The shipwreck, conventionally named Angra D, was lying in only 7 meters of water, some 50 meters away from the town's coastline, oriented on a west to east axis, parallel to the shore, 28 meters away from the midpoint of Angra C.
Most of the site was buried under the sand, with only a handful of timber ends protruding above the bottom's surface, in between some loose ballast stones. Beneath the ballast mound, the hull of the ship and some of its contents have remained well preserved in the sediments, mainly fine sand and silt. Preliminary excavations revealed that the wreck had an average length of 35 meters and a maximum width of 9 meters.
 |
Drawing the wreck. Below the diver, one of the deck beams. Photo: Paulo Monteiro |
Divers first conducted a controlled visual search around the mound area to locate other exposed remains and also to establish the ends of the wreckage, and determine the extent and depth of cultural deposits. As opposite to the winter situation, the area immediately surrounding the mound appeared to have been recently scoured out, since there were more exposed timbers at the north side of the wreck. Numerous concretions were scattered about, all of them belonging to the wreck of the Run'Her.
Once the search for additional remains outside the mound had been completed, reference baselines were installed to facilitate triangulation, elevation recording, grid placement, and photography. Location of the baseline was dependent on the exact position of shipwreck remains and the orientation of the hull structure. Marker and mooring buoys were also used in order to help divers reach and become quickly oriented to the site. The shipwreck site offered divers an advantage with regard to positioning since the excavated main mound was substantial and had identifiable features that could be used as points of reference, as happened with the boiler from the Run'Her, that served as a stationary reference point.
Although the visibility was very low - sometimes reaching zero visibility condition for days in a row - during the winter site examinations, the conditions at shipwreck site were very workable due to shallow depths (7 meters), a coarse sand bottom, and the potential for periods of good visibility (2 - 20 meters, exceptionally), in the spring time. The site's close proximity to docking facilities, a protected harbor, lifting equipment, and storage space was also an advantage.
Radiocarbon dating for three samples recovered from Angra D pointed to a date of loss around the turn of the XVth century.
The Excavation
After the removal of about 10 tons of ballast stone, the team discovered that Angra D presented articulated remains of a lower hull, composed by 9 floor riders, the keelson with the mast step assembly, ceiling planks, a complete assembly of the stern post and several isolated timbers at the stern and bow ends of the wreck. The excavation of the test units provided also a sampling of artifacts to help estimate the number and variety preserved on the shipwreck site.
After the complete removal of Angra C, the 16 crew members were devoted 12 hours a day, 6 days a week, to the complete removal of Angra D. While dealing with the inherent complexity of the task, the team also had to make do with logistical and physical constraints, like the lack of proper facilities for the archaeological studies and analysis. Space was very limited at this facility, with the 16 member team working from a 20 foot container on the first month of the project. This included temporary storage, artifact cleaning, cataloguing, analysis, filling operations, briefing space and changing rooms. Impacts from natural forces were a very real threat, especially during site excavations. A plan of protective measures, such as covering or buffering fragile portions of the shipwreck site, was thought out and developed prior to the advance of catastrophic storms, that did not occur for the project duration.
Hull timbers
Angra D measured 35 meters in length and had a maximum beam of 8.1 meters. Surviving timbers included keelson, 9 floor riders, bow knees, the main mast step, deck beams, stanchions and the sternpost assembly. The wreck was very heavily ballasted and several tons of rocks were removed in the process.
The wreck was listing to starboard, where 5,5 meters of hull had survived, with only 2,5 meters of the port side breadth surviving, abeam of the mast step.
Sternpost
Articulated remains of the stern measured 7,5 meters long to 4,2 meters high on the sternpost. The sternpost assembly was only physically attached to the wreck by the keelson and the keel, all outboard and inboard plankings thorn around that line of fracture. The rake of the sternpost was measured and found to be circa 70º, with sixteen frames recorded in the tail section, as well as several cant frames, deadwood and outboard planking.
 |
A diver inspects the results of the preliminary excavation of the sternpost. Photo: Peter Waddell |
The stern section, which was resting on the sand at an angle of 20º, is massively sheathed in lead, with all the
inner details of construction hidden beneath the sheathing. Three iron gudgeons - the first 85 cm from the keel plane, the second 195 cm from it and the third 240 cm away from it - are still in situ, while some planks resting besides the sternpost suggested that they constituted the remains of the rudder. Further analysis of those planks will confirm,
or not, that suspicion.
 |
The lifting of the sternpost. Note the Y-shaped hawse pieces. Photo: Paulo Monteiro |
The gudgeons were 14 cm sided, projecting outside from the sternpost about 13 cm. Gudgeon orifice was 7 cm in diameter. The sternpost had, in all, 19 hawse pieces. The structural integrity of the sternpost assembly permitted its complete lifting as a whole, without having to be dismantled, as soon as the keel and keelson were sawed.
Floor riders
The floor riders, internal frames set atop the ceiling - averaged 30 cm to 40 cm moulded by 25 cm to 30 cm in sided thickness, and were, in general, spaced apart 1.5 meters. They are 2.5 to 3.5 m long. There were at least 3 footwales, also embraced by the floor riders. Two of these footwales were scarfed with the keelson, at the bow section. Floor frames, amidships, had a sided thickness of 20 cm. The presence of riders was a complete surprise since they are typically found only in XVIIIth century warships. The only known example of floor riders in early post medieval ships is found on the Mary Rose.
Keel
Total length of the keel was about 25,5 meters. It was 45 cm sided, below the mast step, and had 30 cm moulded. The upper surface of the keel was completely sheathed in lead. The severe dislocation of the 3 different parts that composed the keel - as well as the fact that the sternpost assembly was imposed over the east end of the keel - made the in situ measurements slightly uncertain. Total exact length of the keel will be obtained as soon as the keel will be re-assembled on the third phase of the project.
 |
Keelson and keel section. Sternpost section. Photo: Paulo Monteiro |
The collapsing of the ship, during the shipwreck process, caused a severe list of the sternpost to the starboard with the rupture of the first keel scarf. The second keel scarf was severely twisted and bent but the keel was not broken, causing only a severe listing to the vertical of the second segment of the keel, the one which included the main mast step. At that location, the keel measured 44 cm sided with 40 cm moulded. An interesting detail of the keel was that its top face, where it contacted the floors, was also completely sheathed in lead.
Keelson
The keelson, also totally preserved, was notched to fit over the floors. It measured, behind the mast step, 28 cm sided and 32 cm moulded - total height where it was not notched, 25 cm at the notches - and it featured two keyed
hook scarfs, one below floor rider 4 and another between floor riders 8 and 9. The keelson enlarged to 46 cm sided, in order to form the mast step on which stepped the mainmast.
 |
Keelson scarf. Preliminary phase of the excavation. Photo: Paulo Monteiro |
The keelson showed two mortises - 25 cm long, 5 cm sided and 4 cm deep, between floor riders 3 and 4, separated by 80 cm - and another one, just forward of floor rider 6, with the same general dimensions.
Floors
The average floor side thickness, at the keel, was 21 cm - minimum 15 cm, maximum 25 cm - with an average moulded height of 19 cm. The average center spacing was 22 cm, equal to the sided thickness of the floor frames. Limber holes were dead center over the keel and averaged 6 cm high by 4 cm wide.
 |
Keelson section over stern frames. Photo: Paulo Monteiro |
The master frame (#101) was the larger floor, measuring 24 cm sided, with futtocks on both the aft and forward face of the floor, joined to the frame by a dovetail-mortice feature. To either side of the master frame were 7 central floor timbers, all joined to the first futtocks by dovetail scarfs, which lead us to believe that they were pre-assembled before being placed over the keel. The rest of the framing must have been added as planking progressed. If this is true, the floor/futtock sets number 43/44 and 115/116 were the bow and the stern almogamas.
The spaces between the lower ends of the first futtocks and the outer ends of the floors, on the starboard side of the ship, as well as the outer ends of the first futtocks and the lower ends of the second futtocks, were filled by overlapping both joining timbers, so that was a solid belt of wood beneath the stringers. Surviving frames included three berthing deck beams, on the starboard side of the wreck.
Main mast step
Beneath the main mast step assembly, there were five filling pieces, that separated the floors underneath the expanded keelson. The expanded keelson had a mortise to house the heel of the main mast and a chock at the forward end of the mortise to keep the foot of the mast from shifting. The mortise was 61 cm long and 20 cm wide.
 |
Photomosaic of the main mast step, after the removal of the pump well box. Photo: Miguel Correia |
Five pairs of perpendicular buttresses, 20 cm sided and 90 cm long, laterally supported and reinforced the mast step in the immediate vicinity of the enlarged portion of the keelson. These buttresses were lying over the floors and some of them had scarfs that fit into notched stringers on their outboard ends.
At the base of the mainmast mortise was a drilled hole on the floor (#107) right beneath the mortise, which may have been intended for a drain hole to allow sea water to escape into the bilge. There were two pump wells to house the shafts of the ship's bilge pumps, one in each side of the keelson, at the stern extremity of the mast step. The two pumps, one on either side of the keel at the lowest part of the hull, had their foot where water collected in the bilge.
Surviving vertical structure included a pump well, fitted around the main mast step, at least two bulkheads, eight limber boards and several stanchions in situ, that fit over the floor riders. The stanchions had a square section, 8 cm sided and had mortises to fit in the tenons on the floor riders.
 |
Detail of a stanchion. |
The pump well was 3.4 square meters in area and was almost box like in structure, being 1.8 m long on the keel axis and 1.6 meters wide from side to side, centered on the main mast step. The vertical pump well timbers were 2.5 cm thick.
 |
Pump well box. Preliminary phase of the excavation. Photo: Paulo Monteiro |
Sheathing
One method of protection against the shipworm borers was to nail strips of lead to the hull. The hull of Angra D was completely sheathed with lead strips, some of which show weave impression from textile cloth, that might have been pressed between the lead and the hull. All the loose strips have holes left by sheathing nails. After the removal of the outboard planks, a second layer, now of lead, was found beneath, composed of all the lead strips that, still attached to one another, had rusted away from the hull.
Planking
Outboard planking average 5 to 8 cm thick, with sided dimensions varying between 28 and 33 cm. Footwales are 13 cm sided and 14 cm moulded. Ceiling planks were 5 cm thick and were around 4.3 meters long, with a sided dimension 35 cm.
 |
Outboard plank and lead sheathing. Preliminary phase of the excavation. Photo: Paulo Monteiro |
Fasteners
Fastening was done by means of two different kinds of iron nails, both with a square cross-section that could be either 1 cm or 1.5 cm sided. Some treenails were located, in the outer hull planking, 2.5 cm in diameter.
Artifact distribution
Artifacts recovered are representative of the daily life on board a ship of the end of the XVIth century or the beginning of the XVIIth century. Several of these, like a brass fitting for a trunk, a copper pitcher, two staved buckets - one with the rope handle still attached - several wicker baskets and barrel staves and bottoms, represent containers for cargo or ships's supplies.
 |
Excavation of a staved bucket. Photo: Paulo Monteiro |
Others once formed part of the ship rigging like a a two block pulley and several lengths of rope and cable found. Others represent past cargoes, like the mercury puddles found on the hold of the ship or ship's daily life, like a Our Lady figurine carved in resin, or the presence of a thimble with some pins in context, too.
 |
Staved bucket stabilized. Photo: Miguel Aleluia |
Mammal bones recovered include cow, hog, lamb and pig. Chicken and fish bones also were found in the shipwreck. Coconut shell, almonds, seeds, raisins and corn represent the other dietary elements of the on board life. Cockroach wings and exoskeleton remains, as well as bones and skulls of rats are the remains of the unwanted stowaways that competed with man for food resources.
Hundreds of fragments of ceramic containers were found in the hull, the majority of them being several rims and shreds of olive jars, both glazed and unglazed on the inside. Two of the rims had the "IHS" sign from the Jesuistic Society, very much like the ones found at the Atocha wreck site. White and blue porcelain shreds were also found, with the most significant of those being identified as a Ming dynasty porcelain, dating from the 1550´s onward.
 |
Fragment of Ming porcelain. Photo: Paulo Monteiro |
Over 150 milliliters of mercury, were found all over the hull, with metallic puddles concentrated at the mast step assembly and amidships to the portside. The metal was recovered with syringes and stored for further analysis. The presence of mercury in the bilge of Angra D suggests that the vessel was carrying - or had in the past carried - a cargo which included quantities of that metal, which was used mainly as a component for the extraction of silver and gold in Central and South America.
Several items of the ship's ordnance were recovered, including ammunition for the ship's artillery, which included a stone cannon ball, and four iron cannonballs, recovered right at the mast step, suggesting that the shot locker was located there.
Small arms ordnance was represented by the recovery of three wooden gunpowder boxes, triangular in shape, one of those being the one used for carrying the priming gunpowder. A wooden stock for the barrel of a musket was also excavated as well as several dozens of lead shot, circa 2 mm in diameter.
 |
Gunpowder box in excavation. Photo: Paulo Monteiro |
Unfortunately, no coins or other dated materials, with the exception of the Ming dynasty porcelain, were recovered. Olive jar chronology point to a late sixteenth century or a earlier seventeenth century wreck, a fact supported by the white and blue ceramics found. Also, the presence of a stone shot cannon ball also points to the same period in time. No intrusive materials were found, except the iron concretions of the Run'Her, earlier mentioned.
Conclusions
Examinations of elements of the hull remains, and measurements of various parts such as futtocks and floor timbers suggest a very large sailing vessel. Size of frames, floor timbers and futtocks, as well as distance and space between frames suggest a ship of between 400 and 500 tons displacement, and an overall length of between 35 and 40 meters. Field studies provided critical information on the vessel type, period of use, origin, and function for this shipwreck. Investigators also gained some insight into the site's layout, makeup, and surrounding environmental conditions.
The materials recovered and the naval techniques used in the building of the ship provide a late sixteenth-century or early seventeenth century date for the shipwreck, and the overall artifact assemblage, most notably the mercury, matches what might be expected from an Iberian ship of that period.
Several vessels are presently known to have sunk in this area during the second half of the XVIth century and the first half of the XVIIth century, according to historical documents.
In 1542 the portuguese nao Grifo, captained by Baltazar Jorge Pais was lost in Angra bay. In 1555 the portuguese naos Assumpção - captained by Jácome de Mello - and Algarvia Velha, were also shipwrecked in Angra. In 1583, three portuguese patachos were, too, lost there, in a storm.
In 1586, the Santa Maria, a Spanish nao from Santo Domingo, also run aground due to a storm, that also sunk, on the following day, the nao capitania from the same armada. On the same year, two more Spanish naos were wrecked, one of them being the Nuestra Señora de la Concepción, captained by Juan de Guzman. In 1587 the portuguese galleon Santiago, coming in from Malaca and captained by Francisco Brito Lobato, was also run aground by a southeast storm.
In 1589 another galleon from Malaca was wrecked inside the bay as well as another Spanish nao. In 1590, a biscayan ship was wrecked and, a year latter, in 1591, another Spanish ship run aground on the bay.
In 1605 the portuguese nao captained by Manuel Barreto hit some sunken reefs at the entrance of the harbor and was also sunk. In 1618 the nao São Jacinto was wrecked and salvaged by the town people. In 1642 a carvel was sunk by portuguese artillery fire, inside the bay. In 1649, four unidentified ships were run aground by a storm and in 1650 a Spanish nao was again wrecked, in the bay.
The majority, if not all, of these wrecks were of Iberian construction. A number of XVIth century wrecks were discovered on the past several years, all of them sharing some naval shipbuilding characteristics with Angra D. These wrecks include the San Juan, in Red Bay, Canada, the Rye A vessel, in Sussex, England; theCattewater wreck, in Plymouth, England; the San Esteban, in Padre Island, USA; the Molasses Reef Wreck, in the Turks and Caicos Island; the HighBorn Cay wreck, in the Bahamas; the Aveiro A wreck; in Portugal and the Nossa Senhora dos Martires; also in Portugal.
citado por Constança Duarte Gonçalves
