Group Title: INA quarterly
Title: The INA quarterly
Full Citation
Permanent Link:
 Material Information
Title: The INA quarterly
Alternate Title: Institute of Nautical Archaeology quarterly
Abbreviated Title: INA q.
Physical Description: v. : ill. ; 28 cm.
Language: English
Creator: Institute of Nautical Archaeology (U.S.)
Institute of Nautical Archaeology (U.S.)
Publisher: Institute of Nautical Archaeology
Place of Publication: College Station TX
College Station TX
Publication Date: Winter 1993
Frequency: quarterly
Subject: Underwater archaeology -- Periodicals   ( lcsh )
Archéologie sous-marine -- Périodiques   ( rvm )
Genre: periodical   ( marcgt )
Dates or Sequential Designation: Vol. 19, no. 1 (spring 1992)-
General Note: Title from cover.
General Note: Latest issue consulted: Vol. 23, no. 2 (summer 1996).
 Record Information
Bibliographic ID: UF00098800
Volume ID: VID00008
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 26536606
lccn - sf 94090290
issn - 1090-2635
 Related Items
Preceded by: INA newsletter (Institute of Nautical Archaeology (U.S.))


This item has the following downloads:

V20_No4 ( PDF )

Full Text


The INA Quarterly

Volume 20, No. 4 Winter 1993

PO Drawer HG
1993 The Year in Review College Station, TX 77841-5137

3 Letter From the President Hear firsthand of our latest discov-
eries in nautical archaeology. Mem-
bers receive the INA Quarterly,
4 The Shipwreck at Uluburun: 1993 Excavation Campaign scientific reports, and book dis-
Cemal Pulak counts.

13 Weight, Money, and Weight-Money: The Scales and Regular ....... $25
Weights from Serce Limani Contributor .... $50
Fred Hocker
Supporter .... 100
22 Relics of the Revolution and a Schooner Called ..
Benefactor . . $1000
Water Witch
Kevin Crisman Student/Retired .. $15

31 News & Notes Checks in U.S currency should be
31 News & Notes
made payable to INA.

On the cover: A diver works amid the last copper ingots and stone anchors to be removed from the Late Bronze Age
shipwreck at Uluburun, Turkey. The largest preserved section of the ship's hull lies above the ingots (Photo: D. Frey).

April 1994 by the Institute of Nautical Archaeology. All Rights Reserved.
INA welcomes requests to reprint INA Quarterly articles and illustrations. Please contact the editor for permission.

The INA Quarterly was formerly the INA Newsletter (vols. 1-18).

Institute of Nautical Archaeology

Editor: Michael A. Fitzgerald

- Letter from the President -

Dear INA Members,
We can all take great pride in the accomplishments of the past year. In this issue we are happy to report still more
discoveries at Uluburun, Turkey, in Lake Champlain, and in the library, where research on the Serce Liman Glass Wreck
material continues. The year was also special in that INA successfully concluded its National Endowment for the
Humanities Challenge Grant drive. You should take particular satisfaction in the knowledge that you helped make all of
this possible.
Thank you for your assistance to INA. With your support, we can continue a partnership that is creating a World
Center for Archaeology and turning our visions into reality.
Special thanks are extended to our major 1993 contributors INA's Board of Directors and all who gave to the
NEH Challenge Grant Drive, and to our many other supporters, listed below.


These endowments, given to Texas
A&M, further the goals of INA
The George T. & Gladys H. Abell
The Mr. & Mrs. J. Brown Cook
Graduate Fellowship
The INA Faculty Fellowship
The Frederick R. Mayer Faculty
The Frederick R. Mayer Professorship
The Meadows Foundation
The Nautical Archaeology Faculty
The Mr. & Mrs. Ray H. Siegfried
II Graduate Fellowship
The George O. Yamini Family Chair
in Liberal Arts
The Sara W. & George O. Yamini

Foundations, etc.
The Amoco Foundation, Inc.
Andersen Consulting
Conservation Analytical Laboratory,
Smithsonian Institution
The Anna C. & Oliver C. Colburn
The John Brown Cook Foundation
Corning Incorporated
Covington & Burling
Cressi-sub, Italy
INA Films
INA Foundation
The Institute for Aegean Prehistory
The Jewish Communal Fund
Lubrizol, Incorporated
The Meadows Foundation
R.J.R. Nabisco
The Nason Foundation
The National Endowment for the
The National Geographic Society
Philip Morris
Shell of Turkey, Ltd.
Texas A&M University Development
Texas A&M Research Foundation
Texas A&M Hyperbaric Lab
Texas A&M University

Mr. Erik Jonsson

Dr. Allan C. Campbell

Toni & Maria Pia Bassani
Mr. & Mrs. Charles B. Collins
Mr. John DeLapa
Mr. & Mrs. Maurice Duca
Dr. Faith Hentachel
Mr. & Mrs. George W. Lodge
Mr. Isaac A. Morris
Dr. Ernestine A. O'Connell
Mr. & Mrs. Richard L. Rosenberg

Supporting Members
Ms. Emel Akalin
Mr. Thomas H. Anderson
Mr. David A. Batchelor
Mr. Alan L. Boegehold
Mrs. Anne B. Bradford
Mr. & Mrs. Thomas H. Bredt
Mrs. Marianne Buchwalter
Mr. John Camp
Mr. Robert W. Campbell
Mr. Kenneth Cassavoy
Mr. Lionel Casson
Dr. Marco E. Chioffi
Mr. & Mrs. George H. Clippert
Mr. E.H. Cooley
Mr. & Mrs. Frederick J. Cox
Robert P. & Jamie P. Davis
Mr. & Mrs. Thomas A. Deering
Mr. Robert F.'Dennis
Mr. Frank Devenney
Janet & Jules Doner
Mrs. Ruth L. Dugan
Ms. Marilyn S. Easly
Mr. &Mrs. Spencer M. Ehrman
Dr. Cynthia J. Eiseman
Mr. John Ellis
Dr. & Mrs. H.C.M. Erzurumlu
Dr. & Mrs. William F. Farr
Mrs. Thomas B. Fifield
Dr. Hudson D. Fowler, Jr.
Ms. Barbara E. Fox
Mr. & Mrs. Robert J. Frasca
Ms. Judith G. Galvin

Mr. & Mrs. Nathan
Dr. & Mrs. Gerald Green
Dr. R.J. Halbert
Mr. Robert M. Halperin
Ms. Chatten Hayes
Miriam S. Hogan
Dr. Virginia Huffer
Mr. Frederick C. Jessop,
Mr. & Mrs. J. Pierre
Mr. Weldon D. Kruger
Mr. Robert C. Kuzela
Dr. & Mrs. Walter G.
Mr. & Mrs. Hugh R.
Mrs. Helen H. Lindgren
Mr. & Mrs. Stanley R.
Mr. & Mrs. Edward C.
Mr. & Mrs. Thomas R.
Mr. & Mrs. Thomas
McCasland, Jr.
Mr. Brian McLaggan
Mrs. Madeline B. Nelson
Mr. Murray W. Pegler
Mrs. Wilma Jane Perlman
Mrs. Ann J. Pierce
Mrs. Judith Price
Mr. & Mrs. Jerry L.
Mr. & Mrs. Howard A.
Dr. Alice S. Riginos
Mr. Billings K. Ruddock
Mr. & Mrs. Raymond
Dr. & Mrs. William R.
Mr, & Mrs. Norman
Mrs. Joanne G. Starr
Mr. &Mrs. David A.
Mr. & Mrs. John L. Stern
Mr. William B. Tabler, Jr.

Ms. Sylvia E. Thomas
Mrs. Mary A. Tooze
Mr. & Mrs. Ronald A.
Mr. Richard B. Vowles
Sandra S. Weiksner
M. Clark-Werner & Lin
Sue Crobaugh Willis
Mr. J.E. Wilson
Mr. Patrick S. Wilson
Ms. Mary Winch
Mr. & Mrs. Robert L.
Mr. Keating V. Zeppa

Contributing Members
Mr. Richard 1. Abbatiello
Catherine Allegra and
James M. Tanner
Mr. Joseph C. Bates
Mrs. Lucille S. Beck
Dr. Mensun Bound
Mrs. Miriam E. Caskey
Ms. Mary E. Dasenbrock
Mr. Peter L. Engel
Dr. William A. Geffen
Mr. Robert C. Gleason
Mr. Glenn Grieco
Dr. David Hunn
Mr. Edgar W. Jones, Jr.
Ms. Denise Jure
Mr. Harry C. Kahn III
Mr. Yutaka Kondo
Mr. John G. Nelson, Jr.
Mr. James E. Nielson
Mr. Peter Reaveley
Mrs. Marian H. Sagan
Ms. Jane A. Scott
Ms. Robin Y. Sim
Mr. Joseph W. Smith
Mr. William Strayer
Mr. & Mrs. George Thomas,
Anica Walker Turner
Dr. Ronald E. Walsh
Miss Winifred E. Weter
Dr. Sarah U. Wisseman

Photo: C. PuJ*k
Figure 1. Aerial view of Uluburun from the southwest. Anchored over the wreck site, INA's
research vessel Virazon is visible near the cliffface (right center).



by Cemal Pulak, Mr. & Mrs. Ray H. Siegfried II Graduate Fellow

Between May 28 and August 20, 1993, INA completed its
tenth, and shortest, excavation campaign on the Late
Bronze Age shipwreck at Uluburun (now the preferred
spelling of the site), near Kas in southern Turkey. During
the campaign we made 1,908 dives totaling some 607 hours
of excavation time under water. This brings the number of
dives to 20,556, for a total of 6,006 hours of excavation
time on the wreck, more than on any other site in INA's
history. (By contrast, the full excavation of the seventh-
century Byzantine shipwreck at Yassiada, Turkey [1961-
1964], required only 1,224 working hours in 3,533 dives.)
At the end of our four-month 1992 season it was
expected that the completion of all areas of excavation and
the eventual removal of the remaining bulky artifacts (two
stone anchors, a large storage jar or pithos, and fifteen
copper ingots) could be accomplished with an abbreviated
season of just two months. Consequently the "final"
campaign was planned for June and July of 1993, with an
option for extension into mid-August, after which INA's
annual shipwreck survey would commence. The fortuitous

discovery of additional hull remains beneath the last copper
ingots removed from the wreck, and of bronze artifacts and
pottery in areas beyond the site perimeter, however,
showed us that the wreck still had much to offer. Even so,
it is expected that a campaign of two months' duration in
1994 will bring to a close our work at Uluburun.
Work in 1993 may be divided into the following three
categories: excavation in the central part of the site; the
excavation, mapping, and recovery of all wreck material
between ca. 55 and 61 m (180 and 200 ft) deep; and a
systematic metal-detector survey of all areas within and
immediately adjacent to the designated site boundaries.
Much of our effort in the area just north and northeast
of the large, boulder-like rock outcrop located centrally on
the site (grid squares K-P15, L-P16, N-P17, N-P18,
N-P19) was devoted to the completion of previously
unexcavated areas, the removal and raising of the remain-
ing fifteen copper ingots and two large stone anchors, and
the documentation and recovery of the portion of the ship's
hull discovered in 1984 that lay upslope of the copper

INA Quarterly 20.3

ingots. Excavation progress in these areas was rather slow
due to the profusion of beads and other small finds, and
the fine sand and sediment in a deep gully just upslope of
the boulder-like rock outcrop was found to have preserved
many fragments of wood and other organic materials that
required great care during excavation. This latter area re-
warded us with the majority of the important small finds of
the season. Among them are: a blank stone scarab; a
faience cylinder seal that is similar to one found in 1992
(see INA Quarterly 19.4: 10, fig. 9; Dominique Collon of
the Department of Western Asiatic Antiquities in the British
Museum, who is studying our cylinder seals for publica-
tion, believes both were probably made in the same uniden-
tified workshop somewhere in northern Syria, possibly near
Ugarit); a seashell ring, which raises to 23 the number
recovered from the wreck; a small bronze zoomorphic
weight in the form of a recumbent bull and many other
pan-balance weights of hematite, stone, and bronze; and an
unidentified circular copper or bronze object encased in a
wooden sleeve and presumed to be a pan for a pan-type
balance. Also in this area were found many lead fish-net
sinkers; beads of agate, faience, and glass; "quarter Photo: M. Fitzg ld
oxhide" ingots of tin; bronze weapons (KW 4517, fig. 14, Figure 3. Peter Kuniholm of Cornell Uni-
p. 12) and tools (KW 4560, fig. 12, p. 10); Cypriot export- varsity (front) and Cemal Pulak examine a
wood sample's suitability for dendrochrono-
type pottery; an intact Canaanite jar (KW 4502, fig. 2) and logical dating.
sherds for many others. Sieving the Canaanite jar's

contents revealed chunks of the same yellowish material
found previously in other such jars on the wreck and
identified as terebinth resin.
From this area we also excavated and raised a meter-
long (3.28 ft) portion of a tree trunk (not a remnant of the
hull), possibly of juniper, along with two logs previously
identified as either spruce or larch. Eager to have these
pieces and pieces from the hull dated more precisely
through dendrochronology (tree-ring dating), we were most
pleased when Professor Peter Kuniholm visited. He took
samples back to the Malcolm and Carolyn Wiener Labora-
tory for Aegean and Near Eastern Dendrochronology at
Cornell University, where he will compare them with his
master dendrochron6logical sequence for the eastern
Mediterranean (fig. 3). Results are pending.
Without doubt, the most exciting and rewarding aspect
of the 1993 season was the documentation, study, and
eventual recovery of the section of hull wood partly
uncovered in 1984. During that season a sounding trench
was cut in grid squares M-015 to determine if any of the
ship's hull had survived. The sounding revealed a section
Photo: D. Fry of the keel, garboard strake (the plank or series of planks,
Figure 2. Excavation of the last intact Canaanite adjoining the keel), second strake, and fragments of the
jar (KW 4502) on the wreck. Sieving the jar's third strake on the port side, but only a fragmented portion
contents revealed terebinth resin.

INA Quarterly 20.3

r- INA 1992

of the garboard strake on the starboard side. In order to
protect this important material from damage, we have kept
it covered with a protective layer of sand during the
excavation and removal of heavy artifacts over the years.
Based on a cursory evaluation of the remains in 1984
and 1989, we learned that the Uluburun ship was construct-
ed in the same "shell-first," edge-joined-plank technique
found in Graeco-Roman ships. In contrast to the present-
day "skeleton-first" construction technique, whereby the
ship's planking is formed around and fastened to the pre-
erected skeleton of the vessel, the ancients used the "shell-
first" method. This was demonstrated conclusively for the
first time through the excavation and detailed study of the
late-fourth century B.C. merchantman discovered near
Kyrenia, Cyprus. The Kyrenia ship remained the earliest
well documented example of a ship built by this method
until the discovery of our Late Bronze Age shipwreck at
Uluburun. Except for the larger-sized mortise-and-tenon
joints used for the edge-fastening of the planks on the
Uluburun ship, the joinery method is virtually the same as

that seen on the Kyrenia hull. We thus have dated the use
of this construction method to about a thousand years
earlier than previously known! Details regarding other key
elements such as the keel and frames, however, would
remain unknown until the wood was fully excavated and
studied in succeeding campaigns.
It was therefore with much anticipation that we set about
excavating, recording, and raising the hull remains in
1993. When completely exposed, the remnants were found
to be fairly well preserved over an area measuring some
1.8 m (ca. 5.9 ft) along the keel by 1 m (3.28 ft) in width
(figs. 4, 5). After being completely drawn in plan and
sectional views and studied in situ, the keel and strakes
were placed in separate, custom-built wooden trays for
raising and eventual transportation to the Bodrum Museum
of Underwater Archaeology, where they will receive
conservation and more study. The trays were carried
under water from the site to the shallows near shore, where
they were then lifted to the surface (fig. 6). This precau-
tion was taken to prevent damage that could have occurred

INA Quarterly 20.3

S I I I J | K |I L H i N 0 0
I 26
INA 1992 o
0 '

t ,4
o 00 =


0 '


"a 7 -is I J K L Is a

if divers had ascended from the wreck directly to
the surface with the trays.
Soon after we began to clear the wood we were
excited to discover that we were confronted with an
unexpected and hitherto unknown structural config-
uration: in spite of detailed examinations of the
hull remains, both under water and on the surface,
no evidence for any framing has emerged. At this
time we reason that perhaps the preserved section
is not wide enough, or perhaps long enough, to
include frames or bulkheads or evidence for secur-
ing such elements to the planking, especially if they
were not affixed to the first few strakes they
spanned. But this is not the only intriguing con-
struction feature (fig. 5). Upon raising the keel we
found that it is wider (sided 28 cm = 11 in) than it
is high (molded 22 cm = 8.66 in), although the
latter dimension is based only upon what appears to
be the finished surface of a knot in the keel, and so
may not be the original molded dimension. Ac-
cording to what we know so far, the lower surface
of this timber originally protruded beyond the outer
planking surface by only a few centimeters. It
would have served as the ship's spine, and to
protect the planks and support the ship when
beached or hauled ashore. Unlike keels of later
sailing ships, however, it would have done little to
help the ship hold course or point nearer the wind
under sail. In other words, it appears that we have
a rudimentary keel, perhaps more of a keel plank
than a keel in the traditional sense. Not only will
this exciting discovery allow us to understand
better the evolution of ship construction technol-
ogy, but also to consider its possible implications
for sailing capabilities and the nature of trade in the
Late Bronze Age.
During the excavation of the hull remains, a
number of fascinating artifacts came to light near-
by. A group of bronze objects had concreted into
a mass along with wood, bone, and ivory artifacts.
Several are indicative of shipboard fishing during
the journey. Among these are a bronze harpoon
(KW 4254, fig. 7), a netting needle for repairing
torn fish nets (the second recovered from the site)

Figure 4 (Upper left). Plan view of the hull re-
mains first discovered in 1984. The white dots
mark the locations of pegs that secure the mor-
tise-and-tenon joints (upslope [west] is toward
the top of the photo). Figure 5 (Left). Section-
al view of the keel and part of a garboard
strake, looking upslope.

INA Quarterly 20.3

and several fishhooks. In the same con-
creted deposit was an ivory cosmetic or
unguent spoon with a handle in the form of
a feminine forearm with a clenched hand
(KW 4246, fig. 8), a bone tube that may
possibly have served as the spoon's case
(KW 4251, fig. 7), a partly worked long
bone of unknown purpose (KW 4255, fig.
7), and a nearly complete tortoise carapace
(KW 4250, fig. 7) that probably was used
as a sounding box for a musical instrument
such as a lute or lyre. On the southern
side of the keel or keel plank lay a small
limestone anchor (KW 4418, fig. 9) simi-
lar to one raised in 1989 (KW 2339).
Weighing 25.9 and 21.9 kg (ca. 57 and
48 Ib), respectively, they were probably

Photo: D. Frey
Figure 7 (Above). The concreted mass of wood, bone,
ivory, and bronze objects found near the hull wood: the
bone tube (beneath and left of brush); the worked long
bone (one end visible below and right of brush); the
harpoon (overlying the long bone); and tortoise cara-
pace fragments (left of bone tube and behind long
bone). Figure 8 (Below). The cosmetic or unguent
spoon of ivory (length ca. 9.8 cm).

Photo: D. Fry
Figure 6. Divers walk trays of hull wood to the shallowsfor lifting to the

anchors for the ship's boat, but they could have served as
hawser weights or lead (rhymes with reed) sinkers for large
fish nets. (A lead sinker is a weight affixed to the leading
and/or trailing bottom corner of a fish net and to which the
net marker buoy is also attached.) The much larger anchors
being carried on the ship appear to be sandstone.
The third and fourth rows of copper ingots (in N-P17,
N-P18) had served as a barrier that prevented small arti-
facts originally located in the aft portion of the vessel from
rolling down the steep slope upon which the ship eventually
came to rest. The excavation of the last copper ingots on
the wreck therefore revealed many objects, as had the
removal of the second row of ingots in 1992. The freeing
of these heavily concreted ingots required meticulous
chiselling, during which the appearance of unknown
artifacts in the encrustation
and among the ingots had
to be anticipated at all
times. This newly cleared
area yielded many beads
of agate, faience, glass,
and ostrich eggshell.
Other finds include Cypri-
ot ceramic export wares
(White-Shaved juglets and
milk-bowls); disk-shaped
glass ingots colored cobalt-
blue and turquoise; assort-
ed pan-balance weights;
lentoid agate beads, per-
haps intended as seal Pfoto: a. Gre. e, P. van Afen
blanks; a bronze adze Figure 9. The small lime-
stone anchor of 25.9 kg.

INA Quarterly 20.3

Photo: E. Greene, P. van Alfen
Figure 10. A head and feet of the two duck-shaped
cosmetic containers found in 1992. The wing
(length 14 cm) belongs to KW 2818.

blade, a knife, tools, and fishhooks; fragments of a lead (?)
pendant; and tin ingots.
Just downslope of the copper ingots (N-P19), around
and beneath the remaining two stone anchors, were many
glass ingots, some of which had to be excavated and
removed before the anchors could be raised. Most glass
ingots here (as elsewhere) were cobalt-blue in color, but a
considerable quantity of turquoise-colored ingots were also
recovered. Also discovered was a purple glass ingot frag-

Photo: D. Frey
Figure 11. Thorny burnet (dunnage) at right, and remains
of several disarticulated strakes and the keel found beneath
the last copper ingots removed from the wreck.

meant of the same color as a unique purple ingot found in
1992. Whether this fragment is part of the first ingot or
is from a yet undiscovered second ingot of this. unusual
color has not been determined. One perplexing aspect of
the glass ingots found here and in numerous other places
on the site is the occurrence of beautifully preserved
ingots alongside others that have hydrolyzed totally, so
that they have no mechanical integrity and their original
color is no longer discernible. The large number of
hydrolyzed glass ingots was another reason for the
generally slow pace of excavation. Extremely fragile,
they had to be encased in a protective "jacket" and freed
with utmost care before raising.
The sandy area immediately downslope of the stone
anchors (N-P19 and N-P20), mostly excavated in 1992,
yielded additional small artifacts that included more glass
ingots, a variety of beads, seashell buttons or beads, and
additional components of the two duck-shaped ivory
cosmetic containers recovered in 1992 (fig. 10): a wing
belonging to container KW 2818 (see INA Quarterly 19.4:
5, fig. 2), as well as a head, neck, and
feet. Apparently all these artifacts rolled
down from their original positions high-
er up on the slope.
With the removal of the last 15 copper
ingots from the wreck for a total of
354 on the wreck large quantities of
dunnage were exposed, comprising pri-
marily thorny burnet (a low, thorny shrub
ubiquitous to the Mediterranean) but also
grasses and oak leaves. To our delight
we encountered beneath this material
more remnants of the ship's hull, though
they are poorly preserved. This portion
of the ship came to rest on bedrock, not
on sand that could have cushioned and
protected the wood. Thus the hull suf-
fered distortion and damage from the
weight of the copper ingots that pressed
it against the uneven rocks jutting from
the seabed. Nevertheless, it was immedi-
ately apparent that new hull information
was available. In addition to the keel or
keel plank, there are at least six strakes
in evidence on the starboard side. Large p ,: E Greene
wooden pieces rounded in section and P. v = Al
placed perpendicular to the keel or keel Figure 12.
Small chisel
plank appeared at first to be the remains Sm 4 i
(KW 4560)
of poorly preserved irregular frames, but with wood
after all of the readily identifiable dun- handle (overall
nage had been removed, closer exami- length 19.5

INA Quarterly 20.3

nation of these pieces revealed that none was fastened in
any way to the keel (or keel plank) or the planking.
Consequently, they are probably large branches (dunnage)
that were placed under the brushwood for additional protec-
tion of the planks from the heavy metal ingots.
As remnants from the only Bronze Age seagoing ship
known to date in the Mediterranean, all of these pieces of
wood deserve thorough scrutiny. But because they were
found to be in such a poor state of preservation, their
removal promised to result in extensive damage and the
loss of unique information. Accordingly, we decided to
leave the wood in place until we could study and under-
stand it as much as possible a task not
easily performed while fighting the judge-
ment-clouding effects of nitrogen narcosis
at a depth of about 49 m (160 ft). We
estimated that an additional six to eight
weeks would be required to complete this
task, far more time than was left to us in
1993. This was the primary factor in our
decision to push the completion date of
the excavation into 1994. At the end of
the season we carefully covered all the
hull remains with a thick layer of fine
sand to protect them from the elements.
In order to locate artifacts we might
have missed in prior seasons, we initiated
a comprehensive, systematic metal-detec-
tor search of the approximately 450
square meters of site area and another 200
square meters around it. We found an
abundance of modern metallic trash in-
cluding nuts, bolts, wire, hose clamps, Figure 13. A b
nails, and various objects lost overboard (KW 4193), anf
from the Virazon during our ten seasons
on the wreck site (this in spite of our care to avoid contam-
inating the site in anticipation of an eventual metal-detector
survey), as well as pre-excavation modern refuse including
WWII shrapnel. Ancient contaminants such as net and
fish-line sinkers, anchors and anchor fragments, metal ves-
sels, and even a small sounding lead were also found.
The discovery during the survey of only a single artifact
a necked adze within the site boundaries demon-
strated that the site has been excavated methodically and
completely over the years. It had fallen into a crevice in
a rock ledge and had become totally sealed by encrustation.
Concretion deposits are usually chiselled open during
excavation in search of such objects, but this adze was
missed because the concretion was thought to be bedrock.
Two similar necked adzes were recovered during the 1985

campaign, all three belonging to a type found mostly in
Egypt and, to a lesser degree, south Palestine.
To our surprise, the metal detector led us to ceramic
objects as well. This enhanced capability to locate wreck
material benefitted us significantly in the deeper parts of
the site. We recovered half of a pithos and other pottery
fragments that had slid down the slope to depths of ca.
58-61 m (190-200 ft). As these remnants were few and
lay under about 50-90 cm (ca. 2-3 ft) of sand, their
discovery by airlifting alone would have been extremely
slow, labor intensive, and dangerous for divers working at
such extreme depths.

Photo: E. Greene. P. van Alfce
'ronze spearhead or dagger (KW 4217), the Mycenaean sword
d the trident (KW 4160) (preserved sword length ca. 51 cm).

After the conclusion of the metal-detector survey within
the wreck boundaries, the search was expanded to include
adjacent peripheral areas. Our efforts were handsomely
rewarded by the discovery, in a deep sand pocket just
upslope of the site (K8-9) at a depth of only 42.5 m (140
ft), of a unique bronze trident for fishing (KW 4160, fig.
13), a spearhead or tanged dagger (KW 4217, fig. 13) and
a Mycenaean sword (KW 4193, fig. 13). The sword is
very similar to, but slightly longer than, the one found just
2 m (ca. 6.5 ft) downslope in 1985. As with the first
sword, the pommel, or end piece, of the new sword is also
broken off and missing. This is unfortunate, as the
pommel is the key to distinguishing fourteenth-century
Mycenaean swords from those of the thirteenth century.;
Careful examination of preserved structural details on the
two sword handles, however, make us fairly confident that

INA Quarterly 20.3

Photo: E. Greenc, P. van Alfen
Figure 14. A tanged dagger (KW 4517) of Canaanite type not previously noted on the site
(length ca. 31 cm).

both of these weapons are of the fourteenth-century type.
Preliminary study of the ship's cargo and the personal
effects of those on board suggests that the ship was sailing
from somewhere on the Syro-Palestinian coast, perhaps
embarking from Ugarit in northern Syria for a destination
somewhere to the west of Uluburun, before being dashed
against the rocky promontory. The two Mycenaean
swords, however, together with the Mycenaean knives,
chisels, nearly two dozen pieces of pottery, jewelry (beads
of quartz, faience, amber, and glass pendant beads), a
cloak pin of bronze, and the personal seal of a merchant,
point to a near-certain Mycenaean presence aboard the
ship, if only as a passenger or passengers. On the other
hand, the recovery from the wreck of at least five spears of
a type that, when found on the Greek mainland is termed
a "northern type," and a ceremonial stone mace or "axe"
for which the best parallel, albeit of bronze, comes from
the Danube region (see INA Newsletter 17.4: 11, fig. 6;
INA Quarterly 20.3: 13), suggest an Aegean connection
extending all the way into the northern Balkans or Black
Sea basin.
With a final, eleventh campaign of two months' duration
in 1994 we hope to complete the excavations at Uluburun.
We can then direct our energies to the conservation,
restoration, documentation, study, and final publication of
this most magnificent site. The new Kas Uluburun
Shipwreck Exhibit Hall, currently under construction at the
Bodrum Museum of Underwater Archaeology, ultimately
will house all materials recovered from the wreck.

Acknowledgements. As in previous years, the project was
generously funded by the INA Board of Directors, Texas
A&M University, and the Institute for Aegean Prehistory.
The fuel needed for the project was donated by Shell of
Turkey, Ltd., while Cressi-sub of Italy gave us significant
concessions towards the purchase of their diving equipment.
By taking us on an aerial tour in their private plane, Toni and
Maria Pia Bassani of Milan turned five years of dreaming into
the reality of observing and documenting Uluburun from the
Under the overall directorship of George F. Bass, the 1993
team comprised Cemal Pulak, co-director; INA staff Donald
A. Frey, Robin C.M. Piercy, Tufan Turanli, Murat Tilev;

INA staff archaeologist Sheila Matthews; INA counsel James
A. Goold; hyperbaric specialists David Perlman, M.D., and
Tom Sutton, P.A. The excavation would not have been
possible without the enthusiastic and diligent participation of
volunteer archaeologists and art historians Dr. Faith
Hentschel, Jerry Lyon, and Dr. Patricia Sibella; Nautical
Archaeology Program students Peter van Alfen, William
Charlton, Michael Fitzgerald, Elizabeth Greene, Roxani
Margariti, Brendan McDermott, Claire Peachey, Matthew
Pridemore, Edward Rogers, David Stewart, and Mark Smith.
Harun OzdaS of the Bodrum Museum of Underwater Archae-
ology represented the General Directorate for Monuments and
Museums of the Turkish Ministry of Culture. Back in
Bodrum, Uluburun finds continued to be conserved under the
guidance of INA staff conservator Jane Yildinm by G6khan
Ozagagh, Giines (zbay, and Giilser Smaci, with Suzanne
Biehl, and student volunteers Stacey Hild, Barbara van Meir,
and Tuba Tetik. INA staff members Selma Karan, Sheila
Matthews, and Sema Pulak continued to prepare object
drawings for final publication.

Suggested Reading
Bass, G.F.
1986 A Bronze Age Shipwreck at Ulu Burun (Ka4): 1984
Campaign. American Journal of Archaeology 90:
1987 Oldest Known Shipwreck Reveals Splendors of the
Bronze Age. National Geographic Magazine 172
(December): 692-733.
Bass, G.F., C. Pulak, D. Collon, and 3. Weinstein
1989 The Bronze Age Shipwreck at Ulu Burun: 1986
Campaign. American Journal of Archaeology 93:
Haldane, C.
1993 Direct Evidence for Organic Cargoes in the Late
Bronze Age. World Archaeology 24.3: 348-360.
Payton, R.
1991 The Ulu Burun Writing-Board Set. Anatolian Studies
41: 99-106.
Pulak, C.
1988 The Bronze Age Shipwreck at Ulu Burun, Turkey:
1985 Campaign. American Journal of Archaeology
92: 1-37.

INA Quarterly 20.3

Weight, Money, and Weight-Money:

The Scales and Weights from Serge Limani

by Fred Hocker, Sarah W. & George O. Yamini Faculty Fellow

When INA excavated the eleventh-century shipwreck at
Serge Limam between 1977 and 1979, the most readily
apparent components of the site were the major cargos of
glass (both raw and scrap) and wine (represented by over
a hundred amphoras), as well as the hull itself. In the
succeeding years, much of the research on the excavated
material has been devoted to the reconstruction of the hull
remains, piecing together the hundreds of thousands of
glass fragments, and the intensive study of the weight and
capacity systems detectable in the amphoras. But these
large groups of artifacts only tell part of the story of this
medieval trading voyage. The ship was well equipped with
anchors, with tools for repair and foraging, weapons for
defense against pirates, money for purchasing commodities
and supplies, nets for catching fish, and pots for cooking;
the crew and merchants aboard left behind their own
possessions, such as knives, a chess set, combs, even a
sewing kit. All of these small finds have been the subject
of extensive study over the last 15 years, and a large
number of scholars have been involved in the process of
testing, sampling, and analyzing hundreds of artifacts of
clay, metal, bone, glass, and wood. One group of finds,
the weights and scales used to measure commodities and
coin, have revealed some surprising aspects of Mediterra-
nean trade in the eleventh century, and have provided some
clues to the idiosyncratic personality of their owner.
Measurement is essential to the conduct of trade. While
some finished products and raw materials may be sold by

I: A;u


count, this is impractical or impossible for the vast majority
of commercial goods. For these goods, quantities are ex-
pressed in units of length, area, volume, or weight. For
convenience, some dimensional quantities and count may
also be expressed with acceptable accuracy by weight; thus
one may buy wire by either a linear or a weight unit, and
modern nails are often sold by weight if sold in quantity.
Weight is also relatively easy to measure and verify. In
many periods, not only merchandise but also the money
used to purchase it was commonly weighed to assess value.
This was especially true where coins could be easily
altered, where change was made by cutting up coins of
larger denomination, and where money issued by several
different authorities was in mixed circulation. It should
come as no surprise, then, that the Serge Limam ship or
the merchants travelling aboard it were equipped with a
variety of devices for measuring weight: a steelyard, at
least three equal-arm balances, two sets of metal pan-
balance weights, as well as a number of glass coin weights
(the latter are being studied by Dr. Michael Bates of the
American Numismatic Society).

The Steelyard
The steelyard beam, its counterpoise (counterweight), and
the assembly of iron load-hooks, chains, and shackle (figs.
1-5), were all recovered from a large basket of mixed
implements found at the foot of the sternpost. Other items
in the basket included a wide variety of carpenter's tools,

-t '- '
.. C .i %-.~, .: .

.'.t -.
%.I r: e-- I

4 Ell-~ 1 . .

.... .
Photo: F. Hocker
Figure 1. The SerCe Limanm steelyard beam (GW 975) (length ca. 41.2 cm).

INA Quarterly 20.3

GW 975$

Drawing: N. Piercy
Figure 2. Both scales of the Serge Limam steelyard beam.

new and used nails, a lock, and the pans for a fine balance.
The steelyard was found in association with its gear, lying
near the top of the jumbled contents of the basket.
The unequal-arm balance in various forms has been in
use in the Mediterranean since at least Roman times and
remains common in local markets there; when G6khan
Ozagagh and I wanted to measure the weight of a medieval
anchor in the Bodrum Museum of Underwater Archaeolo-
gy, we went up to the local vegetable market and borrowed
a steelyard from one of the vendors. Unlike the equal-arm
balance, with which an unknown weight is balanced against
an equal weight of known amount, the steelyard uses the
principle of leverage to balance the unknown weight against
a single known weight that is moved toward or away from
the fulcrum along a graduated scale. The farther from the
fulcrum the counterpoise is moved along the beam, the
greater the weight it will balance.
The essential components of the steelyard are a beam
incorporating one or more fulcra, a means of suspending
the beam at the fulcra, a movable counterpoise, and a
means of suspending the load from the beam. The beam is
normally a straight bar of bronze with a little more than
half its length covered by the graduated scales. The
counterpoise may be a simple lump of lead or bronze, but
is often decorated or cast in the form of a portrait bust; its
suspension hook is normally hammered out thin and flat to
allow the weight to hang on the thinnest possible bearing
edge, so that its position on the scale can be accurately
determined. In the Byzantine period, the load gear usually

consisted of two or three hooks on chains attached to a
free-swinging shackle suspended from a knife-edged yoke
riding in a groove at the end of the beam nearest the
fulcrum (fig.4). Some balances have a pan in addition to
the hooks and chains, and there is some evidence, in the
form of iron pan remains found in the same tool basket in
which the beam was found, that the Serce Limam steelyard
was so-equipped.
As with the equal-arm balance, one of the keys to
steelyard accuracy is minimal friction in the suspension
apparatus as well as minimal inertia in the beam. In the
Byzantine period, an efficient system of iron knife-edged
fulcrum pins projecting from the sides of the beam and
bearing on metal suspension shackles was in common use;
this can be seen on modern steelyards as well. Accuracy
also depends on the location and uniform spacing of the
scale markings, which can be derived mathematically or
calibrated empirically. The former requires an understand-
ing of the physical principles involved, and theoretical
aspects of equilibrium and the operation and calibration of
balances were the subject of much scientific enquiry in the
Middle Ages; a number of relevant texts and commentaries
have survived in both Greek and Arabic. In practice, these
discussions and texts were largely academic, as most
steelyards were no doubt calibrated empirically by weigh-
ing known or standardized quantities and marking the scale
appropriately. Because the relationship between distance
from the fulcrum and weight is linear, it is not necessary
to calibrate the graduation of every scale on every beam.

INA Quarterly 20.3


Two points can be ascertained on a scale and the desired
divisions interpolated. As with all weighing systems, the
possibility of cheating is ever present (the simplest expedi-
ent for the seller being to shave weight off his counterpoise
after calibration), and so regulation of some sort is desir-
able, although there is no evidence on the Serge Limam
steelyard of any kind of control or certification stamp.

Drawing: Puljk
Figure 3. The counterpoise (GW 1283) (maximum diameter
7.6 cm).

In size, form, marking, and materials, the Serge Limamr
steelyard is fairly typical- of those from the Byzantine
world. It falls into a relatively large group of medieval
steelyards that are between 35 and 45 cm long. Nearly all
of these balances begin with 0 or 1 Xirpa (litra, the Byzan-
tine pound) on the lightest scale, and most have maximum
capacities of 35 to 60 lirrai (11-19 kg). The animal-head
finials, astragals (raised convex rings), and design of the
load gear are actually quite similar to those of the large
steelyard from the seventh-century Yassiada wreck, and the
Greek letters used to denote the markings at 5-litrai
intervals can be seen on nearly all medieval steelyards in
the eastern Mediterranean, even those of Arab origin. The
bronze and iron used in its manufacture are also typical,
although the manner in which the beam is made is unusual,
if not unique. Rather than a single casting filed to shape,
the beam was cast in two pieces and joined by means of a
soldered mortise and tenon at the astragal between the scale
and the fulcrum bar. The joint is not a repair, as the
mortise in the fulcrum bar is part of the original casting.

This steelyard is equipped with two fulcra and is marked
with two corresponding scales, the first reading from 1 to
13 units subdivided into quarters, and the second from 15
to 51 units without subdivisions. The gap of two units
between the scales is a bit unusual, although there are
examples of steelyards with small gaps, usually of a litra
or less, between scales. A two-litrai gap is excessive and
potentially inconvenient, but
doubtless results from an
alteration of the steelyard at
some point in its working life.
Apparently, the owner was not
satisfied with the capacity of
the balance and had it
changed. This involved mov-
ing the fulcrum for the higher
capacity scale closer to the
load collar, filing off the cor-
responding original scale
markings (the remains of
which can still be seen in
places), and re-calibrating the
The unit of weight on this
steelyard, presumably one
litra, is something of a puzzle.
As the instrument is apparently
complete, with load gear and
counterpoise, it is theoretically
possible to calculate the abso-
lute value of the unit for which
it was calibrated. There is
some room for variation here,
as we do not know the original
weight of some of the compo-
nents (the iron load gear is
now represented only by an
epoxy casting, and the counter-
poise is damaged), but we can
at least come close to a mini-
mum. With the existing
weights of beam and counter-
poise, and a reconstructed
weight for the load gear, the Photo: F. Hocker
"pound" for the lighter scale is Figure 4. The yoke, free-
550 g, that for the heavier swinging shackle, and
scale approximately 590.5 g chains (GW 1282) (pre-
(by comparison, the pound in served overall length ca.
common use today weighs 454 33.5 cm).
g). The discrepancy between
the scales, of approximately

INA Quarterly 20.3

7.5 per cent, is not the result of weight loss in the beam or
counterpoise, nor can it be attributed to an inaccurate
reconstruction of the weight of the load gear. It is instead
due to an inherent error in the graduation of the beam,
probably resulting from the alteration of the heavy scale.
At a litra of 590.5 g, the maximum capacity of the steel-
yard is about 30 kg. Although the original marks on the
heavy scale are all but obliterated, it is possible to estimate
the original capacity before alteration; calculation reveals
that the heavy scale began at 12 litrai and had a maximum
capacity of 40 or 41 litrai (22-22.5 kg).
It seems apparent that the original "pound" or litra for
which the Serce Limam steelyard was calibrated was some-
thing more than 550 g, but probably not more than 650 g.
Unfortunately, a litra in this range does not seem to
correspond to any common medieval weight standard of

clearly derived from a litra of 318 g or a bit more (see
below). In order for the steelyard to weigh in Byzantine
litrai, the counterpoise should weigh approximately 1 kg -
a little more than half of its actual weight.
If the original counterpoise is indeed missing, how can
any merchant have tolerated the use of an oversized
counterpoise, one so oversized that the error was easily
detectable? This counterpoise is clearly intended for use
with a steelyard, and was found in close proximity to the
beam, so it is difficult to dispute their association. Could
the provenience of the Serge Limam counterpoise be mis-
leading? Is it possible that the steelyard and its associated
gear were not being carried as usable implements, but as
scrap, due in part to the loss of the counterpoise, and that
the counterpoise was also scrap carried in the same basket?
This seems far-fetched.

Photo: F. Hocker

Figure 5. Detail of the steelyard.

Byzantine, Islamic, Balkan, or Italian usage, even though
the steelyard and its markings are clearly Byzantine in
nature. The Byzantine litra, the standard commercial unit
of the period, ranged between approximately 312 and 324
g in the eleventh century and was ultimately derived from
the Roman pound of only slightly more. It is possible that
the steelyard represents an as yet unknown Byzantine unit,
or a unit in use in a peripheral area, such as the Balkans.
Another possibility that must be considered is that the
counterpoise is not original to this steelyard. It is consider-
ably larger than the counterpoises found with other medi-
eval steelyards of similar dimensions and capacities, and
the maximum load of ca. 30 kg seems excessive for the
relatively light suspension and load gear. In addition, the
other weights on board the ship seem to be based on
common standards, with the large set of disc weights

A final possibility is that the counterpoise has lost
considerable weight, up to 250 g, and was intended to
increase the capacity of the instrument by doubling (a
practice not unknown in the Middle Ages; double pounds
were sometimes called "royal pounds"), so that a reading
of 10 litrai would signify an actual weight of 20 litrai.
The moving of the heavy fulcrum and re-cutting of the
relevant scale demonstrate the owner's desire for increased
capacity, so another alteration to increase capacity further
is not out of the question. Amphoras exceeding 50 litrai
were quite common in this period, and several of the
amphoras carried on the ship's last voyage weighed when
full considerably more than 51 Byzantine pounds but less
than 102 pounds. Perhaps even more significant is Fred
van Doorninck's study that shows that the amphoras on the
wreck were carefully manufactured in standardized sizes

INA Quarterly 20.3

graduated by weight (see "Giving Good Weight in Elev-
enth-Century Byzantium: The Metrology of the Glass
Wreck Amphoras," INA Quarterly 20.2: 8-12). A steel-
yard too small to weigh the common seagoing cargoes of
the day would be of limited use on the ship.
If the last scenario is accurate, then our steelyard had a
long and colorful career behind it before the ship sank.
Fabricated in a Byzantine workshop and calibrated to the
standard pound of the day, it originally had a capacity of
approximately 40 litrai. At some point, the owner had
need of a larger instrument and had it altered, possibly
doing the work himself, so that it could weigh up to 51
litrai, presumably with the original counterpoise. Even this
was not enough, and the original counterpoise was replaced
to double the capacity, to 102 litrai. Could the last alter-
ation be directly related to the instrument's employment on
a merchant ship carrying amphoras and other-bulk goods?

The Equal-arm Balances
The beam of a fine balance (fig. 6) was found approxi-
mately 1 m downslope from the basket of mixed imple-
ments, not far from a concentration of barrel-shaped
weights, but the pans were found, crushed, in the basket
itself. It is likely that the beam originally lay in the basket
in its own container (probably a wooden box) with the pans
and was thrown or washed downslope during or after the
wreck. The arms of a large folding balance were found
over 2 m apart, between 8 and 9 m downslope from the
after part of the keel. A fragment of a third balance beam
lay over 2 m farther downslope.
The equal-arm balance is the oldest and simplest of
weighing devices. The essential components of the balance
are a beam, a means of suspending it on a free central
fulcrum, a means of suspending the known and unknown
weights from the ends of the beam, and a set of masses of

Drawing: N. Picrcy
Figure 6. The fine balance (GW 455) (beam length 15.7

Drawing: F. Hocker
Figure 7. Reconstruction of the folding balance (GW 522,
523), folded up. Only the two arms were recovered (original
open length ca. 50 cm).

known weight, commonly called, simply, weights. A
refinement added in Roman times was a vertical indicator,
attached to the beam, which made it easier to determine
when the beam was exactly horizontal. The accuracy of
the instrument depends primarily on how closely the pan
weights match their theoretical standard, but the precision
of the balance depends on the construction of the balance
itself, with knife-edged fulcrum pins offering the same
advantages that they do in the steelyard. A well made
balance of the Middle Ages was capable of distinguishing
values (that is, producing a visible movement of the beam)
of less than 0.1 g, considerably smaller than the smallest
weights in common use.
The Serge Limam fine balance is reasonably typical of
small balances from Byzantine sites. Small pans were
normally suspended by silk thread, and one of the hemi-
spherical pans bears stains that may be from fine cord.
Such balances were typically used to measure precious
metals, coins, and some expensive commodities sold in
small quantities. The unusually deep pans may also have

INA Quarterly 20.3

Figure 8. Some of the disc weights (note the inscribed conc

been useful for weighing powdered or granular substances,
perhaps even liquids. Fine balances of this size and the
weights that accompanied them are common finds on
medieval sites (including graves) in northern Europe, but
medieval Byzantine fine balances, while widely distributed,
are comparatively rare. A balance of this size is capable
of great accuracy, on the order of a twentieth of a gram.
A balance for weighing money and small goods was an
individual possession, and was normally kept on the person
and protected carefully. Small balances such as the Serge
Limam example were easily transported with their weights,
normally in a fitted wooden case, but larger instruments for
heavier loads could be unwieldy. The folding balance
represents the most common medieval solution to the
problem. The arms are hinged to the central section of the
beam, which also carried the fulcrum pin and vertical indi-
cator, and could be folded into a narrow, elongated pack-
age (fig. 7). This innovation first appeared in Persia or
Mesopotamia in the early Middle Ages and spread as far as
Scandinavia by the end of the ninth century. Such balances
were ideal for weighing groups of coins, as medieval
illustrations show. The Serge Limam balance represents
one of the largest known examples, with a reconstructed
beam length of at least 45 cm.
A folding balance is less accurate than a comparable
fixed-arm balance, as the arms must be heavier, and there
is a minute but unavoidable amount of slop in the joints,
especially as the balance wears with age. The reduced
accuracy is less significant when weighing silver, due to its
lower value per unit of weight, and it is therefore not sur-
prising that folding balances are most common in northern
Europe between the eighth century, when gold coinage was

abandoned, and the thirteenth century, when it was
re-introduced. The use of a folding balance on the
Serge Limam ship, especially a balance of such
great size (and presumably minimal accuracy),
suggests either that its primary use was for weigh-
ing commodities of relatively low value, or that the
owner felt no need for extreme accuracy.

The Weights
The large number of metal weights (all but one are
of bronze) found on the wreck fall into two easily
distinguished groups: 15 discs and 14 barrels. The
former (figs. 8, 9), ranging between 20.64 and
469.4 g in preserved mass appear to represent only
six different denominations, with as many as four
D. Fpy examples of one denomination, and seem to be a
entric collection assembled from several sets or partial
sets with distinctly different styles of decoration.
Much the same is true of the barrel-shaped weights,
ranging in preserved mass from 3.06 to 170.5 g; there are
at least seven and perhaps eight different sizes (figs. 10,
11). Each of these groups has something very interesting
to tell us about the type of commerce in which the ship was
engaged, and even
throws some light on
the personality of the
owner, who was prob-
ably the man who
owned the steelyard.
The disc weights
are typically Byzantine
in proportion and
decoration, and allow-
ing for slight loss of
mass due to corrosion,
fit readily into/a nor-
mal progression of
Byzantine ounces,
with representatives of
1, 2, 3, 6, 12 and 18
ounces in the set.
This combination is all
that is needed to
weigh any number of
whole ounces up to .
the total weight of the
group, 71 ounces
(almost 6 litrai, or
1.88 kg). This pro- w: N. Pie
gression is most com- Figure 9. A six-ounce disc weight
only found in groups (GW 494) (preserved mass 153.55

INA Quarterly 20.3

of weights used in systems based on the 12-ounce
pound, such as the Romano-Byzantine litra.
According to the preserved mass of our weights,
this litra weighed at least 318.5 g, and was proba-
bly nearer 320-322 g. This is quite similar to the
value of the litra derived independently by Fred
van Doorninck from the capacity weights of the
amphoras on the wreck. It shows that even in the
eleventh century there was still a great deal of
consistency in Byzantine weights, and that the litra
was still very near its original Roman value of
approximately 327 g. Earlier studies of Byzantine
weights and coins suggested that the litra had
declined to as low as 318-319 g by this time, but
the Serge Limam finds show that merchants were
successfully "holding the line" against this trend to
a greater degree than formerly suspected.
The barrel weights are actually faceted, rather
than biconical (with two important exceptions). This shape
is considered typical of Fatimid Islamic metal weights, and
the dimensional proportions of our weights indicate that
they were probably made in Palestine rather than Egypt.
Determining the denominations of these slightly smaller
weights is not as easy as it is for the discs, as the barrel

,' .

Drawing: N. Piercy

Figure 11. A typical
barrel weight (GW 516)
from the wreck, probably
a 24- or 25-dirhem weight
(preserved mass 70.28 g).

weights do not fall as clearly
into groups on the basis of
preserved dimensions and
mass. There is also the prob-
lem of the Islamic weight
system itself. Unlike the
Byzantine ounce, which was
based on the weight used for
precious metal and was direct-
ly linked to the weights of
coins, the Islamic dirhem had
two theoretical weights, one
for silver (2.96 g), derived
from the gold dinar standard,
and one for non-precious
commodities (3.125 g). The
difference is slight but signifi-
cant. The Serce Limani
weights are too corroded to
allow clear attribution to one
standard or the other, but there
is the distinct possibility that
weights based on both stan-
dards have been combined in
this very motley group. There
is also some confusion about
the denominations due to the

Figure 10. Some of the barrel weights.

nmow; U. Prey

apparent mixture of two different weight systems, one
based on units of 10, which is more typical of Islamic
weights, and one based on units of 12, which is consistent
with the Byzantine system. Could it be that the latter are
from an Islamic region on the Byzantine frontier, such as
Syria, where a 12-unit system allowed easier interchange
with Greek merchants to the north?
Two of the barrel weights are smoothly biconical, not
faceted. They resemble late Roman/early Byzantine weights
of similar form quite closely, right down to the decorative
rings turned in their upper faces. They also do not fit into
the dirhem system very cleanly, as they are slightly over
weight. In fact they are very nearly the proper masses for
a 10- and a 20 dinar-weight. The problem is that the
biconical form was not typically used for dinars; rectangu-
lar or octagonal forms were preferred, probably so that
dinar and dirhem weights could be easily distinguished.
Could these two weights be old Byzantine weights added to
a set of Islamic weights precisely because they were so
close to a good dinar multiple? A similar practice was
common earlier in the Islamic world. Before the Fatimid
caliphs began issuing their own dinar and dirhem weights,
Abbasid rulers simply validated existing Byzantine weights
and reissued them.

Use and Ownership
The people on board the Serge Limam ship were well pre-
pared to conduct commerce in both Byzantine and Islamic
centers of trade. Heavy goods, such as amphoras, could be
weighed with the steelyard, while fine goods and coinage
could be weighed with the equal-arm balances and theit
associated weights. The available weights could even ac-
commodate the dichotomy of the Islamic weight system:

INA Quarterly 20.3

the commercial standard, represented by at least some of
the bronze barrel weights, is complemented by the glass
coin weights (which are all Fatimid issues for dinars and
dirhems [fig. 12]), and by the Byzantine-style barrel
weights, which represent the monetary standard. Byzantine
coins, since they were the basis of the commercial weight
system, could be weighed with the same weights as fine
goods. It may seem a bit odd that no Byzantine coin
weights were found, but in a unitary weight system, in
which coins were weighed in the aggregate, they were

Figure 12. Two of the 16 glss coin weights of Fatimid issue foi
the wreck

Medieval gold and silver coins were not money in the
modern sense, but merely a convenient format for the
transportation and exchange of bullion. In medieval
Arabic, the word for payment had the literal meaning of
"weighing." While medieval copper coinage might be
accepted at face value, the value of gold and silver coinage
depended directly on its precious metal content. This in
turn depended on the fineness (purity) and total weight of
the coin. The best coinage was struck from fine (pure)
metal, but in many of the economies of medieval Europe,
debasement was a common method of raising revenue, such
that coinage was frequently recalled and reissued. Fineness
could be assayed by use of the "balance of wisdom," as it
is called in Arabic sources (a water balance, i.e., a means
of measuring specific gravity), but the device was hardly
portable and in practice the merchant was forced to rely on
the honesty of the issuing authority and his own shrewd-
ness. In the markets of northern Europe, debasement and
the proliferation of regional or independent mints kept mer-
chants on their toes. Fortunately for the merchants on

board the Serge Limam ship, the eastern Mediterranean
was dominated by a small number of issues, Byzantine
nomismata and Fatimid dinars, which were themselves
relatively stable until well after A.D. 1025, about the time
the ship sank.
Even where the currency was fine and stable, however,
wear and deliberate defacement reduced the total weight of
coins, and the inevitable variation in weight of hand-struck
coinage made weighing necessary. Long-term wear of
coinage could also prompt issuing authorities to lower the
weight of new coins slightly, so that they would match the
existing, worn coins in circulation. In addition,
change was sometimes made by clipping pieces off
of coins. In fact, 15 cut gold dinar fragments were
recovered from this wreck. These fragments were
not simple fractions of whole coins, such as halves
and quarters (four complete quarter-dinars were
also found on the wreck), but irregular bits, and
thus depended purely on their bullion content for
their value.
In this sort of "weight-money" economy, it
behooves the merchant to protect himself from loss
by verifying the value of currency. Coins could be
weighed and verified by bankers and money chang-
ers in most larger commercial centers, but a travel-
ing merchant had to have his own means of weigh-
ing, and he had to be familiar with the fineness of
D. Fr the issues with which he might be paid. The
und on weight of individual coins could be checked with a
fine balance and a set of official coin weights,
which were often made of glass. Yet this was
tedious and impractical, and money for large transactions
was normally weighed in bulk. In such transactions, the
value of the group was expressed as its total weight, rather
than as the number of coins. Transactions recorded in
medieval documents show that groups of gold coins were
described in terms of both the number of coins involved
and their value by weight in "standard" gold coins. For
example, in 1050-1051, a merchant in Tunisia sent to
Egypt a purse containing 53 /2 coins valued at only 49
23/24 dinars.
A final issue to be resolved is ownership of the Serge
Limani weighing equipment. Was it personal property of
one or more merchants, or was it part of the ship's equip-
ment? The evidence strongly favors the latter. All of the
weights and weighing implements were found quite far aft,
in an area that also produced carpenter's and foraging
tools, coins, locks, and some weapons. All of these items
can be associated more easily with the ship than with
individuals. The stem is also the traditional stowage area
for a variety of shipboard property, such as valuables

INA Quarterly 20.3


(money) and bosun's stores, both the steelyard, and the pans
for the fine balance were found in the basket that also con-
tained a large selection of carpenter's tools, a lock, and a
quantity of new and used iron nails. The nails are the
same size as those used in the construction and repair of
the ship, and the tool assemblage is similar to other
shipboard tool kits from Mediterranean wrecks. It is
therefore most likely that the tools are part of the ship's
equipment, and perhaps by association, so is the steelyard.
The large, folding balance was found farther down
slope, broken and scattered, and is less clearly associated
with the ship's equipment. Folding balances were devel-
oped precisely because they were personal possessions and
had to be easily transportable, so it is possible, even likely,
that it belonged to an individual. It is also possible that its
owner had ties with the ship, perhaps as captain or owner;
a medium-sized balance was needed to fill the gap in
capacity between the steelyard and the fine balance.
Moreover, the folding balance is of the correct size for use
with the disc weights and was probably the most effective
means of weighing money. If the folding balance is not
associated with the ship and its equipment, then perhaps the
balance of slightly smaller size represented by the beam
fragment is.
The pan weights themselves tell an interesting story.
That the barrel weights were used with the fine balance and
coin weights, as well as with a quantity of coins, can be
argued on the basis of provenience. Provenience is of less
help with the disc weights; they were more tightly concen-
trated on the site (suggesting a more robust container) and
farther removed from the tool basket, although other
elements of ship's equipment, notably a large felling ax,
were nearby. The heterogenous nature of both sets, with
large numbers of extra weights culled from partial sets,
suggests that both were owned or collected by the same
individual. They also hint at a personal interest in weights
that goes beyond the utilitarian. The inclusion of weights
of different standards within each set parallels the minimal
concern for accuracy evident in the alteration of the
steelyard and its consequently incompatible scales. This
inconsistency, which is also seen in the relatively wide
range of pounds nominally conforming to the same stan-
dard in a given period, is a reflection of the personal level
at which business was done in the centuries before the
changing structure of medieval trade made more consistent
standards necessary. It also seems to me to reflect the
personality of the merchant who owned the weighing equip-
ment. Maximum capacity was his primary goal, not

Suggested Reading
Goitein, S.D.
1967 A Mediterranean Society I: Economic Foundations.
University of California Press, Berkeley.
Kisch, B.
1965 Scales and Weights: A Historical Outline. Yale
University Press, New Haven, CT.
Knorr, W.B.
1982 Ancient Sources of the Medieval Tradition of Me-
chanics: Greek, Arabic and Latin Studies of the Bal-
ance. Institute e Museo di Storia della Scienza, Fir-
enze, Monograph 6. Florence.
Moody, E.A., and M. Clagett (editors)
1952 The Medieval Science of Weights (Scientia de Ponde-
ribus): The Treatises Ascribed to Euclid, Archimedes,
Thabit ibn Qurra, Jordanus de Nemore and Blasius
of Parma. University of Wisconsin, Madison.
Petrie, W.F.
1926 Ancient Weights and Measures, Illustrated by the
Egyptian Collection in University College, London.
Sams, G.K.
1982 Weighing Implements. In Yassi Ada I. A Seventh-
Century Byzantine Shipwreck, edited by G.F. Bass
and F.H. van Doorninck, Jr., pp. 202-230. Texas
A&M University, College Station.
Schilbach, E.
1970 Byzantinische Metrologie. C.H. Beck, Munich.
Spufford, P.
1988 Money and its Use in Medieval Europe. University
Press, Cambridge.

INA Quarterly 20.3

Relics of the Revolution


A Schooner Called Water Witch

The 1993 Field Season on Lake Champlain

by Kevin Crisman

Way back in 1981, during my first week as an M.A.
student in the Texas A&M Nautical Archaeology Program,
I composed a letter to my friend Arthur Cohn in Vermont.
What we needed, I wrote, was a realistic plan of archaeo-
logical research for Lake Champlain. Our successes of the
previous two years the recording of the earliest known

Map: K. Criman
Locations of 1993 Lake Champlain projects.

wreck of a steamboat in 1980 and the discovery of a
sunken War of 1812 ship in 1981 were a good start, but
we needed to systematize our approach. Lake Champlain
was clearly the perfect place to study the history of
American ships and inland navigation: what we now
needed to do was find a wreck of every significant vessel
type and from every major period in the lake's history.
The data we gleaned from these wreck studies could then
be turned into books on the nautical archaeology of Lake
Champlain, books that would describe 300 years of ship-
building, warfare, commerce, and everyday life. Above
all, I said, we had to be realistic about what we were
attempting to do. The fieldwork, research, analysis, and
publication of all these wrecks could possibly take as long
as five years to complete.
Like economic planners in the former Soviet Union, Art
and I have watched our "5 year" plan extend into a "10
year" plan, and onward into a "15 year" plan. Were we
naive in 1981? Yes, hopelessly. Would we do any of it
differently? I don't think so. It has been an incredible 14
years of astounding underwater finds, hard work, and
generous dollops of good luck. In this time we have
documented prehistoric dugout canoes, colonial bateaux, a
British sloop from the French and Indian War, Benedict
Arnold's flagship from the Revolutionary War, a 20-gun
U.S. Navy brig from the War of 1812, sailing canal boats,
steamboats, a floating drydock, and the only known wreck
of a horizontal-treadwheel horse ferry. The lake has been,
for us, a cornucopia of beautifully-preserved and unique
In recent years the work on Lake Champlain has moved
to a new level of productivity and sophistication, thanks in
large measure to the participation of the Institute of Nauti-
cal Archaeology and Texas A&M University. And 1993
was our best year yet, for with the assistance of a team of
graduate students from the Nautical Archaeology Program
we completed three very ambitious projects: the recovery

INA Quarterly 20.3

Reconstruction of the floating bridge between Fort Ticonderoga and Mount Independence.

and conservation of Revolutionary War artifacts from the
waters around Mount Independence, the in situ recording
of the steamboat-turned-schooner Water Witch, and a
preliminary study of the sidewheel steamship Champlain.
Our work on the Champlain will be described by project
field director Liz Baldwin in Volume 21.1 of the Quarterly.

Mount Independence Artifacts of the American Revolu-
During the perilous years 1776 and 1777 the future of
America's war for independence rested in large measure
upon the twin defensive works of Fort Ticonderoga and
Mount Independence near the southern end of Lake Cham-
plain. Here the American army was expected to stop an
imminent British invasion from Canada, an invasion that
threatened to split the rebelling colonies in two and force
them to surrender. Earthwork gun batteries were placed to
repel boats filled with redcoats; barracks, hospitals, maga-
zines, and other buildings were constructed to house
American troops and war materials; and a floating bridge
was extended across the lake to assist the movement of
soldiers and munitions between fortifications. By the

Photo: K. Criman
View across the lake to Fort Ticonderoga, along the path of
the floating "great bridge. "

spring of 1777 the defenses seemed ready.
The strength of Ticonderoga and Mount Independence"
was illusory, however, for through mismanagement and
faulty intelligence there were not nearly enough men to

INA Quarterly 20.3

Photo: S. Paris
Nautical Archaeology Program students David Robinson, Peter
lifted from the lakebottom.

stem the British tide; the American troops on hand were
poorly equipped and riddled with sickness. When the
superior British army under General John Burgoyne began
surrounding the two fortifications in early July, 1777, the
commanding American General Arthur St. Clair decided
that there was nothing to be gained, and a whole army to
be lost, by trying to resist a siege. On the night of July
5-6 his army decamped in the darkness, abandoning
everything and fleeing southward on foot and by boat. It
may have looked disgraceful, but St. Clair's retreat saved
the army to fight another day, and fight it did, capturing all
of Burgoyne's army a few months later at the Battle of
Saratoga. It was perhaps the greatest single American
victory of the war.
Cohn and I first surveyed the
waters between Ticonderoga and
Mount Independence in 1983,
finding the caissons that had an-
chored the floating bridge and a
few scattered artifacts. These
were entered on the Vermont
state archaeological inventory
and the site was monitored until
1991, when a relic-hunter from
Indiana was arrested and con-
victed for mining artifacts from
the lakebottom. The state decid-
ed that it had to take further
steps to study and protect the
site, a goal that coincided nicely
with our interest in returning to
investigate this historic water- -- r
front. The work began in 1992 "
with intensive surveys that re- The cast-iron gun can


ca. 47 cm = 18. 75 inches).

Photo: K. Criummn
Hitchcock, and Joe Cozzi watch as the 12-pounder cannon is

vealed an iron cannon, mortar bombs, bar shot, entrench-
ing tools, and a complete musket. Recording of the bridge
caissons was also undertaken at this time (see INA Quarter-
ly 19.4: 17-21).
The results of the 1992 study so impressed the Vermont
Legislature that it passed a special one-time appropriation
to continue the survey in 1993, and to recover and con-
serve all of the finds for display in Vermont museums. It
was a tall order: complete a survey in the black, muddy
water conditions around the Mount, as well as build a
conservation facility from scratch and treat all of the arti-
facts in just a few months. Fortunately, we had strong
allies in the form of eight highly trained and hard-working
graduate students from the Texas A&M Nautical Archaeol-
ogy Program, as well as a group
of enthusiastic undergraduates
from the University of Vermont.
r.J With this crew and a sprinkling
of experienced Lake Champlain-
area divers and volunteers, much
Should be accomplished in three
busy weeks, from June 14 to
July 1.
We employed the same diver
search techniques in 1993 that
we had depended upon during
the previous year, which includ-
ed examining the bottom visually
:_F. (visibility sometimes extended
up to several inches!), by touch,
W2 and most effectively, with hand-
Yi held metal detectors. Sections
.. Photo. Rbinso of lake floor were surveyed with
age wheel (diameter anchors and guide tapes to allow

INA Quarterly 20.3

Drnwing: S. McLaughlin
(above) and detail of the unidentified crest on the barrel (below).

Drawing: S. McLaughlio

thorough, overlapping linear or circular search patterns.
All surveyed areas and finds were mapped by using transits
to triangulate the locations of buoys or a stadia rod. The
going was slow, but the relatively warm water and shallow
depths (averaging 3 m [10 ft] to no more than 7.62 m [25
ft]) permitted long dives of up to three hours.
Whenever we encountered a cluster of artifacts with a
metal detector, survey markers were placed over the find
and a diver gently removed the overlying mud by hand.
Once the full extent of a feature was uncovered, the snail's-
pace process of plotting the location of each artifact began.
One of the biggest problems we had to deal with in this

regard was seeing the numbers on the measuring tapes and
writing them on the clipboard in next-to-zero visibility.
Nautical Archaeology Program student David Robinson
spent two days mapping a scatter of grenades, mortar
bombs, and cannon shot that was about 2.43 m (8 ft) in
diameter; his carefully measured underwater drawing of the
entire feature was about 3 inches in diameter, with numbers
each the size of a flea's hind leg. Asked why he worked
in such a small scale, Dave reported that it seemed large
enough underwater when he was holding a clipboard,
flashlight, and pencil 1 inch away from the faceplate of his
The high point of the field work came on June 28, when
"feature 1," the 12-pounder cannon discovered in 1992,
and a cast-iron gun carriage wheel were hoisted from the
water by a crane parked on the deck of the Lake Cham-
plain Transportation Company ferry Vermont IV, which
voyaged down to Mount Independence specially for the
occasion. Also on hand was the Lake Champlain Maritime
Museum's replica Revolutionary War gunboat Philadelphia
II, which fired a series of salutes from her muzzle-loading
swivel guns as the cannon rose above the surface. It was
a memorable day for everyone involved.
The cannon, after cleaning, clearly exhibits the broad
arrow and British weight stamps on its breech, but the crest
on the barrel, which indicates the country of origin, has not
yet been identified. One of the cannon's trunnions had been

INA Quarterly 20.3

J ..- - -- .. ...-

1 0o Z 1

1V IT. -I 1992-1993
_________________________________T_____________________T_____________________________ ~

The 12-pounder cast-iron cannon

knocked off with a heavy tool and was found nearby on the
lake bottom; removing a trunnion was a common way to
render a gun unusable when abandoning it.
There was even more excitement. The 12-pounder was
surrounded on the lakebottom by 36 8-inch mortar bombs.
Were they, we wondered, filled with gunpowder, and if so,
was it still dry? The possibility of the bombs still being
dry inside after all this time seemed remote, but getting
blown up by Revolutionary War ordnance was definitely
not on our "to do" list for the 1993 season. I selected a
few sample bombs from the pile, and Art Cohn and
Nautical Archaeology Program student John Bratten lifted
them one-by-one into the inflatable boat, then gingerly
scraped away the corrosion around the fuse holes and pried
out the wooden plugs. Some of the bombs gave a little hiss
when the plug came loose as 200-year-old air escaped;
remarkably, they were perfectly dry inside. Fortunately,
all the bombs that we recovered contained no powder, and
apparently they never had.
Overall, the amount of material we encountered on the
lakebottom in 1993 was considerably more than we had
expected to find, and together with the finds from 1992
comprised a large and varied collection of Revolutionary-
War-era ordnance and tools. Recovered pieces of cast-iron
shot included 42 round shot (for guns firing 4- to 24-pound
shot), 41 bar shot for use against a ship's rigging (for guns
firing 6- to 24-pound shot), 26 mortar bombs (6- and 8-
inch diameter), and several hundred smaller swivel and
grape shot. Much of the grapeshot came from two rectan-
gular wooden shot boxes that were found lying side-by-side
on the bottom. Four hand grenades and a lead musket ball
rounded out the collection of munitions. Other weaponry
included three bayonets, and the musket found in 1992
which, during conservation, was found to have "Pomeroy,"
the name of a Connecticut gunsmith, stamped on its lock
Tools found on the site consisted mostly of entrenching
spades (a total of 18, with many different styles of manu-
facture), two shovels, two picks, two axes, and a file.
Objects relating to the daily life of the troops at Mount
Independence included a brass skillet and many pieces of
cast-iron cooking pots, leather shoe fragments, ceramic
shards, whole and broken dark-green-glass alcohol bottles,
clay pipe stems, a deer antler, and animal bone fragments.
Out in deeper water, not far from one of the bridge
caissons, Nautical Archaeology Program student Joe Cozzi
located the partly-buried remains of a flat-bottomed scow,
a common vessel type on inland waters but one that is
seldom studied by archaeologists. Test excavations around
the sides of the vessel permitted Cozzi to record its
construction details and reconstruct its appearance, but no


I ..,
ML Mri"-

Photo: D Robinon
Top: John Bratten opens the first of
the mortar bombs, none of which
contained powder (bottom).

INA Quarterly 20.3

diagnostic artifacts were found
that would permit us to date the
craft. Nevertheless, the scow's
location near the bridge, in an
area that saw much ferrying
activity during the 1776-1777
period, strongly suggests that it
may be of Revolutionary War
Locating, mapping, and re-
covering these materials was the
relatively easy part of the pro-
cess; John Bratten's task of
building a conservation lab from
scratch and getting it into full
operation in just a few weeks
was one of the truly impressive
feats of the 1993 season. When
John arrived at the Lake Cham-
plain Maritime Museum in late
May the "lab" consisted of a
few stakes pounded into the
ground near the museum's boat-
building shed. As the founda-

.. a. v

t r

-- '-
Pholto: D. Robinon
An entrenching spade (top) and a cast-iron cooking

tion was poured and carpenters
erected the walls and roof of the
lab, John ranged over the Ver-
mont countryside, scrounging
parts from junkyards, borrowing
equipment, and negotiating for
tools and chemicals. By the end
of June artifacts were bubbling
away in electrolysis vats, field
school students and local volun-
teers were cleaning and scrap-
ing, and the conservation pro-
cess was in full swing. John
worked until the end of August,
and then Dave Robinson took
over the direction of the lab; by
the end of October nearly all of
the finds had been conserved
and were ready for exhibit.
The Mount Independence
study is now in the research and
writing phase, with both interim
and final reports in various
stages of completion. Nautical

a III___ I

.....otec.mad MT INDEPENDENCE
JUNE 1993


I 0 10 0 30

010 U F, z 3 4 5 6 7 8e g9

Co. 2-tI
Drawing: J. Coni
Reconstruction of the scow that may date to the Revolutionary War period.

INA Quarterly 20.3

SDr wmg Cm

.. . .- -. . --

";--~~~~~~~~~~ -.... .: ; :- -"i~ -- -- -- -- -: -- --:... ..

-- -

Drawing: K. Crisman

Reconstruction of the schooner Water Witch, A

Archaeology Program student Scott McLaughlin has
undertaken the study of the entire artifact collection for his
Master's thesis, and over the next few years these artifacts
and story they tell of the desperate summer of 1777 will be
featured in articles, books, and other publications, as well
as in museum exhibits around the state of Vermont.

Schooner in Fourteen Fathoms the Water Witch Project
The second major accomplishment of the 1993 season
began on July 6 and continued until July 16. This was the
recording of Water Witch, an 80-foot-long (24.39 m)
schooner that sank in 14 fathoms of dark and very cold (if
relatively clear) water near Diamond Island, in the central
part of the lake. This project differed a great deal from the
preceding work at Mount Independence.
Water Witch had a long and varied career that makes her
beautifully-preserved hull of particular interest to nautical
archaeologists. She was launched in 1832 as a steamboat
by veteran steamer captain Jehaziel Sherman. In her first
three years she navigated Champlain's waters as a passen-
ger-and-freight carrier that competed with the reigning
steamboat monopoly on the lake, the Champlain Transpor-
tation Company. Sherman's challenge was not an entirely
noble cause, however, because his chief purpose in running

an opposition boat seems to
have been to force the mo-
nopoly to buy out his opera-
tion at a fat profit. The
Transportation Company
finally did so in 1835 and,
due to an over-abundance of
steamers on the lake, remov-
ed Water Witch's engine and
boiler and converted her into
a centerboard-equipped mer-
chant schooner. She sailed
in this capacity for three
decades until she was
knocked over by a squall and
sank with a load of iron ore
in April of 1866. According
to all accounts the sinking
was nearly instantaneous.
The captain's infant daughter
drowned in the after cabin
during this tragedy.
Art Cohn and I conducted
a preliminary and very brief
survey of Water Witch during
a stormy week in May of
1990, and planned to con-

tinue the study when time and funding permitted. In 1993,
with special assistance from then-INA Board Chairman Ray
H. Siegfried II and the Lake Champlain Basin Program, we
scheduled a two-week project to document the external
appearance of the wreck and take off her lines record
her three-dimensional shape and thereby gain a better
idea of the hull form of this early lake steamer. Our goals
were extremely ambitious given the time we had, the depth
and the dark conditions on the wreck, as well as the unpre-
dictable weather that can so easily disrupt a dive schedule
on this part of the lake. Nevertheless, we had several
factors working in our favor, including use of the research
vessel Neptune, commanded by experienced lake captain
Fred Fayette, and a small but crack crew of eight Nautical
Archaeology Program students and local divers. We were
also fortunate to have the valuable assistance of the Nauti-
cal Archaeology Program's ship expert, Dr. Fred Hocker.
Water Witch is perhaps the prettiest vessel that I have
yet had the privilege of recording, but she is a most
haunting wreck to visit. She sits upright on the bottom in
perpetually dim, grey-green light, her sides rising high and .
dark above the smooth, silty lakebottom. Whenever we
descended upon the wreck, one of the first objects to
appear in the beams of our dive lights was the bowsprit,

INA Quarterly 20.3

extending out into inky blackness. At the schooner's bow
could be seen two folding-stock anchors, one lying on the
bottom off the starboard side, the other still ready for
deployment on the port side of the deck. In the center of
the bow sits a wooden windlass for raising the anchors, its
drums worn from years of reeling in anchor chain. Two
open hatches on the main deck provided access to a hold
now filled with chunks of iron ore and covered over with
a thick layer of silt. o L
At the after end of the schooner the quarterdeck rises
above the level of the main deck, and is surrounded by a a
low open railing. A flashlight-assisted peek into the open
companionway revealed the stern cabin, complete with a
cast-iron wood-burning stove, reeded panelling painted W
white, and what appears to be the remains of bunks and
storage lockers. About 9 m (30 ft) off the starboard 8
quarter of Water Witch lies the schooner's small boat,
dragged to the bottom by a towrope that has long since
fallen to pieces and drifted away.
We did not, unfortunately, have much time for sight-
seeing on Water Witch. The project schedule allowed for
only one dive per diver per day, and each diver descended
with a long list of measurements and sketches to record in
just a few short minutes on the wreck, which lay at a depth
of 84 ft (25.6 m). During the 10 days of diving the dimen-
sions of endposts, rails, spars, scuppers, planking, hatches,
the companionway, tiller, and rudder were carefully taken,
along with hull offsets, video footage, and several rolls of
color slides. The recording of the schooner's three-
dimensional shape was greatly assisted by our use of an
electronic angle-measuring device, called a goniometer,
devised by Joe Cozzi.
All in all, it was a nip-and-tuck race to see if we could
complete all of our objectives on time, but the weather
held, all of the divers did an exemplary job of getting the
required information, and on the last day we recorded the C
final, essential measurements necessary to reconstruct the o s
lines of Water Witch on paper. In 14 years ofarchaeologi-
cal work on the lake, this was probably the most produc- l
tive project we have ever run in terms of the amount of
data collected per dive.s
What did we learn about the vessel? The reconstructed o u
lines of Water Witch show a long, lean hull with a length- 5 '
to-beam ratio of 4.44:1, a sharp entrance, a relatively full
midship section with slack bilges, and a long, gracefully i
tapering run. Above all Water Witch was a shallow vessel, o I L |
with a rabbet-to-sheer height of only 6.5 ft (1.98 m)
amidships; she could not have drawn much more than 4 ft h
(1.21 m) of water even when heavily laden with iron ore.
The weight of her final cargo probably did not leave much

INA Quarterly 20.3

freeboard, and it is no surprise that she swamped and went
to the bottom as quickly as she did.
The conversion of Water Witch from a steamer to a
schooner was thorough, and we found little evidence of her
earlier career on the outside of her hull. Her original U.S.
Government enrollment documents indicate that the vessel
had no masts or bowsprit when she was built; these were
all added after 1835. Fred Hocker noted that the opening
cut for the bowsprit showed clear evidence of being a later
modification. The centerboard case in her hold was also a
later addition, designed to improve the schooner's lateral
resistance and stability. Remnants of the vessel's steam-
boat past may lie in the hold, beneath the iron ore, in the
form of engine mounts and sister keelsons designed to
distribute the weight of the engine and boiler over the
length of the hull.
What was evident to us was Water Witch's advanced age
at the time she sank. Small fillets of wood had been
inserted in many places in the main deck and at the sides
to replace sections of rotten planking, and the corners
where the rails meet the deck had strips of copper flashing
tacked down over seams to prevent water from leaking into
the hold. She was an old vessel, clearly in the twilight of
her career, when that gust of wind heeled her over in 1866
and the water gushed into the hold. By a strange coinci-
dence the shore nearest the wreck is Fort Cassin Point, the
very place where Water Witch was built and launched in

Acknowledgements. The 1993 field season on Lake Champlain
was sponsored by the Institute of Nautical Archaeology, Texas
A&M University, the Lake Champlain Maritime Museum, the
Vermont Division for Historic Preservation, the University of
Vermont, and the Lake Champlain Basin Program. A special
thank you is due to Mr. Ray H. Siegfried II for his help.
Thanks also to Texas A&M graduate students Elizabeth
Baldwin, John Bratten, Joseph Cozzi, Alan Flanigan, Peter
Hitchcock, Scott McLaughlin, Stephen Paris, and David
Robinson. The staff of the Lake Champlain Maritime
Museum contributed greatly to the smooth running of the
summer's fieldwork activity.

Suggested Reading
Crisman, Kevin J.
1986 Of Sailing Ships and Sidewheelers: The History and
Nautical Archaeology ofLake Champlain. Vermont
Division for Historic Preservation, Montpelier, VT.
Hill, Ralph Nading
1976 Lake Champlain: Key to Liberty. The Countryman
Press, Taftsville, VT.

I.-. I


Photo; D. Robinson

:: 7 MtIndependence rtrtl.t P
Photo: D. Robiamo


Phot: K. Crwmm
Cannon balls and grapeshot (top), bar shot (mid-
dle), and a shot box (bottom).

Palmer, Peter S.
1983 History of Lake Champlain. Harbor Hill Books,
Harrison, NY.
Ross, Ogden J.
1932 The Steamboats of Lake Champlain. The Delaware
and Hudson Railroad, Albany, NY.

INA Quarterly 20,3



News & Notes

Fulbright Scholar Program
Council for International Exchange of Scholars
Washington, D.C.
1995 1996 Competition

Fulbright opportunities are available for university lecturing or advanced research in nearly 140 countries. Funding
is provided by the United States Information Agency, on behalf of the U.S. government, and cooperating governments
and host institutions abroad.
Awards range from two months to a full academic year, and many assignments are flexible to the needs of the grantee.
Openings exist in almost every area of the humanities, social sciences, natural and applied sciences, the arts, and
professional fields such as business, journalism, and law. Applications are encouraged from professionals outside
academe, as well as from faculty at all types of institutions.
Basic eligibility requirements for a Fulbright Scholar award are U.S. citizenship and the Ph.D. or comparable
professional qualifications. For lecturing awards, university or college teaching experience is expected. Language skills
are needed for some countries, but most lecturing assignments are in English.
A SINGLE DEADLINE OF AUGUST 1, 1994, exists for research or lecturing grants to all world areas. Other
deadlines are in place for special programs. For further information and application materials, contact:
Council for International Exchange of Scholars
3007 Tilden Street, N.W., Suite 5M
Washington, D.C. 20002-3009
Bitnet (application requests only): CIES1@GWUVM.GWU.EDU. Telephone: 202-686-7877


Interpretative Research Program

The Division of Research Programs of the National Endowment for the Humanities welcomes applications for projects
in Old World and New World archaeology. The Endowment is particularly interested in projects that focus on
preparing the results of excavations for scholarly and popular publications. Support is also available for work on both
foreign and American sites, survey, excavation, materials analysis, laboratory research, artifact preservation, and field
reports. Funds for excavation are limited to $20,000 in outright funds per year; additional support is available through
federal matching funds. Awards usually range from $10,000 to about $150,000 for up to three years' duration,
depending upon the size of the project. The deadline is October 15, 1994, for projects beginning no earlier than
March of the next year. For application materials and further information, write or call:
Archaeology Projects\Interpretive Research
Division of Research Programs, Room 318
1100 Pennsylvania Avenue, NW
Washington, DC 20506
Telephone: 202-606-8210

INA Quarterly 20.1



George F. Base, Archaeological Director
Gregory M. Cook, Treasurer

John H. Baird
George F. Bass
Edward O. Boshell, Jr.
Gregory M. Cook
Harlan Crow
Claude Duthuit
Daniel Fallon
Danielle J. Peeey
Donald G. Geddes 1l
William Graves
Bcngt 0. Jansson

Robert K. Vincent, Jr., President
Rebecca H. Holloway, Secretary


Harry C. Kahn II
Michael L. Katzev
Jack W. Kelley, Chairman
Sally R. Lancaster
Norma S. Langworthy
Samuel J. LeFrak
Robert E. Lorton
Frederick R. Mayer
William A. McKenzie
William H. Mobley

Donald A. Frey, Vice President
Cemal M. Pulak, Vice President

Alex G. Nason
Ray H. Siegfried, U
Ayhan Sicimoglu
William T. Sturgis
Robert L. Walker
Lew O. Ward, Vice Chairman
Peter M. Way
Garry A. Weber
Martin A. Wilcox
Richard A. Williford
George O. Yamini

George P. Bass
George T. & Gladys H. Abell Professor of Nautical Archaeology/Yamini Family Professor of Liberal Arts
Kevin J. Crisman, Assistant Professor
Donny L. Hamilton, Associate Professor
Frederick M. Hocker, Sara W. & George O. Yamini Faculty Fellow
1. Richard Steffy, Sara W. & George O. Yamini Professor of Nautical Archaeology, Emeritus
Frederick H. van Doorninck, Jr., Frederick R. Mayer Professor in Nautical Archaeology
Shelley Wachamann, Meadows Assistant Professor of Biblical Archaeology

Mr. & Mrs. Ray H. Siegfried I
Graduate Fellow:
Cemal M. Pulak

Mr. & Mrs. J. Brown Cook
Graduate Fellows:
Gregory D. Cook
Joseph R. Cozzi
Jerome Lynn Hall
Tarns P. Pevny
Elizabeth Robinson Baldwin


Selma Karan
Sheila D. Matthews, M.A.
Robin C.M. Piercy
Cemal M. Pulak, M.S., M.A.
Sema Pulak, M.A.
Murat A. Tilev
Tufan U. Turanh
Patricia A. Turner
Jane Pannell-Yiddnm

Jeremy Green
Cheryl W. Haldane, Ph.D.
Douglas Haldane, M.A.
Margaret E. Leshikar, Ph.D.
Kathleen McLaughlin-Neyland, M.A.
John C. Neville
Robert S. Neyland, M.A.
James M. Parrent, Ph. D.
Ralph K. Pedersen, M.A.
Donald Rosencrantz

Cynthia J. Eiseman, Ph.D.
John A. Gifford, Ph.D.
Faith D. Hentachel, Ph.D.
Carolyn G. Koehler, Ph.D.
David I. Owen, Ph.D.
David C. Switzer, Ph.D.
Gordon P. Watts, Jr., M.A.

Michael A. Fitzgerald

Australian Institute of Maritime Archaeology
Boston University
Brown University
Bryn Mawr College
University of California, Berkeley
University of Cincinnati
Cornell University
Coming Museum of Glass
Department de Arqueol6gia Subacuatica de
la I.N.A.H., Mexico
University of Maryland, Baltimore County
New York University, Institute of Fine Arts
University of North Carolina, Chapel Hill
Partners for Livable Places
University Museum, University of
Shell of Turkey, Ltd.
Texas A&M Research Foundation
Texas A&M University
University of Texas, Austin

COUNSEL James A. Goold


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs