by J. COZ Cozzi, Ph.D., RPA
University of West Florida Archaeology Institute

The University of West Florida (UWF) Archaeology Institute has identified shipwreck remains in Pensacola Bay as Nuestra Señora del Rosario y Santiago Apostol, a 44-gun Spanish fragata that sank during a hurricane in 1705. This vessel was built in Campeche and its builder utilized mahogany (Swietenia spp.) extensively in the hull (Figure 1). This abundant and large-growing New World wood provided Campeche shipbuilders with a material not available to their counterparts in Europe. How did the availability of large mahogany timbers influence the design and construction of New World ships? The hull of the Rosario indicates that this wood enabled its builder to fashion timbers of substantial dimensions that were unthinkable in Europe at the time.

Introduction

Investigations of this wreck site revealed an exclusively iron-fastened hull that is well preserved from the stem to a point some seven meters aft of the main maststep (Figure 2). The remains lie on a 35-degree slope and consist of a significant portion of the port bow. The starboard side is preserved for a distance of only two meters from the centerline amidship while the port side continues out more than twice this distance. The stern is also no longer extant. Hull timbers are in remarkably good condition and once sand covering the site is removed tool marks and fastener patterns are easily discerned.

Hull Description

Keel

The Rosario’s keel was encountered in 1999 when the pump wells were examined. At that time only the top surface of the keel was available for study. In 2002 UWF excavated unit 118N406E, which is on the upslope side of the site just abaft the main mast step and is just forward of where the Rosario broke her back. The excavation went under the garboard and first strakes of the starboard side revealing the keel (Figure 3). The first strake was near a butt join in two planks and was easily removed. The garboard had to be cut away to reveal the rabbet in the upper starboard side of the keel. The excavation continued under the keel to the port side, revealing details of the ship’s centerline structure. The keel amidship is sided 27.6 cm and molded 32 cm. The rabbet is at the very top of the keel and is in the shape of a sideways “V” (Figure 4). Caulking material was observed in the rabbet. The top surface of the keel has a V-shaped concavity that created a space between it and the floor timber that sat atop it. This served as the limber hole and allowed water to flow freely to the ship’s bilge pump.

The keel had wooden sheathing preserved on both the port and starboard sides, with fasteners evident on the bottom of the keel indicating that this surface had also been sheathed. The planks of keel sheathing were 25 cm wide and 5 cm thick. Samples of the sheathing were taken and have been sent away for species identification. We are currently waiting on the results of wood identification.

Frames

The Rosario’s preserved framing along the starboard side consists of floor timbers and first futtocks. The paired frames have a room of 44.65 cm and a space of 19.81. The molded dimension of the frames at the keel is 35 cm. The molded dimension of timbers along the eroded starboard side is 25 cm. The floor timbers and first futtocks are joined together with rectangular notches in their opposing faces (Figure 5). This is consistent with the Nuestra Señora de Atocha, which was also built in the New World and sank off Marquesas Key in 1622 (Figure 6).

The space between the frames on Rosario was filled with a lime and shell concrete mixture that ballasted the ship (Figure 7). The Atocha also had this material between frames. This practice continues on American ships into the 20th century as seen on two other shipwrecks in Pensacola Bay dating to 1890 and to circa 1920.

Keelson

The Rosario’s keelson is one of its most interesting features. At the main maststep this timber is sided 80 cm and molded 42 cm (Figure 8). It is made from Mahogany and extends from where it is eroded and worm-eaten two framing stations abaft the mast step and runs forward to but against the fore mast step. The sided dimension of 80 cm is carried forward for a distance of 5.1 meters. Forward of this the sided dimension reduces to 30 cm. At the after end, the keelson begins to reduce in sided dimension along is port side similar to the forward end. This feature is no longer preserved on the after starboard side.

The main mast step mortise is 43.5 cm long, 22 cm wide and 9 cm deep. In 1998 and 1999 a carved “X” was noted in the bottom of the mortise, but exposure of this timber by the loss of sediments on the site over several years has destroyed this feature. The loss of sand coving this site is most likely attributable to the dredging of Pensacola Bay and the strong currents that flow over the site. This situation may also be the result of past beach renourishment projects that have altered the shoreline and the effects of each tidal change. The dive team encountered extreme strong currents on out-going tides. At times diving had to be abandoned especially in 1999. In 2002, however, the timing of the summer tides was extremely favorable for diving and underwater visibility on the site.

The top of the keelson has evidence of five stanchions. At the forward end of the expanded keelson were the remains of the foot of a stanchion. This stanchion had a tongue at its base that sat in a mortise in the top of the keelson. In the mortise a wedge had been placed, which may indicate the method with which the builder erected the stanchions. The stanchion was supported on its port and starboard sides by a chock that was fastened to the keelson with three iron spikes. The chocks were shaped on the upper surface so as not to present sharp corners to anyone walking around in this area. A second stanchion was at one time located farther aft as evinced by a mortise and the eroded remains of the port-side chock and fastener holes where the starboard chock had once been. Moving forward on the reduced keelson a third mortise is preserved along with a pair of buttresses that once supported a stanchion and prevented it from moving to port or starboard. Farther forward a forth mortise was found with no buttresses in place. At the very forward end of the keelson was a fifth mortise. The forward end of the keelson extended up and back to provide additional support for a stanchion no longer preserved at this location.

Immediately forward of the keelson is a unique fore maststep timber (Figure 9). It is triangular in plan view and of great size. The timber is 1.785 meters long to its port side. Along the starboard side, however, the timber has a 1.3-meters-long extension that is scarfed into the footwale by means of a hook and butt scarf . The portside footwale ends in a butt join against the after face of the fore maststep timber. The mortise in this step is 34 cm long, 19 cm wide and 13 cm deep. The mortise is located 1.1 meters from the forward end of the step timber. The shape of this unusually timber beneath the surface could not be determined.

Planking

The Rosario’s hull planking was exposed in a few locations by excavation along the starboard side, as well as at the port-side end of the two athwartship trenches (Figure 10). The planking exposed along the starboard side is probably the third or forth strake, and measures 7 cm thick. The garboard is 29 cm wide and 7 cm thick. The first strake averages 27 cm wide and 7 cm thick. The second strake is 25 cm wide and 7 cm thick.

The Rosario’s ceiling planking was completely exposed along the starboard side and in the athwartship trenches. The ceiling begins with a strake along side the keelson. Where the keelson expands and reduces the ceiling was cut to fit this shape. The ceiling was consistently between 5 and 6 cm in thickness, and varied greatly in width depending on location from narrow filler pieces to a maximum width of 40 cm.

Rider Timbers

The placement of rider timbers atop the ceiling helped strengthen the large hull. Four rider stations were noted on site (Figure 11). These timbers consist of a rider floor timber with a rider futtock to port and starboard. The rider floor timbers have a maximum sided dimension of 30 cm at the vessel’s centerline where they also have a maximum molded dimension of 35 cm (Figure 12). These dimensions reduce slightly at the wrongheads, which terminate at a stringer that marks the orlop deck level. Rider futtocks were noted at all rider stations with the best preserved examples at the forwardmost and aftermost portside stations, whose location coincided with our bow and stern trenches. The rider futtocks overlap the rider floor timbers without the elaborate joins noted on the vessel’s framing. Rider futtocks begin close to the centerline at the vessel’s bows and start a little over one meter from the keelson at the two amidship stations. The ends of the five rider futtocks examined on site were all beveled. All rider timbers were fastened to the hull and to each other.

Other Features

When excavated the Rosario’s hull timbers exhibit the bright red and orange of Mahogany and Spanish Cedar, like the buttress (Figure 13). Collapsed deck planking shows saw marks in its rough surface (Figure 14). Barefoot sailors needed protection from splinters, which was provided by woven matting placed over the deck (Figure 15). This same type of matting was found on the Emanuel Point Ship that dates to the Luna expedition of 1559.

During the 2002 season divers raised a well-preserved stair riser (Figure 16), part of a set of steps that led down into the orlop deck from above. The crew used it when they needed to come to retrieve spare ship’s equipment that was stowed on this deck. The rider’s outer face is adorned with the carving of a Christian cross (Figure 17), undoubtedly serving as a religious reminder to the crew.

Discussion

A royal cédula dated to 27 July 1700, is critical of the lines submitted for the two vessels completed for the Armada de Barlovento (Windward Fleet) in late 1703. In the document a Spanish Fiscal indicates that the hulls in question have too little deadrise, as well as too much room in the vessel’s floor. The European Spanish were pointing out that New World shipbuilders made their vessels with a greater flat in the bottom than did their European counterparts. To gain a clear understanding of this we need to compare the hull lines taken off the Rosario during the most-recent investigations with the lines for a 52-gun frigate from the Album of the Marqués de la Victoria, which describes the construction and outfitting of 18th-century ships. By scaling the Rosario’s midship curve on the 52-gun frigate’s lines we see that the floor of the Rosario extends out beyond the preferred turn of the bilge (shown in red on Figure 18). The New World builders were constructing merchant vessels for the shallow waters of the ports they frequented. Another cédula dated to 17 April 1701 states the need for one or more European shipbuilders who can fabricate warships, as local craftsmen were only versed in building merchant vessels.

The Nuestra Señora del Rosario y Santiago Apostol provides a glimpse into the minds of New World ship builders at the turn of the 18th century. They are clearly building European-style vessels, but with New World timber and with an aim toward use in the shallow harbors of the Gulf of Mexico and the Caribbean. New World hardwoods such as Mahogany grew to great sizes providing ship carpenters with large timbers from which to fashion the extremely heavy keelson at the main maststep and the large unique triangular-shaped fore maststep.

In the future another fruitful avenue of research on this site would be to expose and record more of the shape of this well-preserved hull in order to provide data from which to estimate the vessel’s tonnage. This could help clear up the historical record that has left us unclear as to which Nuestra Señora del Rosario y Santiago Apostol this ship is, either the 450-ton or 600-ton vessel.

Conclusion

With this wreck we see that the archaeological resource supports the historical record in that ships built in the New World were made more flat-bottomed than their European counterparts. This was done to deal with shallow waters encountered in the local environment, where these vessels would spend most if not all of their careers. These same New World shipbuilders had at their disposal plentiful and sizeable hardwoods from which to fashion immense maststep timbers that would have been unthinkable in Europe at this time. The Rosario, therefore, is the product of an American mind tempered by experience and unafraid to depart from European standards.

Acknowledgements

We would like to thank the Florida Division of Historical Resources for funding the current investigations, and the Florida Division of Community Affairs for previous funding. Thank you to our field school students and volunteers. We wish to acknowledge Dr. Roger Smith and Della Scott-Ireton from the Bureau of Archaeological Research. We wish to thank Wayne Childers for finding and translating many wonderful documents concerning the Rosario. We would also like to thank David Moore for sharing his insights into the hull construction of the Atocha.