?C ..' ~~:.-
LI , /1
* t .:
The INA Quarterly
Volume 30* No. 4 Winter 2003
Institute of Nautical Archaeology
P.O. Drawer HG
College Station, TX 77841-5137
3 The Model Reconstruction of La Salle's Ship La Belle Learn firsthand of the latest discov-
Glenn Grieco series in nautical archaeology. Mem-
bers receive the INA Quarterly and
19 The First Season of the Dominican Republic Survey Project other benefits.
A Preliminary Survey of the Southern Coast Researcher (students only).... $25
Katie M. Custer and Sara G. Hoskins Diver.................... $40
Seafarer. ................. $75
23 Just Released: Surveyor ...................$150
Sea Power in the Medieval Mediterranean: Restorer ............... $500
The Catalan-Aragonese Fleet in the War of the Sicilian Vespers Curator ................ $1,000
Lawrence V. Mott Excavator ................ $2,500
Navigator ............... $5,000
24 Review by Filipe Castro Checks, in US. currency, should bemade
Sealed by Time, The Loss and Recovery of the Mary Rose payable to INA.
by Peter Marsden
26 Looking Back, Looking Forward
Donny L. Hamilton
27 Vol. 30 Index
On the cover: One of the models used to investigate the construction of the French ship La Belle, wrecked in 1686 and
recently excavated by the Texas Historical Commission. Photo: G. Grieco.
December 2003 by the Institute of Nautical Archaeology. All rights reserved.
INA welcomes requests to reprint INA Quarterly articles and illustrations. Articles for publication should be submitted in hard copy and on a 325
diskette (Macintosh, DOS, or Windows format acceptable) along with all artwork Please address all requests and submissions to the Editor, INA
Quarterly, P.O. Drawer HG, College Station, TX 77841-5137; tel (979) 845-6694, fax (979) 847-9260, e-mail email@example.com or firstname.lastname@example.org.
The Home Page for INA is at http://ina.tamu.edu
The Institute of Nautical Archaeology is a non-profit scientific and educational organization, founded by George F. Bass, Michael Katzev,
and Jack Kelly and incorporated in 1972. Since 1976, INAhas been affiliated with Texas A&M University, where INA faculty teach in the
Nautical Archaeology Program of the Department of Anthropology. The opinions expressed in Quarterly articles are those of the authors,
and do not necessarily reflect the views of the Institute.
The TNA Quarterly was formerly the INA Newsletter (vols. 1-18).
Editor: Christine A. Powell
The Model Reconstruction
La Salle's Ship La Belle
Fig. 1. Early representation ofa barque longuefrom the album ofJean Jouve dated 1679.
In April 1997, the Texas Historical Commission com-
pleted the excavation of a small French ship named La Belle.
Built in 1684 and possibly given as a gift to Robert Cavalier
de la Salle from King Louis XIV of France, the little ship La
Belle was one of four vessels that left France in an ill-fated
attempt to found a colony at the mouth of the Mississippi
River. In 1686, two years after leaving France, the vessel ran
aground in Matagorda Bay on the Gulf Coast of Texas. The
wreck of La Belle not only provides a unique example of a
poorly-documented ship type, but also serves as an early
example of many new design and construction techniques
in use in the French shipyards. Although only about a third
of the hull has survived, sufficient information can be ex-
tracted to reconstruct the appearance of the original vessel.
The archaeological evidence, in conjunction with naval doc-
uments, contemporary models and drawings, and firsthand
accounts of its appearance, can be used to answer questions
about its assembly and the shipbuilding practices of the sev-
enteenth century. What type of ship was this? How was it
designed and built, rigged and outfitted? For an oceangoing
vessel, La Belle was very small, but the techniques used to
build it were analogous to those used for larger ships, pro-
viding a concise and manageable example for the period.
This article documents the construction of two mod-
els of La Belle. The first model, constructed during the ear-
ly phases of the reassembly of the archaeological remains,
was used to determine the correct lines and arrangement
of the hull. As conservation of the remains progressed, new
discoveries were made that contribute to our knowledge
of its appearance. The second model incorporates these new
findings. The purpose of the models is not only to repre-
sent the overall appearance and layout of the original ship,
but also to recreate the methods and sequence of its con-
Evolution of the Barque Longue
Before an accurate set of lines could be reconstruct-
ed, it was necessary to determine the type of vessel repre-
sented by La Belle. It is widely believed that La Belle is an
example of a vessel type called a barque tongue, but what
was a barque longue?
The term barque longue applies to a range of small ves-
sels first recognized as a distinct type by the French navy in
1675. The classification of barque tongue was officially replaced
by corvette in 1676. However, vessels in both these classes
vary in size and complexity. Between the 1670s and the mid-
eighteenth century, the barque tongue grew and evolved into
what was later considered a sloop of war or light frigate.
Perhaps the earliest representation of a barque longue
is from the album of Jean Jouv6 dated 1679. This small,
undecked vessel carried only a simple two-masted rig and,
at most, a few swivel guns as armament (fig. 1). This was
INA Quarterly 30.4
a~c r ~dc~;~
certainly not avesselintended for a trans-
atlantic crossing. Eventually, with the
addition of a deck, these ships began to
carry light four-pounder carriage guns.
The only official record of La Belle
refers to it as a barque of forty to fifty tons.
Joutel's journal provides the most com-
plete description of La Belle. Referred to
as a bark throughout the journal, Joutel
introduces La Belle as "a little frigate, car-
rying six guns." Falconer's Marine Dic-
tionay defines a bark as "a general name
given to small ships: it is however pecu-
liarly appropriated by seamen to those
which carry three masts without a miz-
zen top-sail" It is difficult to determine
whether the definitions and translations
provided by Falconer's dictionary can be
applied to a vessel built a century earli-
er. However, there is less ambiguity in Fig. 2. Drawing of a
Joutel's use of the word frigate. One of
the defining features of frigates is the use
of a three-masted ship rig. Joutel later mentions a collision be-
tween l'Amiable and La Belle in which "the vessel La Belle would
have been in danger of perishing, but escap'd with the loss of its
mizzen, which came by the board."
Another eyewitness account from the Spanish sailor Juan
Enriquez Barroto provides further evidence that the vessel had
three masts. Coming upon the remains of La Belle in 1687, one
year after it ran aground, Barroto states: "On the beach was found
the other gun carriage and the main yard, which was measured
and found to be sixteen cubits. We brought this yard and that of
the fore topsail for making oars, and from that of the foresail
boom was made four oars. Captain Pedro de Yriarte took
that of the mizzen also." The mizzen yard would have car-
ried a lateen sail and would be easily distinguishable from
the yards for the fore and main masts.
If these eyewitness accounts are accurate, it is diffi-
cult to refute the assertion that La Belle carried three masts.
As a commander on board Le Joly, Henri Joutel had suffi-
cient knowledge of French ships to give a reliable descrip-
tion of the rigging of La Belle. This is significant, for it provides
the only evidence that the barque longue had evolved into a
form that could be considered a light frigate.
What evidence do the archaeological remains hold to
indicate how La Bellewas rigged? Remains of mast steps forboth
the foremast and the main mast have survived. However, the
absence of a mizzen mast step among the recovered material
does not rule out the possibility of a mizzen mast
Taking into consideration the steep rake of its stem-
post, the proposed reconstructed length of La Belle places
the main mast slightly forward of the longitudinal center
point. No reliable contemporary representations of two-
masted vessels place the main mast so far forward, sug-
light frigate from the album of Gu.roult du Pas.
gesting that a third mast was needed to balance the rig. This
would be particularly true if the main mast was square
rigged, as indicated by Baroto's reference to the main yard.
Thus the preponderance of evidence, both historical and ar-
chitectural, points to La Belle being a three-masted vessel.
From Joutel's journal it is known that La Belle was armed
with six guns. It appears that six four-pound cannon composed
the typical armament for a barque Iongue of this size.
When the remains of La Belle were first encountered by
two pilots from Baroto's ship, they reported finding "a lost ship
that has three fleurs-de-lis on its poop; six pieces of artillery,
mounted, woolded, and hove down; two iron swivel guns with-
out chambers, which they brought in our canoe." From this de-
scription, it appears that there were six carriage guns on deck
secured with lashings. Baroto himself observed five swivel guns
the next day "that fire a ball up to four pounds.. still upon their
carriages, lashed to the side of the ship." The discrepancy in the
number of observed carriage guns may indicate a misuse of the
term carriage in the second passage. One swivel gun recovered
from the wreck had a bore diameter cose to that of a four-pound
ball. It is possible that the swivel guns would have been lashed
to the sides of the vessel when not in use. While discussing the
ordnance, he states that the hull has "eight portholes and as many
other flues." If the wordflue indicates the pipe or post attached
to the side of the vessel to support a swivel gun, then seven of
eight guns were accounted for at the time. The single swivel
gun recovered by the Texas Historical Commission could be
the eighth gun that fell from the side of the hull before the ship
was found by Baroto.
What of the eight portholes mentioned by Baroto? It was
common for vessels to have more gunports than the actual num-
ber of guns on board. Cannon could be moved from one side of
INA Quarterly 30.4
Fig. 3. Reconstructed lines ofLa Bellefrom data gathered during the excavation and disassembly. Courtesy Texas Historical Commission.
the vessel to the other to increase the firepower of a broadside or
to transfer weight to adjust the vessel's trim while sailing.
All evidence indicates that La Belie was a frigate, just as
Joutel described. Its appearance may have been similar toa light
frigate illustrated by Guioult du Pas in 1710 (fig. 2).
Reconstructing a set of lines
A large amount of data is available for a reconstruction
of the lines of La Belle. The dockyard manuscript prepared in
December 1686, two years after it was built, gives many of the
overall dimensions needed for its reconstruction A set of recon-
structed lines based on the archaeological remains, drawn by
Greg Cook for the Texas Historical Commission (fig. 3), pro-
vides a good first impression of the shape of its hull
Length of keel = 45 feet
Length from stem to stern = 51 feet
Height of stem = 12 feet
Height of sternpost = 11 feet
The first step in producing a set of lines for La Belle was to
determine the shape of the midship section- Fortunately, a large
portion of the starboard side of the hull at midships has sur-
vived. Adjusted so the centerline of the frame is vertical, the re-
mains of the midship frame form the basis of the reconstruction.
In Joutel's description, the draft of La Belle is given as seven feet
(2.268 m). If this is accurate, the archaeological remains repre-
sent almost the entire midship section up to the waterline. Fig-
ure 24 shows a representative section near midships that
combines the archaeological and archival data.
Once the shape of the midship section was deter-
mined, the next step was to define the longitudinal profile
of the vessel (fig. 4). The length of the keel was known
from both the archaeological remains and the dockyard
Depth in hold = 7 feet 6 inches
Rake of stem = 4 foot 6 inches
Rake of sternpost = 1 foot 6 inches
Height of midship section = 10 feet 6 inches
Drawing: C. Crieco
Fig. 4. Reconstructed longitudinal profile incorporating dimensions from the Rochefort dockyard manuscript.
INA Quarterly 30.4
K..~ b .
-g -I Cg -I W Va aDd I U 1 I
~* S. K A0 S KS K K K
I1 - - - I
uw5 I "T
I _ _ ___ _
~~ __ __
12 V 'J_
6 ~ ~ ~ l ,
r -- -
Fig. 5. Reconstructed set of lines used to build the models,
manuscript, and the rest of the hull was reconstructed from
the remaining dimensions provided in the document. It
appears that the overall length of the vessel was measured
from the rabbet of the stem to the rabbet of the stempost at
about the height of the deck. The value for the rake of the
stem describes the distance between the forward end of
the keel and the forward end of the overall length. The
rake of the sternpost is measured from the after end of the
keel to the aft perpendicular of the vessel. Knowing that
the after face of the stempost is raked seventy-four degrees
and that the rake measures one foot, six inches (48.8 cm), it
was calculated that the height at which the length was tak-
en was approximately eight feet, three inches (2.683 m).
This corresponds closely to the height of the deck. By trans-
ferring this height to the stem and using the known value
for the rake of the stem of four feet, six inches (1.463 m) at
this point, the curve of the bow was established. The pro-
file of the bow was completed by continuing the curve to a
height of twelve feet (3.902 m).
To complete the lines of the stem, the shape of the
transom was determined. The height of the wing transom
is given as nine feet, four inches (3.035 m), but no other
dimensions are provided. In order to determine the length
of the wing transom, the French naval regulations of 1673
were examined. The document mandates that the wing
transom should be two-thirds of the extreme breadth, or
ten feet (3.252 m), if the extreme breadth is measured to
the outside of the midship frame.
For the reconstruction, the shape of the vessel was as-
sumed to be a sloop or light frigate. The main deck ran from the
bow to just aft of the mizzen, where a break in the deck of two
feet (sixty-five cm) forms the floor of a small stem cabin. A short
quarterdeck in the stem and a forecastle deck in the bow were
located three feet (1.138 m) above the main deck.
Based on these data, a set of lines were created (fig. 5).
Sections provided in reconstructed lines drawn by Greg Cook
were used to guide the process. Remains of internal bulk-
heads served as templates and provided a means of check-
ing the curvature of these sections. A series of lines were
drawn and manipulated until a fair hull was established,
conforming closely to the reconstructed lines of the vessel.
Reconstructing the Hull
The most valuable source of information for the con-
struction of the models was the archaeological remains of
the ship. The surviving frames and internal structure of La
Belle provide the most reliable evidence of its construction.
For this reason, it was considered desirable to display as much
of the internal structure of the model as possible. Planking the
entire vessel would have obscured important information. In
order to best exhibit the reconstruction of the actual archaeolog-
ical remains, one side of the vessel was left unplanked, allowing
the internal structure to be viewed within the hull. Because a
significant portion of the starboard side of the ship has sur-
vived, this side was completely reconstructed. The port side
was not planked, leaving the frames exposed.
Modeling the Frames
Before construction of the models could begin, accurate
templates were required for each component of the keel, stem,
stempost, and frames. Templates for the longitudinal timbers
were derived directly from the construction drawings. In order
INA Quarterly 30.4
to create molds for the thirty square frame timbers,
additional drawings were required. Using a set of
lines superimposed over the drawing of the fram-
ing plan, the curvature of each frame pair was lift-
ed from the drawing and plotted as a section.
Because each pair was composed of overlapping
timbers, three faces were plotted to establish the
bevel of the external molded faces. Once the out-
line of the frame was plotted, the molded thick-
ness, as measured from the actual timbers, was
used to establish the internal bevels of the frames.
To complete the drawing, the locations of the butt
joints between each timber were marked. Using
these drawings, templates for the individual floors
and futtocks of each frame were obtained.
Paper copies of each template were af-
fixed to pieces of wood, which were then planed
to the appropriate sided dimensions. Each tim-
ber was then sawn to shape and filed to ob- Fig. 6. Rot
tain the correct bevel (fig. 6). Using the body frames. Tw
plan of the hull as a mold loft, the frames were
assembled from the individual floors and futtocks (fig. 7).
The construction of the keel, stem, and apron was rel-
atively straightforward. Fastened together using the appro-
priate scarfs, the only deviation from the original construction
was the use of bamboo dowels in place of iron fasteners. Al-
though the types of fasteners used on La Belle were known,
Fig. 7(below). Lofting the frames using the body plan of the hull as a guide.
Fig. 8 (right). The hawse timbers preassembled and fastened to the forward-
most square frame.
Phntn- C Crincn
ugh cut model frame timbers before beveling and assembly into
o completed frames can be seen in the background.
data regarding their precise locations was not available at
the time that the models were built. To avoid confusion, only
those iron fasteners whose locations had been accurately
determined were represented in the model The remaining
fasteners were represented by bamboo "treenails" using the
general overall pattern of the original fasteners.
INA Quarterly 30.4
A characteristic feature of
French ship construction is the ab-
sence of cant frames. In keeping with
this tradition the frames of La Belle
are all square frames. To fill in the
curvature of the bow forward of the
first square frame, an assembly of
hawse timbers creates a nearly solid
wall of timber up to the stem (fig. 8).
Bolted together laterally, and fas-
tened at their heels to the forward-
most frame, these timbers were often
reduced in their sided dimension
over part of their length to permit air
circulation between them
The sequence of raising the
frames on the models may be a bit
of a departure from the actual se- Fig. 9. The two wmpleted frame models. Photo:
quence. While constructing the
models, the midship frame was raised first. Then, every third frame forward and aft was fastened into position on the keel.
After every third frame was erected, the intermediate frames were inserted. Recent observations of the remains of La Belle
suggest that the floors and futtocks of the intermediate frames were installed separately after the other frames were erected.
Figure 9 shows the two completed frame models. From this point, the construction of the two models varied in many
ways. A complete description of the construction of each of the models would be repetitive. Instead, this study will focus on
the construction of the second model, highlighting the lessons learned from the construction of the first when appropriate.
The remains of at least fourteen strakes of planking were excavated on the starboard side of the shipwreck. The reconstruct-
ed midship section indicates the likely existence of one more plank below the wale. If this is the case, the seams of all the lower
planks are represented at midships. The widths of the inner faces of the hull planking were recorded for nine frames at the time that
the hull remains were disassembled. Using the planking width, the seams of each plank were transferred to the appropriate frame
on the modeL The locations of scarfs, stealers, and drop strakes were marked on the outside of the frames. Using a batten, it was
simply a matter of connecting the dots to deter-
mine the run of the individual stakes. The scarfs
and plank ends then fell easily into place within
the strakes. In reconstructed areas not represent-
ed by archaeological remains, the batten was al-
lowed to run fair to complete the strakes. By
allowing the shape of the hull to determine the
run of the planking, remarkably uniform hood
ends were obtained in the bow and stem (fig. 10).
Strips of poster board were used as
spiling battens. With the batten damped to the
frames, the contour of the exposed edge of the
previous plank was transferred to the batten
using a compass. The batten was then used as
a template for the next plank.
The fair run of the planking allowed the
use of straight-sided planks. Where the edges of
adjacent planks diverged, stealers were used. The
capacity to use straight, almost parallel-sided
planks minimized waste and increased the effi-
Photo: G. Grieco ciency of the planking. The ease with which the
Fig. 10. Fair lines drawn through the points marked on the frames. hull was planked and the similarity between the
INA Quarterly 30.4
shapes of the spiles and the original planks
indicate that the reconstructed shape of the
model is dose to that of the actual vesseL
Some difficulties were encoun-
tered during the first attempt at install-
ing the ceiling on the model. Although it
had been assumed that the process began
at the keelson and proceeded outward, it
quickly became evident that this was not
the correct sequence. The inner strakes
feathered out to sharp points at the bow
and sten, where they encountered the
four outer strakes, indicating that they
were installed last. In the case of La Belle,
an outer notched strake was inserted first, t ric
followed by two ceiling planks. A second
stringer was then notched to fit like the Fig. 11. Notches cut in forward and after faces of the frames for the filler pieces.
outer timber. With these four strakes in
place, the remaining planks were installed, working from the keelson outward.
The filler piece that sealed the edges of the ceiling was set into shallow beveled notches in the sides of the frames (fig. 11).
Tool marks around the notches indicate they were sawn using the outer edge of the ceiling stringer as a guide. This could not have
been accomplished if the external planking was in place at this height. The hull was planked either ceiling-first or simultaneously
inside and out. Only iron spikes were used to fasten the ceiling planks until the outer planking was attached. Treenails were then
used to fasten the outer hull planking from the outside through to the ceiling. As it was difficult to control the direction of the auger
when boring holes for the treenails, the treenail pattern on the external planking is more uniform than that on the ceiling.
The tight fit between the filler pieces and the external planking
suggests two possibilities. Either the filler pieces were installed prior to
the planking and faired to the depth of the adjacent frames, or the outer
plank was attached after the notches were cut, with considerable care
taken to shape the filler pieces to the complex contour of the inner faces
of the frames and planking. Judging from the tight fit between the fill-
er piece and the external planking, the first method seems more likely.
Figure 12 shows these filler pieces prior to installation, and figure 13
Fig. 12 (left). Filler pieces cut tofit around the frames. Fig. 13 (right). Filler pieces in place along the edge of the outermost ceiling
plank. The riders of the mast step can be seen in the lower right corner. Photos: G. Grieco
INA Quarterly 30.4
n"iiu0: 1J. -iItlO
Fig. 14 (left). Forward and aft rider timbers of the main mast step. The notches in thefaces of the riders will accept the ends of the
mast step partners.
Fig. 15 (right). Installation of the mast step partners.
shows them in place, bordering the ceiling plank on the
model. The riders of the mast step can be seen in the bot-
tom right comer.
Aside from slight erosion at the ends of the port arms
of the floor riders, the main mast step has survived in its
entirety. Figure 14 shows the two rider timbers prior to
installation over the ceiling. The curvature of the lower face
of these timbers was taken directly from the archaeologi-
cal remains. Because they fit snugly against the ceiling in
the model, these timbers helped to verify that the shape of
the hull in this area was accurate. Figure 15 illustrates the
way in which the beveled ends of the mast step partners
locked tightly between the notches in the riders, prevent-
ing lateral movement of the heel of the mast. Figure 16
shows the complete mast step assembly with buttress tim-
bers and filler pieces inserted between the ends of the part-
Although only part of the foremast step has sur-
vived, the symmetry of the structure permitted a full re-
construction (fig. 17). A plank inserted between the notches
in the opposing arms supported the aft face of the heel of
Fig. 16 (left). Completed main mast step assembly.
Fig. 17 (right). Heavy breast hook in the bow serving also as the step for the foremast.
INA Quarterly 30.4
Fig. 18. A. Internal structures be-
fore the installation of deck beams.
B. Installation of deck beams re-
quired by internal structure.
Nothing remains of the original deck structure of La
Belle. Without the benefit of archaeological remains, the
scantlings of the timbers are difficult to determine. Care-
ful examination of models, drawings and treatises such as
the Album de Colbert and the Album del Marquez de la Victo-
ria helped to determine the relative sizes of the individual
timbers forming the deck framing. Once the sizes of the
deck beams were estimated the scantlings of the other com-
ponents of the deck were calculated.
The first step in installing the deck was to determine
the run of the deck clamps. Estimating the height of the
clamps from the deck heights on several early eighteenth-
century drafts, it appears that on a vessel with one contin-
uous deck from bow to stem, the upper face of the clamp
is typically flush with the upper edge of the lower wale at
midships. This point was established for the wale during
the reconstruction of the midship section. Running paral-
lel with the top of the lower wale through the waist of the
vessel, the clamp begins to diverge toward the stern as it
rises on the inside faces of the futtocks to butt against the
forward face of the wing transom. However, if La Belle was
a sloop of war or light frigate, its main deck would not
have run all the way to the stem. Olivier made the follow-
ing observations concerning sloops of war of twelve guns:
"Such vessels should have but a single deck, with the guns
only as far as the mizzen mast, where the upperworks
should be raised two feet, with a quarterdeck three feet
above the upper deck. In the event that we should build
sloops of war of ten, eight, six or four guns, they should be
fitted out like those of twelve guns." A break in the deck
just aft of the mizzen mast, with a deck two feet (sixty-five
cm) below it and a quarterdeck three feet (1.138 m) above,
suggests the presence of a stern cabin. For this reason, the
main deck clamp would have supported deck beams back
to frame XIID. From XIID aft, the deck beams of this small
stem cabin would have been supported by the ceiling plank
two feet (sixty-five cm) below the clamps.
Having less sheer than the wales, the clamp typical-
ly dips below the level of the lower wale in the bow. This
configuration varies from ship to ship. Ships often had a
small bulkhead just aft of the hawse holes to catch water
draining from the anchor hawser. Forward of this bulk-
head was a scupper with its outboard end centered be-
tween the two wales. If the damp dipped too low, this area
would not drain properly. To avoid this problem, the deck
clamp was positioned at a height that allowed a drainage
angle of about thirty degrees from the top of the deck
Next, the locations of the deck beams were deter-
mined. Figure 18A indicates the deck beams whose pres-
ence is indicated by the surviving internal structures. The
lower image, 18B, shows the beams that supported the stan-
chions of the three bulkheads and the pump well, as well
as the beam that supported the upper end on the notched
stanchion amidships. The upper image in figure 19A illus-
trates the addition of beams to support the fore, main, and
mizzen masts. Forward of the first bulkhead and the
notched post, beams were added to frame the forward and
main hatches. The hatch aft of the pump closet required
another beam. The lower image shows the addition of a
beam aft of the foremast to support the bitts of the wind-
INA Quarterly 30.4
Fig. 19. A. Installation of deck beams
necessitated by hatches, mast part-
ners and windlass. B. Installation of
three remaining deck beams to bridge
the remaining spans of deck.
lass. Three additional beams were positioned to break up
spans of more than three feet (97.6 cm). In order to tie the
clamps together at the bow and to add lateral strength, a
breasthook was notched down into the clamps and bolted
through the futtocks. With molded dimensions equivalent
to those of the deck beams, the breasthook also served as a
fastening surface for the forward ends of the waterways
Construction of the deck framing began by dove-
tailing the deck beams two inches (5.4 cm) into the clamps
(fig. 20). The dovetail joints increased the lateral strength
of the hull and tied the sides of the vessel together. With
all the beams placed in their proper positions, the loca-
tions of the hatch carlings and mast partners were marked.
Fig. 20 (left). Dovetailing of beams into the beam clamps.
These timbers had the same dimensions as the deck beams
and were set into beveled notches in the fore and aft faces
of the beams. Next, the locations of the carlings, windlass
bitts, main bitts, and bowsprit step were determined.
Next, the beams were reinserted into the clamps. Fill-
er pieces measuring two inches by four inches (5.4 cm by
10.8 cm) were inserted between the dovetails in the beams
to fill the space between the top of the clamps and the bot-
tom of the waterways. The two waterway timbers were
then placed over the ends of the beams (fig. 21). The wa-
terway timbers measured eight inches by four inches (21.7
cm by 10.8 cm) and were notched over the deck beams to a
depth of two inches (5.4 cm). With the waterways in place,
the carvings and ledges were inserted.
Photos: G. Grieco
Fig. 21 (right). Short sections of the inner waterway timber notched to fit over the deck beams.
INA Quarterly 30.4
On both models, only the starboard side of the deck was planked.
This not only permits the inspection of the internal construction fea-
tures, but also allows the details of the deck framing to be seen. Due to
the absence of archaeological evidence, the deck planking pattern on
La Belle will never be determined for certain. The deck planking of the
model approximates the appearance and planking patterns of models
from the seventeenth and early eighteenth centuries.
Thin strips were cut from black plastic binder cover sheets to
insert between the planks to represent the tarred seams. Each plank
was fastened to each deck beam using two one inch (2.7 cm)diameter
treenails. Figure 22 shows the completed deck planking.
Before the gunports were cut and framed, their locations were
determined. Many factors were considered in determining the placement
of cannon on a narrow-decked vessel like La Belle. Structures such as hatch
coamings, masts, pumps, and bitts, can prevent them from being with-
drawn from their ports. Outside of the hull, the location of the fore and
main mast channels can dictate the placement of gunports. The breadth
of the vessel is also a limiting factor. The barrel of a four-pounder cannon
is six feet (1.951 m) long. With a deck less than fourteen feet (4.553 m)
wide and a foot of tumblehome in the waist, two cannon barely fit abreast
with the port lids dosed. Forward and aft of midships, the narrowing of
the deck would prevent this arrangement entirely unless the guns were
drawn in at an angle. Consideration must also given to the gun crews
who fired the guns. Seventeenth-century French naval regulations re-
quired a minimum spacing of six feet, six inches (2.114 m) between the
guns on larger warships to provide room for the crews to service them.
The reconstructed breadth of the gunports was calculated to
be eighteen inches (48.8 cm) square. This calculation was later con-
firmed by a single gunport lid found during the excavation of La Belle.
The frames were erected on eighteen-inch (48.8 cm) centers, conve-
niently allowing the top timbers of two adjacent frames to be used to
frame the sides of the ports.
Fig. 22. Deck planking on the starboard side of
F ,t Fed Drawing: G. Grieco
Fig. 23. Longitudinal profile of second La Belle model showing placement of three gunports on each side.
[NA Quarterly 30.4
Two possible configurations were for-
mulated. The first configuration relied on En-
riquez Barroto's description of the hull and
assumed the use of eight gunports. Employed
on the first mode), this arrangement allowed the
forward two pairs of guns to be secured inboard .
with the gunports closed, but the space avail-
able on deck between the aftermost pairs of
ports was not sufficient for two guns to be posi-
tioned abreast. This suggests an eight port con-
figuration using only six guns. With four guns
in the forward four ports, the two aftermost
guns could be used one per pair of ports and
moved from side to side as necessary. The sec-
ond configuration assumed that only six ports
were present (fig. 23). In this arrangement, space
would have been tight between the aftermost
pair of guns. However, with a slight angling of
the carriages, both guns could have been run in Fig. 24. Sc
when the ports were closed, side of the
Once the locations of the ports were
estimated, the top timbers at the port openings were cut to
a height of eleven inches (29.8 cm) above the deck. A three
inch (8.1 cm) thick lintel raised the sill to the required height
of one foot, two inch (37.9 cm). A second timber of the same
dimensions formed the upper edge of the gunport.
Before the bulwarks were planked, several fittings
were installed in the framing above the wales. Scuppers were
needed on La Belle to allow water to drain from the deck.
Typically, ships were equipped with several scuppers in the
waist of the vessel and in the manger at the bow. By the
Fig. 25. Close-up photo offairlead block for the main sheets and bra
ing ends of the sheet and brace are spliced to the ring bolts below.
rmoro: ". uneco
uppers composed of bored wood blocks installed on the starboard
early eighteenth century, lead scupper pipes were already
in use aboard ships in the French and English navies. In the
case of La Belle, however, no lead artifacts that resembled
scuppers were found among the archaeological remains. It
is possible that the lead linings were scavenged from the
hull after its grounding, although it is equally possible that
its scuppers were not lead lined. Bored elmblocks with leath-
er linings were commonly used as scuppers during the fif-
teenth and sixteenth centuries. The practice was still in use
aboard English vessels during the late seventeenth and ear-
ly eighteenth centuries and may have been used on vessels
of other nationalities as well. Considering the absence of
evidence for metal linings, wooden scuppers
appear to be the best solution. Contemporary
drawing of small vessels depict five to seven
scuppers in the waist and one at the bow for
the manger. Because cutting a hole for the scup-
pers in one of the wales would have compro-
mised the strength of the timber, contemporary
drawings show the outlet of the scuppers in
the strake between the two wales. Placing the
deck clamp at the height of the lower wale, the
angle from the top of the deck to the outlet of
the scupper is close to forty-five degrees (fig.
24). For the reconstruction of La Belle, five two-
inch (5.4 cm) scuppers were provided in the
waist and one in the bow.
While sailing, several lines of rigging
must run through the sides of the hull. To
assure the proper lead of the sheets and brac-
SG. Geco es, three sets of fairlead blocks were set into
ces. Stand- the hull (fig. 25). Positioned between the sec-
ond and third frames forward of midships, a
INA Quarterly 30.4
single sheave block on each side of the hull trimmed the
fore sheets. Between the second and third frames aft of
midships, a double sheave block is provided for the fore
yard braces. Finally, a double sheave block for the main
sheet and braces is set between the sixteenth and seven-
teenth frames aft.
After these elements were installed, the planking of
the bulwarks was completed. The reconstructed midship
section was used to define the dimensions of the plank-
ing. Externally, the planking was composed of parallel-
sided planks that continued the curve of the wales up the
side of the vessel. Contemporary drawings of small ves-
sels and frigates indicate that the space between the upper
wale and the molding below the waist rail typically con-
sists of three strakes, with a single strake between the waist
rail and the sheer rail, and another between the sheer rail
and the drift rail. By dividing the space between the wale
and the molding, it was determined that three nine-inch
(24.4 cm) wide planks were required. Continuing up the
sides of the stem, two additional nine-inch (24.4 cm) planks
alternating with two three-inch (8.1 cm) moldings provid-
ed the height necessary for the quarterdeck. One additional
plank in the bow provided the height needed for the fore-
castle deck. The reconstructed thickness of the planking in
the upperworks was one inch (4.1 cm).
As mentioned earlier, the armament carried on La
Belle included two types of ordnance: six iron carriage guns
firing four-pound balls, and breech-loading iron swivel
guns. Several artifacts relating to these guns have survived.
Although no four-pounder long guns were recovered from
the wreck, two four-pounders were among the guns exca-
vated from the remains of Fort St. Louis on the Gulf Coast
of Texas. Using scale drawings, models of the barrels were
turned in brass at the scale of one to twelve, and then black-
ened to imitate the appearance of the original iron guns.
Measurements from a single gun carriage recovered from
the wreck were used to reconstruct the carriages for the
model (fig. 26). Interesting features of the recovered car-
riage include the deep mounting holes for the trunnions
and the flat capsquare locking the trunnion in place. It is
more typical to have the trunnion resting in a semicircular
notch half the depth of the trunnion, with the upper sur-
face held in place by a capsquare forged with a correspond-
ing semicircular contour. The reason for the position of
the trunnions on the sides of the gun carriage on La Belle is
A related artifact recovered from the hull consists
of concretions containing a ring bolt and an associated hook
bolt (fig. 27). Clearly intended for securing the gun tackle
to the bulwarks, the conserved artifacts furnish several
important pieces of data. Preserved by the corrosion prod-
ucts of the iron, a large section of the wood through which
Fig. 26. View down the deck of La Belle showing the guns
mounted in their carriages.
Fig. 27. Conserved hook and ring bolt still joined by fragment
of a frame timber. Courtesy of the Conservation Research Labo-
INA Quarterly 30.4
the bolts were fastened has survived. Although the origi-
nal thickness of the wood has diminished, a rove for the
forelock of the hook-bolt has corroded into the position
where it rested against the outer surface of the hull plank-
ing. Neither the rove nor forelock has survived on the ring
bolt; however, the remains of the shank indicate that the
bolt ran through timber approximately two inches (5.4 cm)
thicker than the hook. This suggests that the hook pierced
the outer hull planking while the ring bolt penetrated a
wale. Superimposing a scale image of these artifacts over
the bulwarks of the reconstructed midship section, the an-
gle of the bolts and their lengths provides supporting evi-
dence for the reconstruction (fig. 28).
Also found on the wreck was a loaded swivel gun
attached to its post and mounting hardware (fig. 29). The
conserved gun was 54.55 inches (1.478 m) long from the
muzzle to the tip of the tiller (fig. 30). The muzzle had a
bore of 3.42 inches (9.3 cm), close to the bore diameter of
the four-pounder carriage guns. Although Baroto mentions
that the swivel guns appeared to be capable of firing a four-
pound ball, the breach of the barrel contained a two-inch
ball and the chamber held a powder charge with a wood-
en plug. Primarily used as an anti-personnel weapon, the
gun could also have fired a handful of small shot.
Iron swivel guns were used aboard vessels from the
sixteenth through the eighteenth centuries. Examples of
wrought iron, stave and hoop constructed weapons almost
identical in design and dimensions to the one found on La
Top View Bore View
0 10 20 30 40 50
Inches Drawing: D. L. Hamilton
Fig. 28. Drawing of the bulwarks of La Belle showing how the
hook and ring bolt may have been oriented.
Belle have been found on shipwrecks from as early as the
sixteenth century. The post and mounting hardware still
attached to the gun were more unusual finds. The swivel
was set into a hole in the top of the five-inch (14.9 cm) di-
ameter post. A two-inch (6.1 cm) wide iron band encircles
the top of the post, and four inches (10.8 cm) down from
the band a two-flanged iron strap that went around the
Fig. 29 (left). The swivel gun recovered from the wreck of La
Fig. 30 (below). Photograph of the swivel gun after conserva-
tion. The breech chamber and wedge are not shown in this photo.
Courtesy of the Conservation Research Laboratory.
INA Quarterly 30.4
Fig. 31 (left). One of the larger deadeyes from the wreck. Courtesy of the Texas Historical Commission.
Fig. 32 (right). Topmast fid after conservation. Patterns of wear and discoloration provided clues to the dimensions of the topmast
and trestletrees. Courtesy of the Conservation Research Laboratory.
front of the post and twisted ninety degrees to fix the post to the top of the caprail. Each flange had three holes down the
center for the nails that attached it to the top of the caprail. Examination of many contemporary drawings and treatises
did not reveal any images of a swivel post with this configuration.
At the time the models were constructed, few artifacts from the rigging of La Belle had been identified and
conserved. Therefore, the reconstruction has focused on the hull of the vessel as it would have looked as a bark or
light frigate. The rig employed on the model is a generic rig representative of ships of this class during the late
seventeenth century. Although the heel of the main mast was recovered, not enough of it has survived to deter-
mine its maximum diameter at the deck. Mast dimensions were estimated from the sizes of various deadeyes (fig.
31) recovered from the site. A single topmast fid (fig. 32) provided additional clues for the dimensions of one of
the topmasts. Preliminary dimensions for the spars and rigging were taken from contemporary formulas and
adapted according to the mechanical advantage necessary for the proper function of the component. Where the
dimensions of rigging elements such as deadeyes and certain types of blocks were known, scale copies were
constructed and implemented in the rig.
The primary question of this study was "What is a barque longue and what can La Belle tell us about this
type of vessel?" Existing descriptions of the barque longue were heavily influenced by its long history as a small,
undecked vessel. Every contemporary source describing a barque longue depicts a ship with only two masts. The
problem lies in the fact that La Belle is too developed to fit the accepted definition of a barque longue but too small
to fit the definition of a frigate. With this in mind the problem becomes, "which type of vessel is more similar
If La Belle was a barque tongue, its remains indicate that the development of the vessel type has been
underestimated. Eyewitness accounts of the vessel, as well as the archaeological remains, provide a great deal of
evidence that this ship fit the description of a sloop or frigate in every aspect except size.
Construction of the models was remarkably straightforward, a good indication that many of the assump-
tions made during the reconstruction of the upperworks and rig were plausible. Enough evidence exists to
establish a high degree of confidence that the final configuration of the hull is realistic. Although the rig is
based on only a few surviving artifacts, the results are appropriate in both size and arrangement. This
reconstruction supports the argument that the barque longue evolved into a much more substantial vessel
than previously assumed and that there was little to distinguish this type from the sloops and light frigates
of the French Navy. a
INA Quarterly 30.4
Port of Rochefort Archives, Rochefort, France, cote 1 L3. Registre 19. F1 88.89.
Album de Colbert.
1670 Reprint, Nice: Editions Omega, 1988.
1981 "The Barques Longues," translated by H. Bartlett Wells, Nautical Research Journal 27.
1993 The History of the French Frigate 1650-1850. Translated by David H. Roberts. Rotherfield: Jean Boudriot Publi-
Foster, William C.
1998 The La Salle Expedition to Texas. Austin: Texas State Historical Association.
1966 The Last Voyage Perform'd by de la Sale. 1714. Reprint, Ann Arbor: University Microfilms.
Monceau, Duhamel du
1970 lIdmens de l'Architecture Navale, ou Traite Pratique de la Construction des Vaisseaux. 1755. Reprint, Grenoble:
editions des 4 Seigneurs.
Navarro, Juan Jose
1995 Album del Marquez de la Victoria. Facsimile, Madrid: Museo Naval y Lunwerg Editores.
1992 18*~ Century Shipbuilding. Translated by Davis H. Roberts. Rotherfield: Jean Boudriot Publications.
1998 "A Cache of Cannon: La Salle's Colony in Texas," Southwestern Historical Quarterly 102.
Weddle, Robert S.
1987 La Salle, the Mississippi, and the Gulf. College Station: Texas A&M University Press.
Weddle, Robert S.
2001 The Wreck of the Belle, the Ruin of La Salle. College Station: Texas A&M University Press.
INA Quarterly 30.4
The First Season of the Dominican Republic Survey Project:
A Preliminary Survey of the Southern Coast
Katie M. Custer and Sara G. Hoskins
The island of Hispaniola (fig. 1), currently com-
prising the two nations of Haiti and the Dominican Re-
public, has played a significant role in the history of
European seafaring in the Caribbean. Columbus land-
ed on the northern shores of the island during his first
voyage to the New World and named it Hispaniola af-
ter Spain. He reported back to Spain that it was the "fair-
est land under the sun" and built a settlement there out
of the shipwrecked remains of the Santa Maria. Santo
Domingo (fig. 2), founded four years later by Colum-
bus' brother Bartolomew, quickly surpassed the earlier
settlement to become the island's principal port. The
island was the main center for Spain's exploits in the
first part of the sixteenth century, with Santo Domingo
serving as the administrative center for the New World.
It was here that ships arrived carrying goods from Eu-
rope to sustain newly founded colonies, and from here
the treasures of the New World were brought back to
the European mainland. Spain's interests quickly moved
away from Hispaniola as gold and silver were found in
abundance on the mainland of Central and South Ameri-
ca. Santo Domingo became a target for rival powers and
pirates throughout the remainder of the sixteenth and into
the seventeenth century. A struggle for control of the
island between Spain, France, and England ensued
while Dutch pirates also tried to claim a foothold.
INA Quarterly 30.4
Fig. 1. The Dominican Republic, on the island of Hispaniola, occupies a key position in the Caribbean region.
The frequency of shipping traffic in the
first decades of colonization, the later military con-
flicts, and seasonal hurricanes that tore across the
island, demolishing both city and ship, combined "R.D.
to create a veritable nest of shipwrecks awaiting
discovery. The northern shore of the Dominican
Republic has claimed its share of fame for hiding
the lost ships of Columbus. Archaeological and
salvage teams have spent years trying to unravel
the mysteries of these legendary ships on the am omi
northern coast while the southern coast has been
practically ignored. Cities on the southern coast
were completely destroyed several times by hur-
ricanes, leaving a long list of known shipwrecks ." Southern Coast
there. Several fleets of caravels and naus were lost
in these waters in the sixteenth century along with. P.
many other ships of different sizes and nationali-
Caravels and naus were the primary ships Photo: K. Custer
of discovery for the Iberian Peninsula. They took Fig. 2. The general survey area on the Dominican Republic south coast.
Columbus to the New World and da Gama to In-
dia, forever changing world history. However, lit-
tle is known about these ships outside of literary
references to their attributed sailing qualities and
widespread use. Work is currently underway by one of the au-
thors to collect iconographic evidence of these vessels. The only
remaining task is to find the physical remains of the ships. The
hull of a caravel or nau could provide invaluable information on
Iberian ship construction that has to date been elusive.
The potential for shipwrecks of the sixteenth and seven-
teenth centuries in this area of the island is astounding. The sheer
number of lost ships and known wrecks of caravels and naus lends
credibility to archaeological study. A remote-sensing and visual
survey of the southern coastline is being carried out by a team of
graduate students in the Nautical Archaeology program at Texas
A&M University. The objective of the survey is the creation of a
preliminary database of wrecks in the area with the specific in-
tention of finding sixteenth and seventeenth century Iberian ships,
the most prevalent of which are caravels and naus, in hopes of
filling in the gaps of Caribbean maritime history.
With generous help from RPM Nautical Foundation, the
support of the Institute of Nautical Archaeology, and the permis-
sion of the Officina Nacional de Patrimonio Cultural Subaquatico,
the team was able to run the first field season of the Dominican Re-
public Survey Project from June 3 to June 28, 2003. The goals of this
preliminary season were to obtain local information of shipwrecks
in the area, gain working knowledge of the physical environment,
launch the survey, and assess the practicality of the equipment and
methods used. On all accounts, this season was a success.
Photo: K. Custer
The survey was conducted using a Geometrics 881 Cesium
Fig. 3. The magnetometer used in the survey, magnetometer (fig. 3), which was towed from the stem of a six
INA Quarterly 30.4
meter fiberglass fishing boat. The software,
Maglog Lite, used in conjunction with a Glo-
bal Positioning System (GPS) unit, recorded
the data acquired by the magnetometer along
with the latitude and longitude. When iron
was detected, the reading was marked and
saved as a "hit" on the laptop (fig. 4). The lo-
cations of the hits were transferred to the GPS
unit, which was then used to navigate back
to the site for visual survey and identifica-
tion of the hit.
In order to identify the targets, we uti-
lized underwater search patterns, and added
Erika Laanela to our team to assist with the
diving. We began the search using handheld
metal detectors (fig. 5), but we found that
simple visual inspection was sufficient for
identifying the targets. This method worked
extremely well, and we were able to identify
every target that we dove on. Unfortunately,
none of these turned out to be historic finds, Fig. 4.
but there was only time to dive on a handful survey.
of the logged hits.
At the completion of this season, it be-
came clear that adjustments to our equipment setup
would be necessary to run a more efficient survey in
the coming seasons. The use of the small vessel lim-
ited our time at sea and our ability to run parallel,
evenly spaced lanes. We were operating during the
Fig. 5. Handheld metal detectors were used initially, but proved la
Photo: K. Custer
A laptop computer was used to record data obtained during the
beginning of hurricane season, which usually has sta-
ble weather conditions. This year, however, the is-
land was pounded by several storm systems (fig. 6).
This led to seas with up to three meter swells and
strong winds, conditions that make it difficult to
keep a steady course in a small boat. The sur-
vey was limited to the hours between five
a.m., just before the sun rose, and two p.m.,
when the seas became too rough for the cap-
tain to safely navigate.
Another change that will be made for
the following seasons is the inclusion of a
side-scan sonar. It was originally believed
that the sea floor sloped too greatly and was
covered by too many coral blooms (fig. 7) for
the effective use of such equipment. After this
season, however, it was determined that the
area of seafloor that we are concentrating on
is relatively flat with only scattered coral and
a sandy bottom. These conditions are ideal
for the combined use of side-scan sonar and
The information gathered from this sea-
son's work provides us with promising pros-
pects for the coming years. We have established
an archaeological presence on the southern
:uster coast and a positive relationship with the Do-
minican government, and we look forward to
rgely the coming seasons of work there.
INA Quarterly 30.4
Fig. 6. Stormy skies warned of sea conditions that would
hamper the survey.
Photo: K. Custer
Acknowledgements: We would like to thank the following institutions and persons: the board and staff of the Institute of
Nautical Archaeology and RPM Nautical Foundation, the staff of the Officina Nacional de Patrimonio Cultural Sub-
aquatico and Acquario Nacional de Republic Dominicana, Dr. Jerome Hall, Dr. Donny L. Hamilton, Dr. Kevin Crisman,
Francisco Escoto, Francis Soto, Enrique Pugibet, Erika Laanela, Gustavo Garcia, and everyone who took a chance and
believed in two graduate students with a great idea. v
1946 Caribbean: Sea of the New World. New York: Alfred A. Knopf.
Morison, Samuel E.
1942 Admiral of the Ocean Sea: A Life of Christopher Columbus. New York: The American Past: Book of the Month Club, Inc.
Smith, Roger C.
1993 Vanguard of Empire: Ships of exploration in the Age of Columbus. Oxford and New York: Oxford University Press.
Fig. 7. Coral blooms and other undersea life made a
pretty picture, but complicated the search for ship-
Photo: K. Custer
INA Quarterly 30.4
Sea Power in the Medieval Mediterranean:
The Catalan-Aragonese Fleet in the War of the Sicilian Vespers
by Lawrence V. Mott
Gainesville: University Press of Florida
ISBN: 0-8130-2662-8, 384 pp, 16 b&w figures, 5 maps, 2 tables, appendix, notes, bibliography, in-
dex. Cloth. Price: $59.95.
There are remarkably few books devoted to naval warfare during the Middle Ages. To some extent, this is due to a
prevailing attitude among historians that the relatively small medieval states did not have sufficient resources to build
significant fleets. If so, sea power cannot have been as important during these centuries as it had been for ancient empires
and was to be for modem nation-states. In Sea Power in the Medieval Mediterranean, Texas A&M Nautical Archaeology
Program graduate Lawrence V. Mott provides facts that should dispel this attitude.
Between 1282 and 1302, the royal families of France and Aragon fought what is now called the War of the Sicilian
Vespers. In 1263, the Papacy had ended a long conflict with the Hohenstauffen Holy Roman emperors by granting their
territories in southern Italy-the Kingdom of Sicily-to the French prince Charles of Anjou, brother of King Louis IX.
Charles and his tax collectors became extremely unpopular. On March 29, 1282, Angevin soldiers molested a group of
women on their way to evening prayers (vespers) in Palermo. The ensuing riot developed into a revolution that expelled
the surviving occupiers from the island. Recognizing that they could not withstand the French alone, the Sicilians offered
their crown to King Peter Ml of Aragon. For the next twenty years, there were two Kings of Sicily, one French and the other
Aragonese, with the border between their respective territories in constant flux. Although the war technically ended as a
draw, the peace treaty gave Aragon-and its successor, Spain-effective commercial and strategic control of the western
Mediterranean basin for centuries. By diverting European attention westward, the war also contributed to the final fall of
Christian Palestine in 1291 and to the Byzantine reconquest of the remnants of the Latin Empire.
The most significant fact about the War of the Sicilian Vespers is that it was predominantly a naval war, with both
large fleet actions and amphibious assaults. There are a number of contemporary accounts of the war, and many of the
account books of the Aragonese admiral, Roger of Lauria, have been preserved in the Valencia Cathedral archives. It is thus
possible to reconstruct the logistical and economic infrastructure that supported a major medieval fleet Roger's opponents
almost always outnumbered his forces, but he was able to take successful advantage of superior personnel, material, and
tactics. Lawrence Mott shows us how.
Nautical archaeologists will want to pay particular attention to Chapter 9, which describes the ships that comprised
the Aragonese fleet. In one critical battle of the war, fought in Grand Harbor, Malta, on June 10,1283, the Aragonese were
outnumbered two to one. Nevertheless, there were approximately four thousand French deaths (including almost half the
able-bodied male population of Marseille) against fewer than one hundred Aragonese casualties. This was partly due to
superior training, but seems mostly due to differences in ship design philosophy. The standard Mediterranean galley,
based on Byzantine models, represented a compromise between speed and capacity. It had 108 oars and a low freeboard.
In contrast, the fleet Roger of Lauria commanded was largely composed of 120-oared vessels, which were substantially
slower, but could support fore- and stern-castles to shelter the soldiers who shot arrows and hurled projectiles into an
opposing ship to neutralize its complement before boarding. Three centuries later, the evolution of this philosophy led to
the great unhandy galleons of the Spanish Armada.
The real core of Sea Power in the Medieval Mediterranean, however, are the chapters describing the evolution of the
office of admiral during the Middle Ages. An admiral was not only, nor even primarily, a sea commander, but was respon-
sible for procuring ships, crews, and supplies for the fleet. The Valencia documents provide a detailed look at one particu-
larly successful admiral, but the author gathers information from many available sources to describe how medieval fleets
were administered. The book explains how these navies were funded, manned, and operated.
In the end, the reader is convinced that sea power was much more significant in medieval history than historians
have often been willing to admit. The theories of Alfred Thayer Mahan therefore apply to this era as well as to ancient and
modem times. Control of the seas was a deciding factor in the War of the Sicilian Vespers and in the rise of Aragon to a
position of dominance in the Western Mediterranean. Isabella of Castille's husband Ferdinand, Aragon's last king, project-
ed this dominance into the New World as the co-sponsor of Columbus. ,
INA Quarterly 30.4
By Filipe Castro
Sealed by Time, The Loss and Recovery of the
Trowbridge: Cromwell Press: The Mary
Rose Trust, 2003
ISBN: 0-9544029-0-1, 194 pp, 140 illustra-
tions (including b&w photos maps and
drewings, 22 color plates, appendix, index.
This long-awaited book is the first of a series of five
volumes on the history and archaeology of Mary Rose, an
English warship built between 1510 and 1512, and lost on
July 19, 1545, off Portsmouth, during an engagement with
the French fleet of Admiral D'Annebault.
Built in the early years and lost in the last years of King
Henry VIE's reign (1509-1547) Mary Rose was a large warship
with three decks, large fore and stem castles, gun ports on the
main and upper decks, and four masts. There is a drawing of
Mary Rose, dating from 1546, in the Anthony Roll. Formulas to
calculate tonnage may have changed in England through its
lifetime; its capacity was estimated at 500 tons in 1512, and
increased gradually to eight hundred tons in 1545.
Refitted in 1527 and 1528, Mary Rose was an old but probably very able ship by the time of its loss, which seems
to have resulted more from misfortune than mishandling by its captain or crew. Caught by a sudden strong gale, the
ship listed dangerously to starboard, letting water in through the orlop deck gun ports, which caused its rapid sinking
and the loss of most of its crew. Only a handful of men survived. As a possible human error, several sources mention
too heavy a load of artillery on its decks as the main cause for the rapid listing that led to Mary Rose's sinking.
Salvage attempts followed almost immediately, but little was recovered from the shipwreck site until 1836 and
1840, when the Dean brothers salvaged four bronze guns, nineteen parts of iron guns, and a section of the mast. The
ship remains were found in 1836 when a group of five fishermen hired a diver named Henry Habbinett to disentangle
their net from a seabed obstruction. Soon after, John Dean was called to the site and raised a bronze gun with an
inscription referring to King Henry VII. That same year and again four years later, John Dean and his brother Charles
raised the remaining artifacts.
It was not until 1966, when Alexander McKee found the site, that Mary Rose would become the center of a major
archaeology project that changed the way the British looked at their submerged cultural heritage and eventually led to
the enactment of the Shipwreck Protection Act of 1973. Mr. McKee's team found the first hull timbers in 1971 and six
years later a trench was dug into the sediment revealing the ship's hull, which, we know now, was preserved along
thirty percent of its extension.
From 1972 to 1982 the archaeological excavation was carried out to the highest standards. In 1982 the hull was
raised and moved into the Historic Dockyard at Portsmouth, where the Mary Rose museum was to be built.
The excavation of Mary Rose uncovered close to twenty-six thousand artifacts, including a notable collection of
guns, human remains of at least 179 individuals, and above all a significant portion of the ship's hull. Mary Rose's hull
remains are the archive of an incredible amount of precious information about the ship's design and construction, and
their study will undoubtedly yield great discoveries about English shipbuilding in the fifteenth and sixteenth centuries.
There is incredibly little information on how English ships were built in the period, with almost no clues in the
written sources about the way they were designed and assembled. The few other shipwreck sites from this period-
such as the Hamble River shipwreck thought to be the 1420 Grace Dieu, and the Woolwich ship thought to be the
Sovereign from 1487-are far from being fully studied and published.
INA Quarterly 30.4
As an introduction to the archaeology of Mary Rose, this volume is excellent. Along its 194 pages of very orga-
nized, well illustrated, and clearly written text, the reader is guided through the incredible story of this ship, from its
construction, possibly at Portsmouth, to its final arrival at the Mary Rose Museum in the Historic Dockyard of Ports-
mouth. Following the description of the 493 years of life of the ship, this pleasant book is also a repository of all the
known names associated to it, and their roles in Mary Rose's fate.
The first of fifteen short chapters describes the story of the ship, its construction, sailing abilities, refitting opera-
tions, and life at sea or on ordinary, ending with the circumstances of its loss, illustrated by transcriptions of the known
accounts of the events surrounding its sinking. The following two chapters describe the discovery of Mary Rose in the
nineteenth century, and salvage of a small collection of artifacts, and the re-discovery of the site and hull remains in the
twentieth century, as well as the actions that led to its eventual archaeological excavation.
Chapter four explains clearly and concisely the strategies adopted in the excavation and recording of the archae-
ological context, and is followed by the epic of Mary Rose's raising, an extraordinary story that was followed on televi-
sion by millions of people worldwide. Chapter six tells the story of the creation of the Mary Rose Trust and the struggle
to raise funds to carry on such a project without state support. Chapters seven and eight cover the monitoring of the site
after the raising of the hull, and are followed by a post-excavation study of the stratigraphy of the site and interpreta-
tion of the site formation process.
Chapters ten, eleven and twelve deal with the identification of the ship and a preliminary description of its hull
remains and contents. Special sections on the human remains and artillery give us a first taste of the contents of vol-
umes four and five of this publication. Chapter thirteen reevaluates the circumstances of the loss of Mary Rose and
challenges the traditional version blaming mishandling of the ship by its crew during the turning to fire its port guns at the
French opponents. In fact, most of the starboard guns found on Mary Rose were still loaded, making it implausible that the
ship was attempting to turn without having closed the orlop deck gun ports. It is more likely that the ship was just
caught by a sudden and very strong gust of wind, heeled, made water through the starboard side gun ports, and sunk.
Chapters fourteen and fifteen deal with the historical significance of this shipwreck, the impact of this project for
nautical archaeology as a discipline in the United Kingdom, and the way in which it was perceived by the public and
the government. Avoiding strong statements and criticism of governments and politics, the author explains how im-
portant the Mary Rose project was in showing the public that the objectives of nautical archaeology are not different
from those of other branches of archaeology. Since shipwrecks are an important part of any country's cultural heritage,
they deserved to be protected, studied, and enjoyed by both scholars and the general public.
It is known how the Mary Rose project led to the enactment of the first legislation protecting shipwrecks as part of
the British cultural heritage. But as the author points out, the importance of this project is much wider: by making the
archaeology of Mary Rose its first priority, the Mary Rose Trust-a fully private organization-has sent a cear state-
ment about the importance of nautical archaeology over treasure hunting.
There are other famous shipwreck projects whose stories endure in the public memory, in which artifacts whose
conservation is not worth their value at auction were discarded, hulls were not properly recorded, and artifact collec-
tions scattered at auction. The Mary Rose project aimed at understanding the past rather than making a profit. In our
time, when projects formerly announced as treasure-hunting ventures are now referred to under euphemistic designa-
tions such as "commercial archaeology," these five books make an eloquent distinction between archaeology and trea-
We can hardly wait for the next four volumes of this series. The second volume is expected in 2006, will be edited
againby Peter Marsden, and is dedicated to the study of the ship's hull. The third volume will be edited by Alexzandra
Hildred and will treat the weapon collection. Its publication is scheduled for 2004. Also expected in 2004 is volume four
of this series, edited by Julie Gardiner, dealing with the matters of life and death aboard Mary Rose. The fifth and last
volume will be edited by Mark Jones and is expected soon. Its subject is the conservation of the ship's hull and its
almost twenty-six thousand artifacts.
A must in any nautical library, these five volumes are among a series of long awaited publications on the most
important shipwrecks excavated by archaeologists in the twentieth century. This first decade of the twenty-first centu-
ry looks very exciting for nautical archaeologists: following the wonderful publications of the Danish Viking Museum,
this series on the archaeology of the Mary Rose is expected to be soon joined by the full publications of the fourth
century BCE Kyrenia ship, the eleventh century CE Serce Limaru ship, the sixteenth century Basque galleon excavated
in Newfoundland by Parcs Canada, the ten excavations of the sunken city of Port Royal in Jamaica, and the archaeology
of the Swedish ship Vasa.
INA members should note the enclosed David R. Brown brochure on the series, as they can receive a 20% dis-
count. For further details, see http://ina.tamu.edu/oxbow.htm. a,
INA Quarterly 30.4
Donny L. Hamilton
The Archaeological Committee of the Institute of Nautical Archaeology (INA) meets annually
to review the year's excavation projects and surveys and to approve excavations and surveys for the
next year. For 2003, it reviewed the Oklahoma steamboat, Pabuc Buru, Bulgarian Black Sea, and Pipe
Wreck excavations. In addition, the results of the surveys conducted in Turkey, Italy, Cyprus, Moroc-
co, and the Dominican Republic were evaluated. Progress reports on the Serge Limaru shipwreck in
Turkey, Mombasa Wreck in Kenya, the-Denbigh Project in Texas, and the Casis do Sodre Project in
Portugal were scrutinized. We were very pleased with the accomplishments of 2003! However, we are
not content with resting on our laurels and we have ambitious goals for 2004.
In the Old World, a number of interesting projects were proposed. Dr. George Bass completed
the excavation of the late sixth-century BCE wreck at Pabuq Burnu in Turkey in 2003 and we decided
to devote the survey in Turkey 2003 toward identifying the site to be selected as the next excavation
project in Turkey. Two possible sites, the sixth-century BCE Archaic site at Kekova and the fifth-
century BCE Classical shipwreck at Asian Burnu, will be investigated further. In addition, a survey
will be conducted in the Arab Adasi area where the famous Demeter bronze bust was raised in 1952.
Survey projects in the United Arab Emirates, Italy, Greece, and Cyprus were approved. On-going
post-excavation work will continue on the 1300 BCE Bronze Age wreck at Uluburun, the Kadirga gal-
ley in Istanbul, Turkey, the sixteenth-century Portuguese wreck in Mombasa, Kenya, and the elev-
enth-century CE wreck at Serce Limaru, Turkey.
We had hoped to undertake a new excavation of an early 1500s Spanish shipwreck in Panama.
However, negotiations to secure permission were not successful. We continue to look for a significant
shipwreck to excavate in the New World. Closer to home, we continue to be very excited about INA's
participation with the Oklahoma Historical Society on the early nineteenth-century river boat in the
Red River in Oklahoma. Dr. Kevin Crisman will be taking a team of Texas A&M University students
to start the second of a planned three year project on the river boat. Dr. Jerome Hall will continue his
excavation on the seventeenth-century Pipe Wreck at Monte Christi in the Dominican Republic. In
addition, the second year of a survey in the Dominican Republic is planned. Post-excavation research,
conservation, and report writing continues on the Civil War blockade runner, the Denbigh near
Galveston, Texas. Finally we will be supporting a Nautical Archaeology Program student to assist in
the recording the 1281 CE Chinese junk wrecks found recently in Japan.
It has been a number of years since we have upgraded our diving gear and we cannot put it off
any longer. New diving equipment is needed for both our Old World and New World projects. Sever-
al years back, with a generous gift from the Institute for Aegean Prehistory, we were able to purchase
our own submersible, Carolyn. This has proved to be a valuable survey vehicle. Now we would like to
make it even more versatile by installing on it a sub-bottom profiler, a side-scan sonar, and a computer
to operate both pieces of equipment. Also the addition of lights and a remote controlled video camera
would be useful.
It is obvious that we have ambitious plans. This is where members' support is so vital. v
INA Quarterly 30.4
Vol. 30 Index
Bass, G. E, "International Workshop in NauticalArchaeology, Bodrumr
2002," 301, 9-11
Bass, G. E, "Titanic 2003," 303, 3-9
Castro, ,"AReview of Saled by e Tn he Loss and Recovery ofthe Mary
Rose by Peter Marsden," 30.4, 24-25
Cox, S. and K. Custer, "Post-Excavation Techniques for Recording Ship
Crisman, K, "The Horse Ferry Model," 30-2,17
Criman, K. and W Lees, "Beneath the Red River's Waters: The Okla-
homa Steamboat Project, Part I," 30.2,3-8
Custer, K, 'The First Season of the Dominican Republic Survey Project
A Preliminary Survey of the Southern Coast," 30.4,19-22
Greene, E, "Endless Summer The 2002 Excavation Season at Pabuq
Burnu, Turkey," 30.1,3-11
Grieco, G., 'The Model Reconstruction of La Salle's Ship La Belle," 30.4,
Hamilton, D., "Looking Back. Looking Forward," 30.4,26
Livadas, G. KI, 'The Ninth International Conference on Graeco-Orien-
tal and African Studies," 30.2,16
McManamon, J., M. YDAgostino, and S. Medas, "Excavation and Re-
cording of the Medieval Hulls at San Marco in Boccalama
Nasti, A. and H. Bado, "A Witness of Trafalgar The Cannon of the
HMS. Agamemnon," 303,14-17
Swanick, L. A, "Maritime Source Material in the United Kingdom and
Ireland," 30.2, 12-15
Trakadas, A, "Morocco Maritime Survey: The 2002 Season," 30.1,12-
Aragon and Catalonia, 30.4,23
Archaeological Theory, 30.2,19-20
Carolie, 30.1, 5; 30.4, 26
Conference on Graeco-Oriental and African Studies, 30.2,16
Cultural Resource Management, 302, 20-21
Dominican Republic, 30.4,19-22
Cais do Sodr, 303, 10
Mary Rose, 30.4,24-25
Pabu Burnu, 30.1,3-11
Red River Wreck, 302, 3-
San Marco in Boccalama (Venice), 30.1, 22-28
Galveston County Historical Museum, 303,18
Horse Ferry, 302,17
Cais do Sodre, 303,10-13
La Belle, 30.4, 3-18
Pabu< Bumu, 30.1,9-10
San Marco in Boccalama (Venice), 30.1,26-28
Kahn, Harry C. L, 30.2,22
LeFrak, Samuel J., 30-2,23
Lodge, Marilyn H., 30-2,21
Peterson, Mendel L, 303,19
Institute of Nautical Archaeology
Annual Meeting 2003,30.1, 29-31.
President's Letter, 30.4,26
Kahn, Harry C. I, 30-2,22
Keldysh, R/V Akademi Mstislav, 303,3-9
La Belle, 30.4,3-18
La Salle, 30.4, 3
LeFrak, Samuel J., 30.2,23
Lodge, Marilyn H-, 302, 21
Maritime Source Materials, UK and Ireland, 30.2,12-15
Mary Rose, 304, 24-25
Mir I and 2,303,3-9
Model Construction, 30.4,3-18
Nelson, Admiral Horatio Lord, 303,14-17
Oklahoma Steamboat Project, 302,3-8
Pabug Bumu, 30.1,3-11
2002 Excavation Season, 30.1,3-7
Artifacts, 30., 7-9
Hull Remains, 30.1,9-10
Peterson, Mendel L., 303,19
Polymers for use in conservation, 302,18
Research Materials, UK and Ireland, 302,12-15
Red River Wreck, 30.2,3-8
Sagalevitch, Anatoly, 303,5
Horse Ferry, 30.2,17
La Belle, 30.4,3-18
Sicilian Vespers, 30.4 23
Steamboats, Western, 30.2,3-4
Dominican Republic, 30.4,19-22
Swink, Oklahoma, 30.2,3
Trafalgar, Battle of, 303,14-17
Virazon, 30.1, 4-5
Workshop in Nautical Archeaology, Bodrum 2002,30.2,9-11
Reviews and "Just Released"
Bums, J. M., The Life and Times of a Merchant Saior, 30.2,19
Marsden. P, Sealed by Tome The Loss and Reovery of the Mary Rose, 30.4,
Mott, L V., Sea Power in theMedieva Mediterranean The Catalan-Aragonese
Fleet in the War of the Sicilian Vespers, 30.4,23
Smith, C. W, Archaeologiad Conservation Using Polymers, 30.2,18
Spirek, J. D. and D. A. Scott-Ireton, Submerged Cultural Resource Man-
Staniforth, M, Material Culture and Consumer Society, 30 219-20
INA Quarterly 30.4
INSTITUTE OF NAUTICAL ARCHAEOLOGY
Christine A. Powell
Donny L. Hamilton, Ph.D., President"
Donald A. Frey, Ph.D., Vice President* Cemal M. Pulak, Ph.D., Vice President
Claudia F. LeDoux, Chief Accounting Officer and Assistant Secretary
Michelle Chmelar, Assistant Accounting Officer
Tufan U. TuranL, Administrator, Bodrum Research Center
William L. Allen
John H. Baird
George F. Bass, Ph.D.,
Edward O. Boshell, Jr., Chairman*
John A. Brock
Elizabeth L. Bruni
Allan Campbell, M.D.
William C. Culp, M.D.
Robin P. Hartmann
BOARD OF DIRECTORS
John Cassils, M.D. Charles Johnson, Ph.D."
Gregory M. Cook Mustafa Kog
Lucy Darden Francine LeFrak-Friedberg
Thomas F Darden* Robert E. Lorton
John De Lapa Alex G. Nason
Claude Duthuit, George E. Robb, Jr.
Danielle J. Feeney' Lynn Baird Shaw
Robert Gates, Ph, D. Ayhan Sicimoglu"
James A. Goold, J.D.
Secretary & General Counsel'
Faith D. Hentschel, Ph.D. Thomas McCasland, Jr.
Susan Katzev Dana F. McGinnis
Wiliam C Klein, M.D. Michael Plank
George Lodge Molly Rei y
J. Richard Steffy
William T. Sturgis
Frederick H. van Doominck, Jr., Ph.D.
Robert L. Walker, Ph.D.'
Peter M. Way, Vice-Chairman'
Garry A. Weber
George O. Yamini
Sally M. Yanini
Betsey Boshell Todd
NAUTICAL ARCHAEOLOGY PROGRAM FACULTY
Deborah Carlson, Assistant Professor
Fihpe Castro, Ph.D., Assistant Professor, Frederick R. Mayer Faculty Fellow of Nautical Archaeology
Kevin J. Crisman, Ph.D., Nautical Archaeology Faculty Fellow
Donny L. Hamilton, Ph.D., George T & Gladys H. Abell Professor of Nautical Archaeology, George O. Yamini Family Professor of Liberal Arts
Cemal M. Pulak, Ph.D., Frederick R. Mayer Professor of Nautical Archaeology
C. Wayne Smith, Ph.D., Assistant Professor and INA Faculty Fellow
Shelley Wachsmann, Ph.D., Meadows Associate Professor of Biblical Archaeology
NAUTICAL ARCHAEOLOGY PROGRAM FACULTY EMERITUS
George E Bass, Ph.D.,
George T. & Gladys -L Abell Professor of Nautical Archaeology, George O. Yamini Family Professor of Liberal Arts, Distinguished Professor Emeritus
Frederick H. van Doorninck, Jr. Ph.D., Frederick R. Mayer Professor of Nautical Archaeology, Emeritus
J. Richard Steffy, Sara W & George 0. Yamini Professor of Nautical Archaeology, Emeritus
Mr. & Mrs. Ray H. Siegfried Hl Graduate Fellow: Alexis Catsambis Marian M. Cook Graduate Fellows: Peter D. Fix
J Barto Arnold, M.A.
Aye Atauz, M.A.
Kroum N. Batchvarov, M.A.
Arthur Cohn, J.D.
Nergis Giinsenin, Ph.D.
Jerome L. Hall, Ph.D.
Australian Institute of Maritime Archae
Bryn Mawr College
University of California, Berkeley
University of Cincinnati
Donald G. Geddes III Margaret E. Leshikar-Denton, Ph.D
Jeremy Green, MA. Maria del Pilar Luna Erreguerian
Justin Leidwanger John McManramon, Ph.D.
Faith D. lentschel, Ph.D. Frederick Hocker, Ph.D
Fredrik T. Hiebert, Ph.D. Carolyn G. Koehler, Ph.D.
ology Corning Museum of Glass Partners for
Departamento de Arqueologfa Subacubtica de University N
la I.NA.H., Mexico Texas A&M
University of Maryland, Baltimore County RPM Nautic
New York University, Institute of Fine Arts Texas A&M
University of North Carolina, Chapel Hill The Univers
INSTITUTE OF NAUTICAL ARCHAEOLOGY STAFF
Zafer Gui Sheila D. Matthews, M.A.
Bilge GiUnesdogdu Sheree Moore
G(lser Kazanctoglu Asaf Oron, M.A.
Bayham Kosar Muammer Ozdemir
Nurgiil Kulah Robin C. M. Piercy
Ralph K. Pedersen, M.A.
Brett A. Phaneuf
Athena Trakadas, M.A.
Jeff Royal, Ph.D.
Cheryl Ward, Ph.D.
Gordon P. Watts, Jr., Ph.D.
museum, University of Pennsylvania
ity of Texas at Austin
A. Feyyaz Subay
Fred Van de Walle