The Roebuck Relics

The reasons for believing that the materials identified were an indicator of the final resting place of HM Ship Roebuck were:

1. The location of the bell in such a diverse wreckage field while inconclusive, was a clear pointer to the loss of a significant early vessel in the bay.

2. The location of the concreted grapnel near the shore, and in such a configuration as to indicate that its rope was once tangled and that it was irretrievable when abandoned, was considered highly significant.

The Captain’s Log of HMS Hastings for 5th April 1701 reads as follows:

'Satterday 5. . . Wind and Weath[er] Do [ditto]. In y afternoon came on board seven Of y Roebuck men at 10 at night got on board y Roebuck anchor being her small bower and fourtenn fath[om] of cable n giting of which o[ur] Longboat Lost her Grapnel'

3. Cr Les Moss, the President of the Shire of Shark Bay, and Hugh Edwards, noted wreck diver and author of a recently-published book on the bay, both considered that the clam was very similar to those found on the north west coast of Western Australia and in nearby tropical waters.

A diver photographer a partially covered clam shell

Geoff Kimpton and the author excavating the clam.
Image copyright WA Museum

As there were no others seen throughout the inspection and on the familiarisation dives conducted earlier, it was considered to have been brought to the island by human hand, most likely from the Indo-Pacific region. When Dampier’s accounts of his collecting shells while on the Australian coast were considered, the possibility that it was part of his collection became evident. In his account of events at Shark Bay that was published subsequent to his return to England after the loss of his ship, he wrote:

'The Shore was lined thick with many sorts of very strange and beautiful Shells…. I brought away a great many of them; but lost all except a very few, and those not of the best.'

(Wm Dampier, Quoted in the Wilkinson Edition, 1929:87)

Here Dampier refers to the loss of his ship and most of his shell collection. If not a Shark Bay shell, possibly it was one of the cockles or clams varying upwards from 10 pounds that were recovered from the waters of Irian Jaya a short while before Roebuck turned back for home. If it could be confirmed as non-indigenous to Ascension Island waters, and unless it could be proved that the clam was introduced to the area by others e.g. by American or British Service personnel or others travelling from the Indo-Pacific region, here was a clear indication that it was part of Dampier’s cargo. On the balance, the bell, the clam, the grapnel, the  concreted ironwork and other debris, including the ceramics in the rocks, were sufficient as an assemblage to conclude that the team had possibly located the wreck-site of HM Ship Roebuck. The finds and these conclusions were reported that afternoon to the Administrator HH Geoffrey Fairhurst  and the video record was shown to him, to Mrs Wendy Fairhurst, Museum Curator, and to other museum and island officials. After being advised by the author that the exposed objects were at risk and after faxing the British Admiralty (owners of HM Ship Roebuck) of his thoughts, Mr Fairhurst, as Island both Administrator and Receiver of Wreck, requested that the objects be raised.

The Movement of Sand Cover (A Fortuitous Event)

In describing the extent to which rocks were visible on the beach and underwater, we were advised that the sand movement on the beach was unprecedented in HH Fairhurst’s memory. This was subsequently confirmed in discussions with Mr Jimmy Young, a St Helenan, who after fourty years in residence (Since 1960) is the longest-serving resident on Ascension Island. He indicated that he had never seen the beach so eroded and that at one time it was possible to walk in almost a direct line from the Turtle Farm at the south end of the beach to the area now occupied by the ‘Saints Beach Club’ barbecue hut at the north end. That confirmed the understanding that the remains were only recently uncovered. Mr Young also advised of the removal of vast quantities of sand for construction work during the Falklands conflict, an activity that was halted by concerns over its effect on the turtle population which regularly laid their eggs there.

He had also dived with the 1985 RAF expedition, and indicated that they had tended to concentrate their searches in the deeper waters around the island, as they were prevented by the swell and their tow search methods venturing in too close. Finally Commander John Bingeman, leader of the 1979 search has advised that photographs of the beach taken in the 1940s when the Americans were landing equipment show the beach extending much further out to the west than is the case today. He also recalled that the Saints Beach Club was much further back from the sea on the occasion of his visit in 1979, and that there was also a smooth incline from the beach down to the sea - and not the sand cliff that the 2001 team encountered.

The Ceramics

After the dive resulting in the recovery of the bell and clam, Mr Young showed the Museum team ceramics that he had recovered just a few weeks prior to their visit, when the sand cover was even lower. He did not report finding them earlier, as he did not consider them possibly from the Roebuck.

Digital photographs and prints of the ceramics raised by Mr Young and the Museum team were examined by Dr Christiaan Jörg of the Groningen Museum, a specialist on 17th/18th Century export wares for the Dutch market, and Dr Michael Flecker, researcher and member of numerous expeditions involving the recovery of similar ceramics. They were also examined by Rosemary Harper and Jeremy Green, both involved in the examination of similar ceramics from south-east Asia. In respect of the jar, Dr Flecker’s comment read thus:

These jars are quite common. They have been found on the Witte Leeuw, which wrecked off Saint Helena in 1613, and on the San Diego, which was lost off Fortune Island, south of Manila, in 1600 [2]. I have also seen them at antique shops in the Philippines. This type of jar seems to have been manufactured for some time with little change in form. Shipwreck evidence shows that the form remained the same throughout the first half of the 17th century, and in all probability it changed little right through to the 18th century. They were most probably produced at kilns in Guangdong Province, southern China. The blue and white ware provided much more information, being more easy to date.

The blue-and-white jar lid and the shards are products of the Jingdezhen kilns, Jiangxi Province, China. The jar lid is very similar in form to hundreds of jar lids recovered from the Vung Tau Wreck of c. 1690 [3]. The distinct panelled decoration brings to mind kraak porcelain of the early 17th century, although this technique carried on through the transition period at the end of the Ming Dynasty and, in a less severe form, right through the Kangxi period of the Qing Dynasty (1662-1722). … While your lid is nowhere near imperial quality, it is of fairly high export quality.

Small piece of ceramic pottery

A shard from the recovered ceramics
Image copyright WA Museum

Two people holding items recovered from the wreck

Mr Jimmy Young and Flt Lt Richard Burke examine the ceramics.
Image copyright WA Museum

There is a possibility that the two shards are slightly later. While Green and Harper concurred with Flecker, Dr Jörg felt that, though it was not of imperial standard, the blue and white ware was of high quality indeed, the sort of materials that officers would acquire. All agreed that Batavia (now present-day Jakarta) was the most likely source of the ceramics, an important observation as Roebuck spent a considerable time there effecting repairs for the voyage home. Dr Jörg’s comments on the ceramics and the female figures;

The ladies in a garden setting are quite common motifs in this period. They are the so-called "Long Elizas", the long slender ladies, in Dutch "Lange Lijzen" which became Eliza's in English translation later. They walk in a garden with flowering trees, banana trees, large rocks with many holes specially collected for garden architecture and often you see a small dancing boy, in Dutch traditionally called a "zotje" (a fool). It is a standard motif, and on these pieces executed with great care and nice detail. It was certainly good quality stuff, not just the ordinary run-of -the-mill object and can be regarded, I think, as a personal belonging of someone like Dampier himself, or one of the ship's officers.

The Clam

The excavation team gathered around the bell and recovered

The Ascension Island and Museum team with the bell and clam.
Image copyright WA Museum

The clam, which was found lying on its face inshore of the ceramics in a turbulent location, had a heavily eroded back. While clearly Indo-Pacific in appearance, and similar to those found near Shark Bay and further north, expert comment on its provenance and the possibility that it was native to the Island was obtained from Ms Shirley Slack-Smith and Dr Paddy Berry of the Western Australian Museum’s Division of Natural Sciences. Ms Slack-Smith stated that;

From that character and from its general shape and size, I can say that it belongs to the genus Tridacna and possibly to the species Tridacna squamosa Lamarck, 1819. Therefore I can be certain that the shell would have been collected (by whoever) in the Indian Ocean or in the western part of the Pacific Ocean. The genus Tridacna has a distributional range within the tropical and sub-tropical waters of the Indo-West Pacific Region as far south as about Shark Bay and southern Queensland in mainland Australia, northern South Africa, and Lord Howe I and Pitcairn I in the Pacific. Northwards the genus extends up the E coast of Africa and reaches the north of the Red Sea, south and south-east Asia and Southern Japan. To the east, its range extends to the Line Islands, the Marquesas and Pitcairn.If the clam is of the species T. squamosa then it would have originated within a more circumscribed area. This probably doesn't extend southwards beyond the Ningaloo Reef and the southern groups of the G B Reef in Australia, Mozambique in the west, halfway up the Red Sea, southern India, the SE of SE Asia and southern Japan in the north, and eastwards only as far as the Marshall Is, Samoa and Tonga. These limits may be a little out of date but the pattern would hold good. So, the shell did not come from the Atlantic, for sure.

As indicated in the account of the Roebuck voyage reproduced earlier, there is no doubt that clams and other shells were on board and that many were lost when the ship was wrecked at Ascension Island. It is also known that Dampier collected shells at Shark Bay and LaGrange Bay on the Australian coast. When in the vicinity of the north-west end of New Guinea (Irian Jaya) not far from what is now known as Selat Dampier the crew also recovered a variety of ‘cockles’, which from their description are clearly clams, varying from 10 pounds through to 258 pounds weight

There is little doubt then that this particular shell is part of Dampier’s lost collection.

The Bell

Of the vessels expected to have carried a bell with a broad arrow, only HM Ship Roebuck is known to have been lost in the vicinity of North West (Clarence Bay) on Ascension Island. Also, a comparison of archaeological and other data indicates that the lesser importance given to a Royal Navy vessel, the more spartan were its decorations and fittings.

The bell as it first discovered on the sea floor

The bell as found. Note the lack of weed on the surrounding rocks.
Image copyright WA Museum

In respect of the sort of vessel indicated by the Ascension Island find, reference is made to the bell of the contemporary ‘fifth rate’ Dartmouth wrecked off Mull in 1690. It was 394 mm in height and it carried the broad arrow or ‘pheon’ and the inscription DH 1678. As indicated above, this 266-ton vessel, which was constructed in 1655 underwent a major overhaul in 1678 when the bell was fitted.

Drawing of the vessel Dartworth

HM Ship Dartmouth. The broad arrow on its bell appears on the previous page.
Image copyright WA Museum

Further pertinent comparisons can be made with the bells recovered by the Thanet Archaeological Unit from HM ship Stirling Castle (1703) and possibly the Northumberland (1703) wrecks on the Goodwin Sands. These two bells carried the date 1701, while that of the Stirling Castle also bore the broad arrow. The unadorned Clarence Bay bell (300 mm high by 340 mm wide) with a broad arrow, but with no other inscriptions, indicates that a Royal Navy vessel of lesser import than HM ship Dartmouth was lost in the vicinity of the find in Clarence Bay. Of importance are the similarities in the form and size of the broad arrows on each bell.

The Wreck Site of HM Ship Roebuck

It is evident that the first anchoring place of Wm Dampier’s HM Ship Roebuck was close to or at the present tanker mooring just off Long Beach in Clarence Bay, Ascension Island. No other place on the island fits the bearings and descriptions given by Dampier. His rescuers all anchored in a location not far from that same spot. Dampier’s record of having run his vessel in with what appears to have been a short-lived sea-breeze and his having anchored in seven fathoms before sending a small anchor ashore, allowing him to warp in fasten the ship after it grounded in three and a half fathoms of water a cable’s length from shore, would place the initial grounding point of his stricken vessel on or against the line of submerged rocks lying c. 200 metres offshore in North West Bay. These rise 1-1.5m off the seabed lying in that same three and a half fathoms or c. 6-7m (21-24 feet) of water and they would have prevented the waterlogged ship being warped in any further.

While John Penton’s evidence at the court martial shows that Roebuck was ‘full up to ye hatches with water’ when it grounded, Charles Harbree’s use of the words ‘having gott wthn a cables length of ye shore in sum short time after ye ship sunk’ at the same hearing indicates that the grounding and the sinking as the water flowed over the coamings into the ship were two separate events. Robert Sexton’s advice to the author that if one adds the c12-14 foot depth distance from the bottom of the keel to the lower deck to the c.5-6 feet distance from there up to the next layer of plank on the upper deck with the lip of the hatch comings a foot or so atop, then a distance of 21 feet from keel to the top of the hatch coaming is realistic.

This provides some indication that Roebuck lay on top of the rocks a cables-length from shore and a short while after, apparently when the hull was pierced by the rocks, it settled down onto the seabed a metre or so below. This also indicates why it is not necessary to further consider the possibility that Roebuck drifted back out so sea after it was abandoned, despite the fact that the cable holding it eventually parted, or was cut as described above. The ship Hall being ‘dashed to 10,000 pieces’ in the swells on the same beach in the August of 1834, provides a useful clue why Roebuck was not seen by the rescuers and this could also provide some clue why the cable would have parted as the stricken ship ‘worked’ back and forth in the swells before totally breaking up. The unadorned bell with a broad arrow indicates that a substantial, yet not a major, Royal Navy vessel was lost in the vicinity or to seaward of the find. Of all those service vessels lost on or near the island, only William Dampier’s vessel HM Ship Roebuck is known to have been grounded and abandoned in Clarence Bay, and from all deductions this occurred a short distance south and west of the bell find.

Mr Young’s late 17th Century Chinese ceramic finds, his report of timber, the other grapnels, assorted wreckage, fastenings, other ceramics, the chain plate, and the Indo-Pacific clam in shallow water inshore (east) of the last calculated position for William Dampier’s ship Roebuck are all consistent with materials expected from that vessel. The rolling of the clam soon after it was freed from the rocks to which it was attached, even in the moderate swells experienced by the Museum team, attests to the strength of the seas, even on good days, and to the ability of materials, even those as heavy and compact as a clamshell, to move inshore from a location to seaward.

It is concluded on the basis of this evidence that the assemblage emanates from HM Ship Roebuck and that the main portion of the wreck lies under the sand, abeam the bell which lay to the north and west and slightly to seaward of the other finds reported above that are located inshore and to the east. The area containing the large grapnel is considered the most likely place, and there is a distinct possibility that it lies on what remains of the wreckage mound, which will include the iron ballast that Dampier records loading into Roebuck before it departed England. Some of Dampier’s shell collection, including the 258 pound specimen recovered near what is now Selat Dampier off north-west Irian Jaya, may also lie nearby. Having agreed with both the Admiralty and the Island Administrator to perform a non-disturbance study only, this conclusion was not tested. Initially, the Dampier materials were stored in an on-site facility on Ascension Island until a decision was made on their conservation. Hearing of the finds, Mr Charles Barker, Managing Director of the Mary Rose Archaeological Services Ltd of Portsmouth offered to conduct the work and by consensus of the Island Administration and the Museum team it was agreed that the objects be sent there for treatment by Conservation Scientist, Ms Sue Bickerton and Head Of Collections.