This page is adapted from "The Search for Columbus's Last Ships: The 1991 Field Season."
 by J. Parrent, J. Neville, and R. Neyland, INA Newsletter 18.4 (1991): 16-19.

Edited for online publication January 2003.

The second field season of the Columbus Caravels Archaeological Project (CCAP) was conducted at St. Ann's Bay, Jamaica, between June 15 and September 1, 1991.

During these 11 weeks, archaeologists from the Institute of Nautical Archaeology and the Jamaica National Heritage Trust, as well as Texas A&M University field school students, surveyed the area of St. Ann's Bay in search of the remains of Columbus's last command--the caravels Capitana and Santiago.  The fatigue and unseaworthiness of these ships forced Columbus to conclude his last voyage by grounding his vessels on the shore of St. Ann's Bay in 1503.  Columbus and his men lived on the decks of the ships, beached side by side, for over a year before being rescued (see INA Newsletter 16/4 and 17/4).

CCAP team members take measurements  prior to remote sensing.   Photo: INA.

1991 survey

The 1991 field season consisted of two phases: a remote-sensing survey of the area to the west of Reader's Point followed by test excavations of several significant anomalies.  We conducted the remote-sensing survey with a "chirp" subbottom-profiler unit developed and operated by Dr. Steven Schock of Florida Atlantic University.  Named after the distinctive sound it makes during operation, the "chirp" unit uses advanced sonar technology to produce an image of subsurface sediments and any objects buried within them.  This season, 21 potential shipwreck sites were found with the "chirp" unit, giving us a total of 27 sites located in two field seasons.  Each site discovered by the "chirp" sonar was tested by probing with steel rods to verify the type and depth of detected objects.  The most promising sites were further tested by using a vibracore machine to take 3-inch diameter core samples.  Probing and coring allowed us to differentiate materials such as gravel, ballast stone, wood, and pockets of organic gas detected by the system.  (The reflective properties of submerged organic gas make such targets look like solid objects buried beneath the seabed.)

Our success in identifying potential wreck sites was a mixed blessing.  Our surveying methods allowed us to cover a large section of the bay in a short period, but we located far too many sites to excavate during a single field season.  Of the 21 new sites found, probing and coring indicated that 14 were associated with wood.  Four were confirmed as shipwrecks when cores taken from them proved to contain fragments of ship's wood such as frames, ceiling planking, and hull planking.  Small artifacts, including ceramic sherds and a small brass buckle, also appeared in the cores.

Fragments of a porcelain bowl probably date  to after 1700.   Photo: INA.

In the final analysis, we identified 11 sites that fit within the broad criteria established for locating the resting place of the caravels.  The sites lie in protected areas of the bay that would have been shallow enough in the 16th century for Columbus to beach his ships.  Some sites also appear to represent two contiguous wrecks, a configuration that agrees with historical accounts of Columbus beaching his ships next to each other and permanently fixing their position by shoring up the hulls.  These sites were given the highest priority for test excavations.

1991 excavations

Test excavations were conducted on four sites in 1991.  These sites were buried beneath as much as 10 feet of sediment.  Texas A&M geoarchaeologists Mike Waters and Rick Giardino have determined that most of the sediment was deposited as a result of agricultural activity beginning with the Taino Indian occupation around 1100 BP and dramatically increasing with the colonization and development of the St. Ann's Bay area after the English conquest of Jamaica in 1655.

The first site chosen for testing is known as Mangrove Site 3 (MS3).  Originally located in November 1990, MS3 was examined briefly during the final days of the 1990 season.  The site seemed especially promising due to an early radiocarbon date for a wood sample from the hull.  Work at MS3 in 1991 began with a thorough subbottom-profiler survey, which we refined by probing and coring.  After consideration of the data, a test trench was excavated through the approximate center of the wreck, uncovering a maststep and revealing a cross-sectional view of the hull remains.  Preliminary analysis of the hull reveals a shape consistent with that of an early vessel.  The maststep displays a complex and unique construction, being composed of six oak timbers fastened to the keelson with iron bolts.  So far, no parallels to the composite construction of the MS3 maststep have been found.

The only evidence of cargo in the MS3 wreck was a number of logs stowed amid the ship's ballast.  The logs appear to be of different species, one of which has been tentatively identified as lignum vitae (Guaiacum officinale).  All of the logs had been cut into approximately 4-foot lengths and retained marks of the axe used to fell them.  Part of the mandate of early explorers to the New World was the collection of wood samples to determine their viability for commercial exploitation.  Lignum vitae, for example, was found to have a wide variety of uses.  Its hardness and durability quickly made it a favored material for the construction of sheaves for blocks, and a drink prepared by boiling shavings of the wood was used as a remedy for syphilis.  The small logs from MS3 may have been intended for shipment to Europe as trade goods or perhaps for research purposes.

The radiocarbon date for wood from the hull and the presence of a single Taino Indian bead suggest a relatively early date for MS3; however, the wreck is more likely to date from the late 17th or early 18th century since several ceramic sherds and kaolin pipe fragments dating to this later period were recovered from the site.  The data collected at MS3 does not provide an immediate identification of the wreck or its age, but further analysis of the hull remains and the artifacts will produce a more accurate picture.

The second test excavation was carried out at Site 21/22.  This area was chosen because it appeared to contain two targets situated side by side when first located with remote-sensing equipment, and cores from the site contained wood from a ship's hull.  A trench was excavated between the two sites so that sections of both could be examined simultaneously.  Excavation revealed a single large wreck lying parallel to the shore on an east/west axis with the bow probably facing west.  The section of the hull uncovered was badly damaged but it appears to be aft of midships and near the stem.  The ship evidently was a deep drafted seagoing vessel having a heavy keel and deadwood assembly with V-shaped floor timbers resting on the deadwood.  In the area tested, the hull was largely swept clear of artifacts, probably due to its proximity to a relic stream mouth.  The few ceramic artifacts recovered include three porcelain sherds and a single piece of tin-glazed earthenware, possibly delft, which together suggest a late 17th- or 18th-century date for the ship.

The third site tested is designated Site 16.  This vessel, like the other two wrecks, was oriented in an east/west direction, with the bow pointed east.  A 13-foot section of its starboard bow was uncovered.  This well-preserved hull is characterized by a radial cant frame construction.  Artifacts such as salt-glazed stoneware sherds, a case bottle, and a kaolin tobacco pipe date the wreck to the late 18th century.

During the last two days of the field season, Site 14 was examined.  Subbottom-profiler, probe, and core data suggested this site contained ballast but did not show any evidence of hull remains.  Test excavation revealed a lens of thick gravel containing some ballast stones and a few artifacts, including a leather shoe sole, encrusted iron nails, a cargo hook, sherds from a creamware pitcher, and fragments of worked wood.  These artifacts do not represent a shipwreck site and are more likely the result of ships dumping their ballast in the bay.

Although the caravels were not identified this season, the sites that were tested represent significant discoveries.  The three shipwrecks partially excavated are well preserved, and MS3 exhibits unique construction features.  Although possibly salvaged at the time they were originally lost, these wrecks have not been disturbed by modern looters.  Their pristine archaeological condition and their location in shallow water make them good candidates for future research.

Conservation and analysis of the artifacts and data recovered from St. Ann's Bay was conducted under the supervision of field directors Jack Neville and Bob Neyland.  Nautical archaeology graduate student Amy Rubenstein conserved, recorded, and cataloged artifacts that were brought to College Station, and graduate student Greg Cook worked on reconstruction drawings of the hull remains discovered at Site 16.  Graduate student Bess Manning analyzed the data recovered from 136 cores taken this summer, while Dr. Mike Waters reconstructed the geoarchaeological history of the bay with special emphasis on the position of the 16th-century coastline. 

The discovery, exploration, commerce, and development of early Jamaica are all represented in the archaeological sites of St. Ann's Bay.  The high concentration of shipwrecks in the primary search area indicates that it was an attractive, protected anchorage in use for several centuries.  A solid foundation for further research in the bay was laid, and sites were located and evaluated with speed and accuracy.

Acknowledgements

The following groups and individuals are recognized for their contributions to the 1991 CCAP: The John Brown Cook Foundation; Cambridge Seven Associates; American Way Magazine; Texas A&M University; Orel Hershiser; INA Board Members Don Geddes , Bruce Heafitz, and Robert Lorton; Kaiser Bauxite; Air Jamaica; and the men and women of the Jamaica National Heritage Trust, Jamaica Defence Force Coast Guard, Jamaica Customs, JAMPRO, and the Jamaica Tourist Board.

James Parrent, Jack Neville, and Bob Neyland
2003, Columbus Caravels Archaeological Project: The 1991 Field Season
Edited by Ralph K. Pedersen
URL, http://ina.tamu.edu/CCAP/CCAP1991/CCAP1991.htm

Design by Ralph K. Pedersen