HMHS Britannic Expedition Summary 1976-1999

Copyright ©1999-2005 by Simon Mills

Posted with the kind permission of William H. Garzke, Jr, chairman of the Marine Forensics Panel of the Society of Naval Architects and Marine Engineers

Ken Marschall's paintings reproduced here with permission from Trans-Atlantic Designs, Inc.

Introduction: The following report is based on the combined findings of five expeditions to the wreck of the British hospital ship HMHS Britannic, located in the Kea Channel in the Aegean Sea. In that a more detailed forensic analysis has not yet been carried out, it does not at this stage attempt to suggest a definitive explanation for the scale of the damage located to date, or even its cause, but by combining the information from the recent explorations of the site with selected historical documents and other discoveries made in the world of marine forensic analysis, it is possible to put forward a more considered opinion of what might have happened from a historian’s point of view. A more detailed forensic analysis of the wreck will hopefully follow in due course but, in the meantime, this report acknowledges the work carried out on the Britannic wreck by the following personnel and organisations:

1976: Captain Jacques Cousteau
1995: Dr. Robert D. Ballard
1997: Kevin Gurr (IANTD)
1998: Nick Hope (Starfish Enterprise)
1999: Jarrod Jablonski (Global Underwater Explorers)

Greek Diving Liaison:
Hellenic Navy General Staff (1995)
Sotiriou Diving College, Athens (1997)
Mr. Kostas Nizamis (1998/99)

Historical Background

The circumstances surrounding the sinking of the Britannic have remained clouded since the day she went down on November 21st, 1916. Unlike her sister, Titanic, whose demise was examined in a very public arena, the lack of information pertaining to the loss of the Britannic has proved particularly frustrating to historians. This is not to say that there is anything to hide. So many ships went down during the First World War that any idea of a thorough investigation into each shipping casualty would be impractical, while the unstable political situation faced by the Allies in Athens during November 1916 would seem to justify the speed with which the survivors were repatriated. In addition, it should be remembered that few survivors actually saw anything of particular relevance, as the vast majority of the crew and medical personnel were at breakfast at the time of the explosion. Only three individuals actually gave what might be considered as useful evidence and, curiously, two reported seeing the wake of a torpedo. These sightings, however, were at different ends and on opposite sides of the ship. Not forgetting the vague (but historically tantalising) references by lookout J. Conelly that, shortly before turning over his watch at 8.00 a.m., two “suspicious objects” were in sight, it is not surprising that once the final report was completed some three days after the sinking, it remained inconclusive. It concluded by saying:

“The effects of the explosion might have been due to either a mine or a torpedo. The probability seems to be a mine.” [8]

Due to a number of allegations published by the Central Powers in January 1917 in order to justify their re-introduction of unrestricted submarine warfare, unfortunately the Britannic’s reputation has suffered greatly. These accusations originally concerned alleged Allied misuse of their hospital ships (including Britannic) for the purposes of transporting illegal military personnel and supplies, but over the ensuing years the legend has grown up that the Britannic was, in fact, carrying an illegal arms shipment. Even though extensive examination of the wreck has revealed nothing at all suspicious to date, some historians still persist in using this wartime propaganda to put forward what has now become a relatively standard “conspiracy theory”. Even the mere hint of munitions being on board was enough to endanger the dive permit obtained for the 1995 Ballard expedition, but fortunately the expedition eventually went ahead as planned.

Nor is the Britannic’s case helped in any way by the fact that for some time the wreck was not in its reported sinking location. When Jacques Cousteau first attempted to locate the wreck in 1975, he could find no trace of it in the position marked on the British Admiralty chart for the area. It was only after widening the search area and employing the new side-scan sonar technology developed by Dr. Harold Edgerton of the Massachusetts Institute of Technology, that Britannic’s true position was eventually found on December 4th 1975, some 6.75 nautical miles north-east of the charted position.

The reasons for this error are unclear, but an examination of the records at the British Hydrographic Office, Taunton, Somerset, indicates that the incorrect position was first recorded during a survey of the area carried out in May 1947. Needless to say the conspiracy theorists had a field day and went so far as to suggest that the wreck was deliberately misplaced in order to impede any future investigation of the site. It has to be said, however, that any serious attempt to locate the Britannic wreck would also have entailed a detailed examination of the ship’s log, and this document has never been classified in any way. Not only that, but it quite clearly states that Britannic went down approximately three miles north-west of the Port St. Nikolo light on the Greek island of Kea, which is actually a reasonably accurate position. The log is still available for viewing at the British Public Record Office, Kew, in London. The file number is BT 165/1569.

The exact position of the wreck was refined by the American Navy’s submarine NR-1 during Dr. Ballard’s expedition in August 1995. For legal purposes it is not advisable to publish the data here, but suffice to say that it was located approximately two and a half miles NNW of Port St. Nikolo, close to the position indicated by Captain Bartlett on the day that she sank.

According to the individual divers who have visited the site, the extreme depth is reported to be 119 metres. Statements according to the least depth vary. In 1976 Captain Cousteau calculated it to be 80 metres, while the 1997 expedition placed it closer to 90 metres.

Orientation of the Wreck

The wreck lies on a heading of 253°, which would indicate that the ship therefore seems to be headed back in the approximate direction from which it was approaching prior to the explosion. In that Captain Bartlett ordered a southerly course following the explosion, in an attempt to beach the ship in shallow water before it was too late, this seems doubly odd, yet this curious heading might help to provide a possible explanation to justify the scale of the damage seen in the hull area beneath the forward well deck.

During the course of the sinking, it is likely that Britannic’s heading would have been affected by two factors. One sentence from the report of the Britannic’s senior medical officer, Lieutenant Colonel Henry Stewart Anderson, is particularly informative:

“The wake showed that the ship was moving in a wide circle to the right.” [11]

This circle to the right may well have been initiated by the list to starboard and may explain why Britannic’s rudder is turned considerably to port (by at least 10°), almost as if the captain was trying to compensate for this drift to starboard.

The second factor is the prevailing current in the Kea Channel. The closest estimation of the tidal conditions experienced in this area would probably come from the 1997 IANTD expedition organised by Kevin Gurr, as this expedition was carried out during the month of November, the same time of the year that Britannic was sunk. Mr. Gurr reported that some of the currents experienced were so strong that during one in-water decompression period (which usually lasted approximately three hours) the decompression platform drifted some seven miles from its exclusion zone – a dangerous situation in what is now a particularly busy shipping lane. This would therefore suggest that the prevailing North Easterly currents in the Kea Channel at this time of the year can be particularly strong and this would almost certainly have been enough to help turn the ship. However, once the Britannic’s bow had touched the seabed it would have been impossible for it to pivot with the bulk of the hull itself, which still remained above water. With the entire weight of the fifty thousand ton hull concentrated in this area, not to mention the fact that it had been weakened by the effect of both the explosion and the high-speed dash for the shoreline, the stresses exerted in this area were massive. The resulting video images seem to confirm that a twisting motion has caused the bow to become almost severed from the main body of the wreck itself. The hull still remains in one piece, but only by virtue of a number of deck plates at the level of the well deck, which still seem to be intact. Interestingly, as on the Titanic, the two 50 cwt cargo cranes manufactured by Stothert and Pitt (Bath, UK) remain comfortably in situ, which might suggest that the overall structural integrity of the surviving deck in this area is stronger than might be supposed. However, in that Britannic actually lies on its starboard side at an approximate angle of 85°, while Titanic is sitting upright, the fact that these cranes are still in position is perhaps a little more surprising.

The Mine

The mine responsible for the initial explosion which sank the Britannic is thought to be the standard C/12 type A mine, one of twelve laid by the German submarine U73 on the morning of 28th October 1916. This mine, which contained an explosive charge of 150 kg., was anchored to the seabed and had an operating depth of up to 150 metres. The mine was deployed to its set depth by a pre-set control that measured the water pressure. Mines designed to damage the larger ships were usually set to a depth of approximately six or seven metres.

However, the scale of damage that has been observed beneath Britannic’s forward well deck is considerably greater than is thought possible to be inflicted by a single weapon of this type. A number of theories have been put forward to explain the reason for this, which, up until more recently, usually centred around a secondary internal explosion, similar to the one reported on the Lusitania. Factors involved in this secondary explosion include:

Each of the aforementioned scenarios were discussed in detail by Dale Ridder, in the preliminary forensic analysis prepared for the 1998 Symposium, sponsored by the Society of Naval Architects and Marine Engineers and the American Society of Naval Engineers [5], and it is not the intention of this report to summarise these theories again. For now, however, it should suffice to say that in view of the fact that no munitions have been located in either the main body of the wreck, or in the resulting debris field (limited though it is) by any of the five previously mentioned expeditions, it is perhaps now time that this particular allegation be forever disregarded.

Nevertheless, the reasons for the magnitude of the damage still need to be determined. If munitions did not cause a secondary explosion then why is the hole so large? To date no one has reported evidence of any cargo in this particular area, so it is not currently possible to formulate a definite theory as to any contributory factors. This may prove possible once the mine anchor is located, when an investigation of the debris, which detached from the ship or fell from the hull during the earlier stages of the sinking can be carried out. However, it should also be remembered that much of the hull area in question might well have been taken up with open ward spaces, so in terms of cargo there may not actually be very much to find.

Riveted Seams

Because the effects on the area of the initial blast on the starboard side remains hidden to us, to date we do not have definitive evidence that the explosion caused any failure of the riveted seams of the Britannic. Nevertheless, there is documented evidence to suggest that the American aircraft carriers Yorktown and Enterprise suffered such damage from near miss bombs during engagements in May 1942 and October 1942 respectively. Similarly, the British battle cruiser HMS Hood was also narrowly missed by German bombs in September 1939, resulting in a number of sprung rivets in the torpedo bulge. Also, HMS Prince of Wales sustained damage to her shell plating from a near miss bomb while completing at the Cammell Laird Shipyard in Birkenhead in January 1941. The hull plating was distorted and rivets failed, causing the riveted seams to be opened up.

The cause of this extended underwater damage, sometimes referred to as “explosive water hammer” is not confined exclusively to ships. A more dramatic demonstration of the principle would be the development of the famous bouncing bomb by Barnes Wallis. Wallis realised that the effect of an explosive charge would be considerably magnified as the shock wave travelled through the water, and it can quite clearly be argued that if such structures as the Ruhr dams could not withstand these forces, less permanent structures such as riveted ships would be equally vulnerable – if not more so.

Nor were the weaknesses of riveted seams lost on the sea captains of the early twentieth century. The diary of Kapitän zur See von Egidy, who commanded the German battle cruiser SMS Seydlitz at the Battle of Jutland, was very precise on the matter when analysing the damage sustained by his ship:

“The torpedo bulkhead held, but it was seriously strained, as were parts of the armoured deck. Where the rivets had gone completely, the holes could be stopped with wooden pegs. Where they only leaked, which they did in great numbers – more than enough for our needs – they became a distinct menace because there was no way to plug them effectively.”

We should also consider that it was not unknown for warships, which have far greater underwater protection than any commercial passenger liner, to succumb to a single mine. The British battleship HMS Audacious and armoured cruiser HMS Hampshire were both sunk during WWI following single explosions, and bearing in mind the fact that both of these vessels would have had better protected hulls and considerably greater internal subdivision, not to mention better organised damage control parties, this might perhaps help to put the supposedly unsinkable Britannic’s loss in its proper perspective.

These facts, taken in conjunction with the forensic work carried out on the Titanic and Lusitania, as well as the reports of the “unzipped” hull plates on the wreck of the SS Arabic [7], would suggest that it is not unreasonable to assume that Britannic sustained damage of a similar nature. In view of the Britannic’s much-vaunted watertight skin, the possibility therefore remains that a seam of rivets may indeed have opened up sufficiently to allow additional flooding to one side of the ship. Such an event may well have contributed substantially to the increasing list to starboard, which the ship took on as she headed for the island of Kea.

There is, nevertheless, evidence of a failure of riveted seams on the port side. It occurs forward of the actual blast damage and probably results from the stresses exerted on the area from a hull whipping response caused by the detonation of the mine, or those stresses which occurred during the sinking process (due to the slight corrugation of the affected area), though marine growth in the area might possibly impede the retrieval of any rivets.

Metallurgical Analysis of Rivets

Much has been made of the forensic work carried out on the Titanic wreck in 1996 and 1998, and although it has not been possible to retrieve any similar samples from the Britannic to date, it might also be of interest to consider an additional and related report which has recently come to light. As a result of an independent analysis carried out at the University of Sheffield on a wrought iron rivet retrieved from the wreck of the SS Arabic, another White Star liner built in 1902, the following information was discovered on its chemical composition:

Acc (%)
+/- 0.01
+/- 0.002
+/- 0.02
+/- 0.02
+/- 0.02
+/- 0.02
+/- 0.003

The report is reasonably wide ranging, but, in a nutshell, its conclusions basically suggest that the carbon content (0.03%) of the rivet was about the level normally expected in wrought iron, while the sulphur level was low. This is good because higher levels would lead to increased brittleness, if iron sulphide is formed rather than the sulphur going into the slag. The phosphorous level, however, was somewhat higher than expected which would therefore have a detrimental effect on the rivet’s impact properties, high temperature performance and high ductility.

The rivet was also examined by optical microscopy, and point counting of the sample revealed that the volume fraction of slag was particularly high, ranging from ~16% at the top of the rivet head, whereas the bottom of the rivet contained approximately 25% slag. Admittedly the results of an examination on a single Arabic rivet can by no means be regarded as truly representative, though further samples are currently being tested at Sheffield, which will hopefully widen the database. However, bearing in mind that American and Canadian metallographic examination of Titanic’s rivets referred to in the 1998 Preliminary Forensic Analysis [5] showed that they contained 9.3% slag, which is higher than a desired 2-3%, this might suggest that the quality of the rivets was steadily improving throughout the first decade of the twentieth century.

Of course, it is all too easy for us to look back with hindsight and say that the quality of the materials used in those days is not up to the standard of those of today, but a closer examination of the Britannic’s specification book might provide considerable insight into the technological understanding of the day, for it contains a number of important observations which would seem to illustrate that the advantages of steel over iron rivets were appreciated even then [10].

The decision as to which parts of the ship were manually or hydraulically riveted was clearly laid out. The shell was to be hydraulically riveted from the keel to the turn of the bilge over the midship half length (0.4 of the ship’s length) for the length of the wing tanks, and forward and aft of these out to a line tapering in to the keel within the three-fifths length. Within the midship half-length, the longitudinal seams were triple riveted and were double riveted outside of that length in accordance with classification rules of that period. The Bridge sheer strake, Shelter Deck sheer strake and the strake below were to be hydraulically riveted in way of doublings etc. for about 3/4 length amidships; also the inner and outer Bridge stringers, the Shelter Deck stringer and the inner Shelter Deck stringer at the breaks. The keel plate was also hydraulically riveted as far forward and aft as was possible. The various types of rivets to be used were listed as follows:

(a) In punched holes, not countersunk.
(b) All rivets 7/8” diameter and above

(a) In drilled holes.
(b) For all internal work for sizes up to and including 3/4”.

Pan headed:
(a) For connecting Frames and reverse frames to Floors.
(b) In shell and in punched holes, not countersunk.

Iron Rivets: To be used for all hand riveting and to be of best quality. All rivets of and below 1 1/8” to be iron, except where otherwise specified.

Steel Rivets: To be used for Hydraulic Riveting, and for Rivets through Tank Top in way of Engine and Thrust Seats. Steel Rivets to be used for the plating of the Bridge Deck also in way of beams and foundation bars of large deck houses.

Expansion Joints

The arrangement of the expansion joints on the Britannic differed somewhat from those on the Olympic and Titanic. The two earlier ships each had two expansion joints located in the superstructure, whereas Britannic had three, not to mention a fourth further aft where the additional decking had enclosed the aft well deck. The reasons for this change in the design are not clear, though in that Olympic was such an experimental ship (due to her very size), it is quite possible that a number of additional problems were highlighted during her post-Titanic reconstruction throughout the winter of 1912/13, which necessitated additional changes in the design of the Britannic. It is also possible that Second Officer Charles Lightoller’s testimony at the British Inquiry about the opening of the forward expansion joint may have prompted a change in the number of expansion joints.

One of the more likely reasons for the change might be due to the unique gantry davits, which were to be installed on the ship. As a result, the lifeboats were actually stacked at strategic points around the boat deck, and with the weight of both the davits and the lifeboats concentrated in these areas, rather than spread evenly along the entire length of the boat deck, this may well have been enough to bring about a re-think in the ship’s technical designs. Figure 3 clearly shows the difference in the arrangements of the various expansion joints, though it would seem that the forward expansion joint on all three ships was located in the same area.

Because the forward expansion joint on the Titanic was found to be open by some 9 to 18 inches, there is some reason to believe that such might be the case on the Britannic. The images seen to date are not that close, but would seem to suggest that the same structural failure that occurred on the Titanic did not occur to such an extent on the Britannic. There were, of course, a number of differences in how the ships sank. Britannic never attained the angle of trim of Titanic’s near vertical position, while Britannic’s heavy list to starboard also differed somewhat from the slight list to port experienced on the Titanic. The large stresses in the Titanic might therefore have occurred over a longer period of time in the Britannic. Also, the Britannic’s bow pivoting on the seabed would probably have concentrated the structural stresses somewhat differently, and also prevented the forceful impact with the seabed experienced by Titanic’s stem. Two video flyovers of the boat deck would seem to reveal that the second expansion joint across the raised roof of the lounge was unaffected, while the forward expansion joint, visible on the side of the officer’s deck houses, is only marginally more noticeable. In that a penetration of the promenade deck in 1998 also shows the forward expansion joint having the same uniform opening (a few inches), it would therefore seem that the joints on the Britannic were not affected in the same way as those on the Titanic. Comparison of the visible openings of today with a period image of the officers’ ward on the B deck port promenade of the Britannic [3] shows that the expansion joint was particularly noticeable even before the sinking, thus supporting the premise that any additional opening up of the forward expansion joint is negligible.

The Hull

Generally speaking, aside from the area beneath the forward well deck and the lower sections of the bow, the external appearance of the hull appears to be in excellent condition. It is possible that the high level of marine growth might be concealing additional damage from the naked eye, but indications at this time would seem to suggest that rather than destroy the wreck, these growths have, in fact, helped to protect it.

Further examination of the hull has revealed a number of open portholes, though the pattern seems to be random and it is not possible at this stage to say whether or not they would have had any impact on the ship’s sinking. In any event, bearing in mind the fact that the important potholes on the starboard side are hidden, it might be argued that any assessment of the open portholes on the port side would be irrelevant, as by the time they were submerged the ship was already doomed.

None of the smokestacks are in position, though this comes as no great surprise as they were reported to have broken away as the ship sank. When Jacques Cousteau visited the wreck in 1976 he could only confirm that one of the stacks had been located, but fortunately the technology provided by the NR-1 submarine on the 1995 expedition enabled us to locate the other three very quickly. The missing stacks lie between five and six hundred feet to the north. Two of them look to be in a particularly good condition, still retaining much of their original oval shape. According to Mr. George Perman, who in 1916 was a 15-year-old boy scout and aft lift operator on the ship (and probably now the last survivor), the funnels detached from the ship in the order that they touched the water, with the No.1 smokestack being the first to collapse, and the fourth smokestack (the dummy) being the last.

Boat Deck & Deck Houses

Since the Britannic has been resting on the seabed for over eighty-four years, it is quite remarkable that all of the deckhouses (with the exception of the captain’s bridge) are completely in situ, and shows no signs of collapsing even now, though isolated patches of deterioration (particularly the port side children’s playroom) are visible. Even on the bridge there are identifiable remains of what once existed. At least three of the engine telegraphs (manufactured by J.W. Ray Ltd. of Liverpool) still hang there, though they have fallen from their mountings, while the wheelhouse telemotor (built by Brown Bros. & Co. Ltd., Edinburgh) still remains upright. The bridge telemotor is still in position, but has fallen from its mounting and hangs at the angle of the deck.

Nearly all of the davits are still in position and the only noticeable signs of degradation in the area seem to be in parts of the solid bulkhead, which runs along the entire length of the boat deck. It is interesting to compare the fact that Britannic, lying in warm and relatively shallow water, seems to be surprisingly better preserved than Titanic, which lies in an infinitely darker and colder environment. It seems likely that the thick natural barrier provided by these marine growths has somehow protected the wreck by turning it into what is effectively a man-made coral reef.

The Interiors

Due to the possibility of Dr. Ballard’s proposed underwater museum concept, divers wearing open circuit tanks have been instructed not to penetrate too far into the wreck if they discover any areas with preserved wood etc., for fear of disturbing any anaerobically preserved interiors. The superstructure, while intact, seems to be completely open, so no restrictions apply here, although inside the ship a number of areas now appear to have provided a few pleasant surprises, particularly areas within the forward hold, which features a number of well-preserved linoleum floors.

Divers on the 1999 expedition managed to penetrate the forward cargo hold and, as far as their video reveals, it is completely empty. In view of the generally excellent condition of the hold itself (even what looks like some reasonably well preserved decking), this might possibly raise questions as to the nature of the so-called “secondary explosion” which may have occurred. Even in the extensively damaged area of what was once hold numbers 2 and 3, parts of the hull look to be in extremely good condition, which could help to strengthen the argument that most of the visual damage was caused during the sinking process, rather than because of the blast.

Slightly further aft it is a different story. In 1998 American diver John Chatterton attempted to penetrate the broken fireman’s tunnel, in the hope that he would be able to observe some of the forward watertight doors which failed to function. Unfortunately the area was considered too hazardous to proceed too far inside and, although Mr. Chatterton observed what he thought was a partially open door, it was not possible to obtain any images.

Subsequent dives have also shown that any attempt to enter the stokeholds through the funnel casings is unlikely to succeed. Whilst the four smokestacks may have become detached during the sinking, divers who have attempted to enter the hull via this route have found the top of the funnel uptakes to be completely intact, and have been unable to penetrate any deeper than the saloon (D) deck. However, a number of additional features in this area have proved interesting, including numerous crew access ladders and the steam exhaust pipes. Clearly if there were any boiler explosions (as reported by several survivors) then it would appear that the blast damage was effectively contained within the hull.

Internal penetrations within the main body of the hull have generally been confined to the area of the superstructure, and to the higher levels of the first-class forward entrance. Whilst this area is primarily intact, with many of the support stanchions still in place, there seems to be very little evidence of what was once the grand staircase. This does not mean to say that it has completely rotted away. In light of the large quantities of wood panelling sold off after the war, there has always been some question as to just how much work had been done on some of the Britannic’s interiors (even Harland and Wolff cannot be specific), and it could be that a number of the wooden interiors deeper within the wreck are somewhat better preserved. In view of some of the pleasant surprises found in the forward hold, this still remains a distinct possibility.

In spite of the fact that the weather cover above the forward staircase, save for one broken pane of glass, is completely intact, unfortunately the ornamental glass dome itself has been found to be broken. Parts of the iron framework and white glass are still very recognisable and free of excessive marine growth, though the fact that it is not completely intact is something of a personal disappointment.

Video Analysis of the Break

The scale of visible damage on the Britannic wreck is so large that it is unthinkable that a single German mine could be responsible for the entire chasm. Therefore, after an analysis of the available historical evidence and recently obtained video material, I would like to offer the following theory as to the sequence of events:

  1. The initial explosion occurred on the starboard side in the area of the forward well deck at 8.12 a.m., resulting in extensive damage in the areas of hold nos. 2 and 3.
  2. The immediate hull whipping response, referred to by many survivors throughout the ship as a “shudder”, may have further added to the magnitude of the damage, though there is no detailed contemporary evidence as to exactly what was happening in the key areas of the ship at this time.
  3. Following the explosion, Captain Bartlett ordered a southerly course and headed at the fastest available speed towards the nearby island of Kea, in an attempt to beach the ship in shallow water before she foundered. An analysis of the official report into the sinking [8] seems to indicate that during this dash the forward holds began to fill more rapidly. It therefore seems probable that the already-weakened bow was exposed to additional forces, which inflicted further damage to the area.
  4. As Britannic continued to founder, so the ship’s stem eventually came into contact with the seabed. From this moment on any further forward movement was impossible, and the angle of the ship’s sinking hull probably reached its zenith at this time, or shortly after. It was at this point that a combination of the ship’s prior forward momentum, combined with a strong north easterly current, served to swing the main body of the hull around to its final 253° heading. Bearing in mind the stresses which would have been concentrated in the area of the bow at this time, it is not surprising that the already weakened structure beneath the well deck would suffer further damage as the bow was forced deeper into the sandy bottom.
  5. Once the ship had fully settled on the bottom, while the main body of the hull would have rested on the seabed, the area of the foc’s’le, assuming it were still attached to the hull at the time, would have remained unsupported. Whether or not the subsequent collapse was immediate or occurred at a later date is not clear, but video analysis of the affected area would indicate that the stresses exerted upon the hull by the now unsupported bow caused it to crack, so that the foc’s’le, when it finally detached, fell forward and settled on the seabed, causing the damaged area to be further prised open. As a result, weakened parts of the structure in the area of the break, which might hitherto have remained intact, collapsed in a random and unidentifiable manner.

In addition, there is an interesting piece of video evidence, which might argue against the exploding coal dust theory. An analysis of the 1999 video shows that the hatch to hold no. 3 (the bunker hold) in the forward well deck seems to be substantially intact and in its correct position. Mr. Jablonski has since confirmed this information. Based on the assumption that the force of an explosion would follow the least line of resistance, had there indeed been an explosion in the reserve coal bunker then it seems likely that it would have vented upwards, so destroying the hatch in question. In that none of the officers on watch at the time of the sinking seem to have reported any such occurrence, combined with the fact that that hatch cover still seems to be intact, this might be considered a strong argument against the exploding coal dust theory.

Future Operations

While Governcheck Limited is the current UK guardian of the wreck, there are, nevertheless, a number of important external factors, which have to be taken into consideration. The British Ministry of Defence are becoming increasingly sensitive to underwater operations carried out on any wreck classified as a war grave, though they do seem to have been very supportive of our activities carried out to date. The British Government also retains what is referred to as “Sovereign Immunity” regarding their sunken warships and, in addition, because the Britannic lies in Greek territorial waters it is not possible to carry out any activities at the site without the agreement of the relevant authorities of the Government of Greece. The general experience to date seems to be that as long as the work carried out is for educational or scientific purposes then there should be no problem, but any idea of making retrievals of artefacts for commercial purposes should not be considered. Therefore, in terms of raising the necessary finances for future expeditions, the sale of any artefacts to help offset the costs is not an option.

On a point of information, I understand from Captain Ioannis M. Fournarakis (HCG), the Maritime Attaché at the Embassy of Greece in London, that the maximum penalty for the illegal removal of artefacts from Greek territorial waters is fifteen years imprisonment. This, I hasten to add, is probably for more extreme cases, but it serves as a useful point if only to illustrate the fact that the Greek authorities take a very serious outlook on any activities of this sort.

As a result of the images and information retrieved by Dr. Ballard and the three subsequent manned diving expeditions, we now have much more information on the Britannic than we did five years ago. There is still progress to be made, but it is now reaching the stage where a reassessment of the approach taken might now be necessary. Thus far, both ROV and diving technology have proved useful and these technologies should continue to be employed in a properly co-ordinated manner, as and when practicable. However, we should face the fact that while further investigation of the interior may prove interesting, as far as a forensic analysis of the wreck is concerned, we may have already obtained as much evidence as we reasonably can from video analysis alone. Also, in that the divers are also carrying out the dives for their own recreational purposes, they often have a different agenda and might have a tendency to miss the important points that are of interest to the more scientific community. This is in no way intended as a criticism and I would like to go on record by stating that because of these individuals we now have considerably more information to hand than might otherwise have been the case. However, if the Marine Forensics Panel wish to continue the research in more detail, then it would have to consider specifically organising and funding an expedition which, in turn, would result in greater control of the underwater activities.

Analysis of the video to date has proved both illuminating and frustrating. On the one hand, we now have considerably more information on the interior of the wreck, but in terms of analysing the blast explosion damage, it has its limitations. Because the camera angle is so narrow, it is rather easy to get lost in the jumble of twisted wreckage, electrical cables and marine growth. As a result, reverse engineering regarding the appearance of the explosion damage is probably both complicated and limited. In an ideal world it would be nice to carry out a digital photographic mosaic, similar to the one already carried out on the Titanic. More than anything this will enable us to look at the Britannic in more detail, without having to continually resort to disjointed video images. Such a technique would also be extremely useful in mapping the entire site. Coincidentally, I have recently been contacted by Mr. Paul Matthias of Polaris Imaging (Rhode Island), who has actually expressed an interest in carrying out such a survey.

One possible short-term answer might be to study some of the SIT (Silicon Intensified Target) camera images taken during the 1995 Ballard expedition. The resulting black and white wide-angle shots taken above the wreck by the NR-1 will probably help to fill some gaps, and I understand that these tapes are still currently held by NOVA/WGBH, Boston, in Massachusetts.

Sonar technology has also been used to some good effect, though, like the Loch Ness Monster, finding any definitive proof has proved illusive and frustrating. Figure 6 illustrates that there is no lack of underwater objects to search for, but when we utilised the NR-1 for this purpose we often found that these targets were nothing more than oil drums or other assorted pieces of debris thrown overboard from passing ships in the Kea Channel. During the Kevin Gurr expedition in 1997 there was a potentially interesting sonar contact located quite near to the wreck but, due to a power failure at the GPS base station on Kea, they could not be certain of the exact position. Unfortunately they were unable to re-establish contact with the target, so there is still no definite evidence of having located either a mine or, more likely, its anchor.

The NR-1’s attempt to locate the U73’s minefield was based on the plans drawn by Kapitänleutnant Gustav Siess in October 1916, but in that a search of the area in question has revealed nothing of importance, it seems reasonable to assume that Siess’ position is not accurate. This does not seem unreasonable when you stop to consider the conditions under which he was labouring at the time the mines were laid (submerged and navigating by dead reckoning), but it does illustrate the fact that any future systematic search will have to encompass a somewhat larger area than hitherto examined, and that adequate time should be allowed for this exercise alone.

It is possible that over the ensuing years a number of these mine anchors may have been dragged by fishing nets at some time or other, but it is equally likely that some (if not most) are probably still there. If the initial blast point of the mine’s detonation can be located then there is an excellent chance of resolving the debate of mine versus torpedo once and for all, as we tried to do in 1995.

In any event, aside from the retrieval of a number of rivets for forensic metallurgical analysis, there is a convincing case emerging for future investigations to start covering in more detail the area away from the main body of the Britannic wreck itself. Bearing all of these aforementioned points in mind, it therefore seems clear that any future sonar sweeps of the area should not only have the ability to detect these objects, but should also the technology to be able to identify them. It would also be helpful for future expeditions to consider comparing the wreck of the Britannic with that of the 12,481 GRT French passenger liner SS Burdigala. On 14th November 1916, while serving as a trooper, this ship also succumbed to one of the mines laid by the U73 in the channel between the islands of Mikonos and Tinos, and an examination of the damage sustained by the two ships from similar weapons might provide further useful evidence.

As things currently stand, there are no definite plans for a detailed forensic analysis of the site in the coming year, while Titanic, once again, seems to be dominating the headlines, both on the television and in the court room. Recently, however, things have begun to change. On December 6th 1999 the History Channel aired the fifty-minute documentary Doomed Sisters of the Titanic, while Partisan Pictures of New York are about to enter the editing stage of Lost Liners, which, according to the latest information, is scheduled to be aired in America sometime this summer. In addition, on January 9th of this year Fox Family Channel aired a TV movie set on the Britannic, which, although somewhat less than satisfying from the historical point of view, might nevertheless help to raise the profile of the Britannic to the general public. Clearly interest in Titanic’s sister ship is on the increase, and, as a result, the time might now be right to publish the useful and low-profile work carried out to date in a more public arena.


[1] The White Star Liner Britannic. Engineering (27th February 1914).
[2] Mills, Simon. HMHS Britannic: The Last Titan (2nd Edition). Shipping Books Press, Shropshire, England, 1996.
[3] Fleming, Rev. John A. The Last Voyage of His Majesty’s Hospital Ship Britannic. Wordsmith Publications, Buckinghamshire, England, 1998.
[4] Cheetham, Michael. The Sinking of the RMS Titanic – Puncture or Brittle Fracture? Mechanical Properties of Early 20th Century Ship Steels and Ship Safety. University of Sheffield, Department of Engineering Materials, England, 1999.
[5] William H. Garzke, Jr., Simon Mills, Robert O. Dulin, Jr., F. Gregg Bemis, Dale Ridder, Dr. Timothy Foecke & David K. Brown. The Saga of HMHS Britannic: A Preliminary Marine Forensic Analysis. Proceedings From Research to Reality in Ship Systems Engineering Symposium (American Society of Naval Engineers & Society of Naval Architects and Marine Engineers, September 1998).
[6] The Titanic Commutator. (Fall 1977, Winter 1978, Vol. 15 No. 2, Vol. 15 No. 3, Vol. 20 No. 1, Vol. 23 No. 146 and Vol.23 No. 147 editions).
[7] Louden-Brown, Paul. Steel Ships and Iron Rivets. The Titanic Commutator (April 1999, Vol. 23 No. 145 pp 4-13).
[8] Official Inquiry into the Loss of HMHS Britannic, by Capt. H. H. Heard and Cdr. G. H. Staer, aboard HMS Duncan, 24th November 1916. Public Record Office, Kew, England. (File No: ADM 137/2171)
[9] The log of HMHS Britannic. Public Record Office, Kew, England. (File No: BT 165/1569)

[10] SS Britannic specification book, Harland & Wolff (1911-1915).
[11] Report to DMS Malta, by senior medical officer at sinking of HMHS Britannic. Lieut. Col. Henry Stewart Anderson (RAMC), 5th December 1916.
[12] Calypso’s Search For the Britannic. Warner Home Video, © The Cousteau Society, 1977.
[13] Titanic’s Lost Sister. Varied Directions for NOVA/WGBH, Boston, 1996 (TV/video).
[14] Doomed Sisters of the Titanic. MPH Entertainment for The History Channel, 1999 (TV/video).

Unreleased video and stills images resulting from the 1976, 1995, 1997, 1998 & 1999 expeditions.

Acknowledgements: (listed alphabetically) Dr. Robert D. Ballard, F. Gregg Bemis, Jr., William H. Garzke Jr., Kevin Gurr, Nick Hope, Jarrod Jablonski, Cathy Offinger, Dr. Toby Whillock and all of the divers, photographers and support personnel involved in the 1995, 1997, 1998 and 1999 expeditions who have helped to make this report possible.

Britannic home