crossed, or whether he simply knew and didn’t tell them, the reality is of course that Triumph was at the start of her patrol, and was completely unable to take 18 passengers on board.  We don’t have to picture the disappointment, probably bordering on anger, felt by the 18 Australians, for we have Francis Jones’ hearsay account of it in his book.  To put it mildly, the Aussies were not pleased.  It seems that Huddart decided not to have a debate on the beach about air consumption and food and water, but instead simply to quote a change of orders (a convenient white lie) preventing him from taking the evaders aboard.  He softened the blow by telling them that he would be back in ten days to pick them up.  And so Triumph left, never to be seen again.

But the Antiparos story had hardly begun.  After Triumph’s departure the HARLEM base contained 18 evaders, Sgt Redpath, Lt Atkinson, DIAMOND/Diamantes, Capt Jim Craig, and of course Grammatikakis.  We have no record of what steps Grammatikakis took to make arrangements to get to Athens.  What we do know, though, is that Grammatikakis had one girlfriend in Antiparos, and another on Paros itself, and that his Antiparos girlfriend's mother felt that he was taking unreasonable advantage of her daughter.  One source I came across suggested that she rather fancied him herself.  Jim Craig's report says that he had got one of the girls pregnant.  One of the girls' mothers decided to make trouble for Grammatikakis by shopping him to the Italian garrison on neighbouring Paros.  It seems she did this without thinking too hard about the consequences for the 21 Allied soldiers and one Greek agent hiding out on Antiparos, but shop she did, and on the afternoon of the 5th January 1942 a small Italian patrol led by an officer arrived at the Tzavellas house and demanded entry.  The patrol was seen c, and Craig locked up the house and led his party into the countryside behind.  The patrol hung around for an hour and then left "seeming very suspicious" in Craig's words.  Craig and Atkinson had a heated conference.  Craig wanted to evacuate the house and move islands.  Atkinson wanted to fight it out if they came back.  Eventually Craig prevailed.  However, before they could act, at 2 am on the morning of the 6th January, there was a beating on the door and loud Italian voices outside.  After a short delay to allow the group to prepareTzavellas first spoke to the Italians, and eventually opened the door.  Tzavellas and Diamantes (DIAMOND) were taken prisoner.  A few minutes later the patrol began to search the house - with Craig and Atkinson hiding nervously in an upstairs room.

When the searchers came into the room Artkinson started firing - hitting the Italian officer in the chest with a bullet from his Colt 45 pistol.  The wound was not immediately fatal.  The searchers ran out, "helped on their way by a burst of machine gun fire from Redpath", leaving their officer on the floor.  The Italian was fumbling for a hand grenade.  Atkinson calmly shot him a second time, this time dead.  The searchers had not gone far.  A few moments later two hand grenades came flying through the window.  Atkinson opened a window and jumped out, followed closely by Redpath and Craig.  As they ran for the beach in the dark , with Atkinson leading by 50 yards, another grenade burst between them (this may have been what injured Atkinson's leg).

Craig lost track of Atkinson at this point.  Craig and Redpath briefly conferred and decided to go to the house of Baba Manoli, where the 18 or so other ranks were sleeping.  The men were swiftly roused, and carrying what supplies they could were hidden on a high rock promontory by 3 am.  An hour later Diamantes appeared.  he had escaped from the Italians in the confusion, and returned to the Tzavellas house to rescue his radio and Atkinson's suitcase - loaded with cash, gold and a bottle of Curare poison.  These had been too heavy to carry any distance so he had left them with Baba Manoli to hide.

Later that morning (the 6th) the Paros garrison came to Antiparos in force.  They began to comb the island for evaders, arresting Baba Manoli.  The Italians searched all day, but without finding the evaders.  The 7th January passed without incident.  Craig sent a patrol to ask Baba Manoli's friend if he knew where the suitcases had been hidden, but the man was too frightened to cooperate.  The Italians continued their search , but left the evaders undisturbed on the 8th and 9th of January.  Craig describes that Triumph was due to return on the night of the 9th/10th.  Other sources say the 10th January, but they might be referring to orders which allowed Triumph discretion between two dates.  Given that Triumph was attacking with a torpedo on the morning of the 9th, she might have been planning to come back on the 10th (so fitting in with the standard 24 hour sanitation period before a beach landing) but Craig would not have known this.  Craig set up torch signals flashing out to sea for three hours on the night of the 9th.  There was no reply.  He tried again on the 10th, and the 11th.  The evaders' food had run out on the 8th January.  By the 13th, still uncaptured, Craig's party had not eaten for nearly a week. 

On the afternoon of the 13th two evaders spotted a couple of sheep in a pen, went to steal them, but were spotted and fired on by an Italian patrol.  This seems to have led the Italians to the vicinity of Craig's hideout, for he realised that they were surrounded.  Ordering his party to split into groups of four or five and escape, Craig took Redpath, Diamantes and Lt Clarke through the Italian cordon and went in search of a boat.   The search was fruitless - the garrison had removed all boats.  Craig remained in hiding until the 17th January, when his party was betrayed by a local who had given them some food the night before.  Capture swiftly followed.

Red Cross records show that Atkinson was captured on 11 January.  There is a report that Grammatikakis, having escaped the initial roundup, stole a fast launch and returned to Antiparos to try and rescue some evaders, was captured, escaped again from under the noses of his guards by diving into the sea, and then vanished.  I also came across a hearsay report that the Italians discovered Triumph’s intended return date from evaders captured early in the roundup, and had anti-submarine schooners on hand ready to ambush her if she did indeed appear.  Presumably signal fires were also set.  Craig's account, though, confirms that none of his people  were captured until the 13th at the earliest.  Atkinson was captured on the 11th.  There might have been another evader captured earlier who wasnt in Craig's party, but it is difficult to see how that would have happened.  Craig reports that an E boat appeared on the 14th hoping to catch Triumph on the surface, but makes no mention of schooners.

Either way, Triumph did not return, having been mined somewhere en route.  The Antiparos story is not over yet, but I would like to jump briefly to the reason for all this research, namely the search for her wreck.  Once I had got this far into the story my heart sank.  So far from making her neat passage from Sounio to Piraeus, giving me a tidy track to search, Triumph had actually been heading in the diametrically opposite direction.  My beautiful track plot had disintegrated in front of my eyes.  If I had the exact date of Triumph’s planned return to Antiparos then maybe I could construct a new track plot from the Sounio torpedo report to Antiparos.  However, the initial analysis did not look good.  If the pickup was on the night of 9/10 January then normal operating procedures would lead Triumph to arrive off Antiparos no later than 1600 on the 9th, dived, for a periscope reconnaissance.  This in turn would require a dived approach at 3 knots, putting her dawn diving position at most 30 miles from Antiparos.  Adding up the speeds, times and distances meant that Triumph could not fire a torpedo in daylight off Sounio on the 9th, and reach the Antiparos beach by 1600 on the same day.  Something had to give.  One possibility is that Lt Huddart decided  that a periscope reconnaissance was not necessary, given the presence of allied soldiers on the island and the use of fires to signal all clear.  If this was the case, then Triumph could have fired her torpedo, made a dived passage southeast to Antiparos, surfaced after dark and headed in on diesels at 14 knots with plenty of time to arrive before midnight on the 9th.

 Another possibility is that the Sounio torpedo report was wrong.  I recently obtained a translation of the war diary for the Kreigsmarine high command.  This fascinating record summarises naval activity throughout the Reich for every day of the war, with commendable german thoroughness.  Each day's report has a section on the Aegean.  The diary for January 10th 1942 records "a submarine made an attack on one of our lighters off Sunium (sic) on 9th January".  I take that as pretty conclusive evidence that the attack did indeed take place.  Intriguingly, the diary for 9th January records a submarine sighting off Antiparos.  If Triumph was firing her torpedo in daylight off Sounio on the 9th she cannot possibly have been at Antiparos on the 9th as well.  Sightings are not definitive events - observers can and do make mistakes - so the report might simply have been wrong.  Another possible explanation is that the day was wrongly recorded, and that the sighting was on the 8th and only recorded and reported on the 9th.  Triumph might have been lurking off Antiparos doing periscope reccoinaissance the day before her attack at Sounio.  Either way, the sighting report does not add much, but the torpedo attack certainly did happen.

A third possibility is that Huddart was planning to come in on the night of 10/11 January (with “ten days” meaning “after ten days” rather than “in ten days”).  This interpretation fits Craig's report, and also some second hand information on Triumph's orders.  A pickup on the 10th would allow Triumph to follow her normal pattern of periscope reconnaissance in daylight and a clear 24 hours with no attacks beforehand.  Depressingly, if this was the correct date then Triumph would have had a clear 24 hours in which to wander about the central Aegean, making the creation of any kind of track plot rather difficult, and substantially enlarging the area that we would have to search.

 In any case, my neat Piraeus plot was now history.  Dr Simosi was right.  Doubly depressing was the thought that while I would be willing to spend a great deal of time and money searching a small area with a high probability of success, I was frankly less keen to double or triple the time and double the expense to go searching for a needle in an Aegean haystack, and I doubted Dr Simosi would give me permission to do that anyway.

Depressed, I returned to ISINGLASS.

By 17th January all the evaders, with Atkinson, Redpath, Craig and DIAMOND, were in the bag.  Disaster struck.  Atkinson had taken his operation order ashore with him.  Perhaps he forgot he had it in the rush to offload stores.  Perhaps he just didn’t trust his memory with a large number of unfamiliar Greek names.  Craig was convinced that he took it ashore deliberately  Either way, when he was captured his operation order was captured with him.  The Italian Second Bureau immediately recognised what they had acquired, and passed the list of names to the Gestapo in Athens for rounding up.  No warning could be given, but the word must have passed quickly around the Athenian underground, for only four of the fourteen were captured by the Gestapo.  However the Kanellopoulos cell was thoroughly blown, along with the credibility of the SOE and MI9 as safe comrades in arms.  In the words of the SOE narrative history written after the war: “..this affair had pretty effectively knocked on the head the only party, that of Kanellopoulos, whose programme in the least degree met the wishes of the Foreign Office, or which had any chance of acting as a bridge between the (Greek) King and exiled government and the resistance.”[8]

After capture Atkinson naturally tried to present himself as simply trying to help escaping allied soldiers, and therefore not a spy.  His operation order sadly condemned him.  After almost a year of hospitalisation (he was badly wounded in the leg, perhaps by the last grenade, we dont know), followed by movements around various Italian prisons, Atkinson ended up on trial for his life in a Military court on Patission Street in Athens on a charge of espionage and murder (the law of war did not allow an escaper to kill the enemy), in front of a panel of two Italian judges and a general. 

The court was held in a building only a couple of doors up from the Ioannou House, where Atkinson had hid in the summer of 1941.  Efi Ioannou sneaked into the public gallery to see the trial, but was then firmly told not to go again in case Atkinson recognised her and accidentally blew her cover.  It must have struck him as bizarre to be tried for his life two doors away from friends and sanctuary.

Charged with Atkinson were 25 Greeks who had been implicated in the Kanellopoulos cell.  Atkinson was convicted, and condemned to be killed, and was shot a month later at the Kaisariani shooting ground on 28 (or possibly the 23rd, according to Red Cross records) February 1943.[9]  Reports suggest that the Italians also shot several villagers on Antiparos - presumably Tzavellas and Baba Manoli.

There is a memorial to Atkinson and his fellows on Antiparos, by the chapel on Antiparos.


photo courtesy of Andrew Loudon

The names on the memorial are that of Atkinson, George Kapoutsos, three Patelis brothers, Spyros Tzavellas (whose house they were hiding in), Tsantanis, and Kosta Arvanitopoulos.

photo courtesy of Andrew Loudon

 The Patelis brothers' father was also captured and died of starvation on Rhodes.  Tsantanis and at least one of the Patelis brothers were killed by a German firing squad on the same day in February 1943.  Another of the Patelis brothers was imprisoned rather than killed, on the grounds that he had six children to support.  It is possible that the three Patelis names are the father and two sons, rather than three brothers.

I have not yet been able to visit the war memorial myself, but am indebted to Andrew Loudon, who has a house on AP and kindly sent me these photos.

Craig and Redpath survived the war, escaping again from an Italian prison camp.  Craig won the MC twice.  Grammatikakis, by now thoroughly blown, requested permission from the Allies to relocate to the Congo, where he eventually went in December 1942.  To his credit, he had made every effort to rescue his comrades, and retired from British service with credit.  There is a memorial to the incident and the Antiparos dead by the Tzavellas house, which still stands.

My track plot had been demolished, and my search is back to square one, but one question lingers – why did none of this enter the British records (for this is the first time that the Antiparos incident has been written about anywhere in English, as far as I can find out)? There are several possibilities:

 -         the Antiparos affair was a fairly large intelligence disaster.  When the Staff History was being written, in the 1950s, it is possible that an active decision was made to suppress it in order not to destabilise Anglo Greek relations at a time when the political future of Greece was far from certain.

 -         Triumph’s depot ship, HMS Medway, was sunk in 1942, presumably along with all the staff documents about the operation.  It is possible that the Staff Historian simply didn’t know about the Antiparos incident.

 -         All special operations were classified Most Secret.  The Staff History was only classified Confidential, so if the Historian did know about the Antiparos incident he might have censored it on the grounds of classification, and used the vague reference “near Piraeus” to cover the Most Secret actual location of Antiparos.

 -         MI9 procedures were similarly classified well into the 70s and even the 80s, against the possibility that they might need to be re-used if the Warsaw Pact ever did decide to roll over the Inner German Border.[10]

 -         Neither SOE nor MI9 were keen to publicise what was in essence a joint disaster.  Hence both of their postwar narratives skip lightly over the Antiparos incident, and the Staff Historian might therefore have missed it.

 We probably wont find out why Antiparos and Atkinson disappeared from the record, but at least they are now back in it.

As for the search, sadly we wont be putting a sonar into the water in 2015, but if the torpedo report comes up as accurate, with a time and a place, then we might be able to establish a new track plot.  We have now bought a boat suitable for the search, and hope to deploy her to the Aegean in 2017.  Each year sonars get better and slightly cheaper.  We will find Triumph.


The Mystery of Severn and Child

Commando records and the CWGC both suggest that two more men died in Triumph - Corporal Clive Severn of the Northamptonshire Regiment, and Bombardier Alfred Robert Child RA.  This presents a bit of a mystery.  No record has come to light of what these two were doing, and Triumph's official crew list leaves them out.  Both men had made recorded submarine landings in the past.  So what happened?  We are thinking about where we can find some more information, but meanwhile some speculation is possible.

The two commandos cannot have spent the whole of Triumph's patrol aboard - that would simply not have been tolerated by the, or by the men themselves.  So, it looks more likely that Triumph landed them early in her patrol, and then picked them up again before Sounio (ie before 9th January).  This might be an explanation for the Official History record, that agents were landed near Piraeus.  We had always discounted this report as either a mistake or a cover.  But what if Triumph did indeed land two more agents - Severn and Child - after the Antiparos landing operation?  She might easily have done so.  She would not have gone anywhere near Piraeus by sea - the waters to the north of Sounio were considered highly dangerous and too risky for a boat - but she might have landed them at her old beach of Kaki Thalassos, which is in fact quite near Piraeus as the crow flies.  Severn and Child could have carried out some sabotage ops and been back to be picked up within a few days.

Somewhere out there a record exists which will answer the question.


SOE and MI9

As I researched a nagging question kept coming up - why were SOE and MI9 so antagonisitic?  After all, they were both on the same side, fighting the same enemy.  It seemed odd to me that there should be friction, even outright infighting, between them.  I have found some insights into that in Artemis Cooper's excellent book on Cairo In The War.

SOE's headquarters in Cairo were at Rustum Buildings, not far from GHQ.  Its politcal objectives came direct from London, but its actual operations were controlled by a Special Operations Committee in GHQ, on which SOE representatives were treated "more as prisoners in the dock than as committee members" (Sweet Escott again).  Rustum was known to every Cairo taxi driver as "Secret Building". 

Bickham Sweet Escott worte a history of the SOE, and remarked of Cairo "Nobody ...can possibly imagine the atmosphere of jealousy, suspicion and intrigue which embittered the relations between the various secret and semi-secret departments during that summer of 1941"




[1] Kostas Thoctarides, who discovered the wreck of Perseus, among many dozens of others.

[2] HS5 /524

[3] In a wonderful note written at the end of the war Colonel Simonds calculated how much each evader had cost to recover, coming out at a very modest £30 a head.

[4] The Germans made frequent attempts to feed their own agents into the escape channels in order to destroy them.

[6] Either three or five tonnes, depending on which source one believes, but in both cases a large quantity to squeeze into a submarine.  One source suggests that some of these were destined tor Athens.

[7] Date Time Group, which give the time the signal was sent

[8] Kew HS7/151 page 241

[9] He is buried in the Commonwealth Cemetry in Athens

[10] Not my view, but that of Airey Neave, in his MI9 memoir “Saturday at MI9”


HM Submarine Triumph's last patrol - December 1941

  Site design and content copyright © 2011-2019 HMS Triumph Association, all rights reserved.