The Quiet Year

In a normal year, the Smiths Island Archaeology field school would be halfway finished at this moment. The students would be finishing up their collective work (standardized training, really) at Oven Site and dispersing into smaller teams at the other sites - Cave, Smallpox, perhaps the dock/wharf complex at Pitcher's Point. This year, however, we have suspended operations in order to catch up on database and paperwork and also for me to further develop some of my other research projects - particularly Virtual St. George's, which will feature 3D historicized reconstructions of Bermuda's first capital at various moments in the town's four-hundred-year history, developed using a videogame engine. Indeed, during a flying research visit in February, I shot more than 19,000 photographs to use in this digital reconstruction, and I supervised several students creating ArcGIS layer maps for St. George's in 1750, 1775 and 1815. 

Seaward facing side of Elmina, Portuguese from 1482 to 1637 and Dutch thereafter

Great Britain's Cape Coast Castle

New research opportunities have also taken me beyond Bermuda in both time and space.  In January, my University of Rochester colleague Renato Perucchio and I began a long-term collaboration with the University of Ghana to develop 3D models of Ghanaian Transatlantic Slave Trade castles and forts, starting with Elmina, which was built by the Portuguese in 1482. We also visited Fort Amsterdam (aka Koromantin), Fort Patience, James Fort, and Fort Usser and worked with U. Ghana Archaeology Ph.D. students to record them. The sites are enormous and challenging, both in terms of their physical size and complexity and as sites of history, shaped by clashing cultures and interests through which some Bermudians' ancestors doubtlessly passed.

Entry to Fort Patience

All of these coastal sites also reflected a further core element of Bermudian history: shipbuilding, fishing, and boat handling, which underscores the shared African and European roots of its maritime traditions. 

3D model of Elmina, derived from UAV drone footage

Fort Amsterdam model, mid-processing

More recently, I completed an intensive week-long digital archaeology field project in Oplontis and Herculaneum, using a variety of digital archaeology approaches to recording and recovering eroded examples of Roman graffiti and frescoes on First Century AD sites destroyed by the eruption of Mount Vesuvius in AD 79. My U of R colleague Nick Gresens and I led a small team of three undergraduates (including Smiths Island Field School veteran Cameron Barreto) to work at Oplontis, the 95-room Villa of Poppea (Nero's wife).

photogrammetry model of Roman graffiti, appx 5 mm

In addition to conducting photogrammetry surveys and laser scanning villa walls with a FARO focus and hand-held Artec Spider scanner, we employed a FLIR infrared thermal imaging camera and a one-of-a-kind UV cross-polarization camera recently invented at Rochester Institute of Technology (our neighbor) and thus far only used to image planets and nebulae. We are just back and starting to process the 460 GB of data we compiled.  

UV imaging of an amphora's painted labl

Active IR imaging: applying heat (via a halogen lamp) and studying the differential absorption within the wall fabric

Cameron watching the FARO scanner do its work

Oplontis Villa A stratigraphy: the site was buried under 24 feet of volcanic ash, rock, and pyroclastic
flow (lava). Kinda makes Oven Site look easy, since we can excavate it with trowels instead of jackhammers.

Digital Archaeology and Database Initiatives in 2015

3D digital model of Smallpox Bay ruin and excavations,
 based on 373 photos and with a dense cloud of 21.6 million points.

Stemming from Principal Investigator Michael Jarvis’s recent appointment as Director of Digital Media Studies at the University of Rochester and his increased access to new technologies, software, and devices, this season’s work included training students in a variety of new cutting-edge digital archaeology initiatives. Based on the 2014 season’s success in developing an interactive 3D model of Oven Site using Agisoft’s Photoscan software, the P.I. and Digital Archaeology specialist trainees Miriam Beard and Alice Wynd conducted photogrammetry surveys of all five sites, with Oven Site and Smallpox Bay being surveyed at the beginning, middle, and end of the season.

  
Although highest resolution processing awaits a return to the University of Rochester and access to the BlueGene supercomputer, we were able to generate excellent medium-resolution point cloud and textured models during the field school. Subsections of sites (the fallen wall scatter of Smallpox Bay; the interior of Oven Site’s oven, the Oven Site cistern) were also modelled separately.

Sparse Point Cloud preliminary model of Oven Site, including the Cistern in the upper left corner.
 Model derived from 722 photographs. View faces east.

 The 3D models derived from the photogrammetry series for all sites will be archived at the University of Rochester, provided to the Bermuda National Trust and Bermuda Government (if requested) and made available to interested scholars and the general public through Michael Jarvis’s professional webpages.

During the final days of the 2015 field season, we also attempted a fine-resolution 3D surface capture of Oven Site using a Kinect for Windows sensor, laptop running Skanect software, and a boat battery and AC power inverter to make it work in the field. Although a trial scan of the interior of the Oven Site’s ovens was successful, attempts to scan the rest of the site (open to natural daylight) failed, due to the reflective nature of white Bermudian limestone. 

The 3D Kinect laser scan trial proved a failure, but there are several alternatives that may yield better results. The Kinect may have greater success with another program, such as Artec Studio 9.2 or FARO SCENE or SCENE LT. Also, a higher resolution 3D scanner such as the FARO Focus X130 or X330 promises to be able to capture white surfaces in direct sunlight.

Finally, after reading a review in Make magazine, the P.I. obtained a Seek Thermal infrared camera compatible with IPads and smartphones capable of detecting slight temperature changes within rooms and environments. If suitable, the Seek Thermal promised to potentially identify buried features visible through heat signature differentials a little after dawn and dusk, since this would constitute the same dynamics upon which infrared photography remote sensing site detection is based. A partial survey was conducted at 6:00 am (just before sunrise) to 6:30 am on Saturday, June 27 in the wooded Cottonhole Bight valley area, with no discernible results (other than detecting the movements of wild brown rats) for site identification. Additional trials may yield better results, however, at cooler times of the year when night-to-day temperature swings are more pronounced.

Finally, Lab Supervisor and Data Manager Leigh Koszarsky oversaw the entry of nearly all artifacts recovered to date into ArtiBase, a Microsoft Access database of her design, based on established SGARP context sheet protocol and artifact inventory standards established by Dr. Brent Fortenberry. Veteran students Alice Wynd and Miriam Beard did much of the data input and also trained first-year students in proper entry techniques. Although the artifacts in 200 contexts had been entered in an Excel datasheet in 2014, a quality control audit found numerous inaccuracies, prompting the wholesale re-entry of these contexts afresh. The data architecture Koszarsky designed maintains separate forms and tables for each Smiths Island site and cross-references context site information (layer/feature fill/cut, unit size, Munsell color, soil type, etc.) with artifact inventory information. As a result in the future, site supervisors can have immediate access to profiles of artifacts recovered in all previously excavated areas as they proceed, allowing better real-time interpretation of new patterns of material in new contexts as they are recovered. With the exception of totals for contexts excavated between June 26 and July 4, ArtiBase is complete for all contexts excavated in every site since 2010.

Site	Excavated	Artifact Count
Oven Site	2010-15	25,137 (incl. 12,383 bone, 3992 metal & 3434 charcoal)
Smallpox Bay	2013-15	3,772
Cottonhole Bight	2012, 2014	1,732
Cave Site	2014-15	843
Limekiln	2015	134

Artifact Totals, by Site

Cave Site and Limekiln Site Summaries

Cave Site

Discovered in 2010 and 2013, Cave Site is completely undocumented in the historic record. It is located approximately a tenth of a mile south of the 1872 stone farmer’s cottage ruin at the center of Amenity Park but does not seem to be associated with it. Excavations commenced during the 2014 season with a three-meter-long exploratory trench bisecting the southernmost of the cave’s two openings to determine the depth of stratigraphic deposits within the cave. This work revealed a posthole situated midway across the front of the southern opening as well as a 30 cm step down to a flattened floor just within the cave’s interior. Few subsurface artifacts were recovered, but two sherds of Astburyware suggested occupation in mid-18^th-century. We also observed that the cave roof had been laboriously smoothed as an improvement and exhibited numerous tool marks.

This season’s investigations focused on completing the exploratory trench through the back wall of the cave in order to assess the extent of usable space and discern additional cultural modification. Numerous metal detector hits along the back edge of the cave suggested that deposits naturally sloped down to the rear area and concentrated there, so this strategy promised to reveal additional information about the site’s occupants and their activities. Historical evidence and cave sites found on Barbados and other British islands suggest a connection between caves and enslaved Africans; this year’s excavations particularly sought evidence of this and of the sorts of activities that occurred in this obscure, hidden location. Finally, we sought to assess whether Cave Site is an isolated example or whether there are additional rock outcrops and occupied cave sites nearby in this area of Smiths Island, including a promising but nearly completely buried location approximately eight feet from Cave Site’s northern opening.

Two units were excavated sequentially within the cave (N5 E2-3), completing a lateral east-west profile of the cave interior’s extent. The stratigraphic sequence followed that of N5 E4 excavated in 2014 but the units were far richer in artifacts, particularly large mammal and fish bones. Evidence of rodent burrowing and the mixing of 18^th and 19^th-century artifacts within the cave deposits reveal disturbances mitigating against a temporally fine-tuned reconstruction of occupation activities. Datable artifacts ranged from additional examples of Astburyware and plain white saltglazed stoneware to Creamware annularware, suggesting intermittent occupation spanning circa 1725 to circa 1825. Charcoal, a bone-handled knife, peach and other pits, and large quantities of both cracked and sawn mammal bone reveal that occupants regularly consumed food here - perhaps even feasted - and that the site served as a socializing/gathering  place. That said, the near complete absence of hand-blown glass bottle fragments and tobacco pipes suggests that this socializing did not apparently extend to drinking and smoking.

This season's excavations revealed that both the cave floor and ceiling had been deliberately smoothed by humans. Both follow natural limestone bedrock cleavage but exhibit numerous tool marks. The floor has been flattened but has a slight step up in unit N5 E4. The rubble layer lying atop the flattened floor seems to relate to the modification of the roof after the floor had been altered, suggesting that these events did not occur simultaneously. Unfortunately, no material was recovered atop the flat floor sealed by the stone detritus - artifacts that would help us date the earliest usage and first modification of Cave Site; perhaps the first occupants kept this floor clean and deposited rubbish outside the cave, in which case it may be possible to recover this through future excavations.

Cave Site, western end of interior trench (N5 E2), facing west. Note the deliberately
smoothed floor with tool marks and the south profile stratigraphic sequence,
showing a stone rubble layer as the bottom-most deposit.

Southern profile of Cave Site interior trench.

This season's work inside the cave has revealed a considerable usable space made through human improvement, with about four to five feet of head clearance when the floor was first  levelled. It also provided specific evidence of socializing and significant food consumption using a few strikingly high quality items (like a bone-handled knife and an Astburyware vessel) but not smoking or drinking. A few small rounded non-Bermudian stone pebbles and flint stone flake recovered seem culturally significant, since they were deliberately brought into the cave, but otherwise we recovered no culturally/ethnically diagnostic artifacts to associate the site with African or African-Bermudian usage.  Future archaeological investigation will target the area to the south of the interior cave trench and the open area in front of (to the east of) the cave's southern opening to both determine if the 2014 posthole relates to a structure of some sort and to gather more evidence of the cave's occupants and their activities. 

Testing near Cave Site included a pedestrian survey to the south and excavating a meter-square unit (N10 E5) near another rock outcropping. The survey found a significant, high exposed rock ridge approximately 120 to 150 feet south of Cave Site, with several partly filled deep declivities and 19th-century surface finds - an area that warrants further investigation in future seasons. The N10 E5 unit yielded charcoal, bone, and 20th-century artifacts but hit natural bedrock after 30cm and did not reveal the nearby rock cavity to be large or deep enough to constitute a new cave entrance.

Smallpox Bay Overview

Smallpox Bay Ruin, pre-excavation

The easternmost of our SIAP sites adjoins two small bays on Smiths Sound. The site is a standing stone roofless building ruin whose architectural features indicate an 18th- or early 19th-century construction date. Historical research suggests that the site may have been used as a quarantine station for inbound sick passengers and crew, since Smiths Island was one of two official mandated quarantine anchorages mentioned in Bermuda acts from 1731 through the early nineteenth century. Previous archaeology found evidence of mid-18th-century occupation extending into the mid-19th century, the latter with a distinct British military character. The 2014 season identified "G.R." (George Rex) and a broad arrow carved into the inside north wall, indications of an imperial presence. Intriguingly, artifacts recovered also included toys such as marbles and a cast copper alloy cannon barrel, suggesting the presence of children on an otherwise military site. Excavations within the ruin also revealed several postholes that predate the stone structure, as confirmed by the northern stone wall overlying and sealing a large posthole. A midden or heavily concentrated artifact layer discovered approximate fifty feet to the southeast of the ruin found at the very end of the 2014 season yielded several early 17th-century ceramics as well as numerous mid-18th to mid-19th-century artifacts consistent with the established ruin occupation dates.

This season's investigations had three main objectives: 1) deepen our understanding of the site's usage and occupants in the c. 1750-1860 main activity period, 2) attempt to discern a pattern among postholes cut into the bedrock that suggest a multi-period site with one or more timber-frame buildings predating the standing ruin, and 3) attempt to date any identified earth-fast structures in the interest to testing the hypothesis that Governor Richard Moore may have begun to build Bermuda's first town at this location in late July and early August 1612.

The collapsed southern wall of Smallpox Bay ruin. 

Note "Salisbury" on the lowest course of the wall.

Smallpox Bay Ruin suffered significant damage from Hurricanes Faye and Gonzalo in October 2014, which resulted in the south wall of the ruin collapsing. Curiously the interior side of the collapsed wall was revealed to have the word "Salisbury" carved into it through exposure to the elements, graffiti suggesting either the surname of an occupant or a personal or regimental affiliation with this southern English town. With the east and west walls currently not tied together and given the vulnerability of the site to southerly winds (the direction typical for a hurricane strike), the future integrity of the standing stone ruin is very much compromised. As such, we prioritized digital recording of the structure by tasking Miriam Beard and Alice Wynd, two veteran field school students, with creating 3D models before, during, and after excavations commenced (see below).

Site Supervisor Leigh Koszarsky directed excavation of a total of 26 contexts, completely exposing the ruin's interior in a clearing excavation designed to reveal posthole patterns. Strata was shallow and jumbled, and the presence of floor joist footings carved into the walls reveals that the house had a wooden, rather than an earthen floor.

Artifacts recovered from these units further strengthened the interpretation of a military-domestic blending: additional regimental and military buttons (XXth Regiment, Royal Artillery, Royal Engineers) were recovered alongside numerous civilian metal and bone buttons, copper alloy straight pins, hooks, and eyes, a thimble, a bone-handled brush, and a door key. Numerous mammal and fish bones and ceramics provide clear evidence of food consumption, despite the absence of a chimney and fireplace within the ruin. Leigh Koszarsky's additional historical research on the yellow fever epidemics that struck the 42nd and 56th Regiments lead her to interpret the Smallpox Bay site as a place where healthy soldiers - and perhaps their families - were sent to separate them from Barracks Hill, the epicenter of the contagion. This reading nicely accounts for the range of artifacts recovered and explains the presence of women and children at this intermittently used military site.

The clearing excavation revealed pronounced overlapping linear patterns among closely spaced postholes of uniform dimensions, as well as the building trench cuts for the construction of the standing ruin and a natural fissure.

Posthole patterns within the Smallpox Bay ruin, by size and depth. View facing west. A seventh posthole in the orange series was found in 2013 in a unit on the other side of the north wall.

Because the orientation of the standing ruin shares only a north-south orientation with the posthole patterns, it remains for future excavations to reveal the full extent of the earth-fast structures that formerly occupied the site. Posthole fill failed to yield any datable or diagnostic artifacts - which would be consistent with construction on a previously uninhabited location and a brief occupation followed by the abandonment of the structures. The midden site's lowest strata yielded both North Devon Plain earthenware and Surrey Borderware (aka Tudor Greenware), dating to the late 16th and early 17th centuries.

In sum, the 2015 season found clear evidence in the form of early artifacts and features that supports the hypothesis that Governor Richard Moore's briefly established 1612 town may have been situated at or near the Smallpox Bay site. The location makes sense, given the documented presence of the Plough in Smiths Sound and the ease with which a smaller shallow draft vessel like it could have moored close to shore at either of the bays immediately to the south. (A fifty-passenger ship like the Plough was probably in the 50-70 tons range and likely drew only eight to ten feet; accounts of the English landing at Jamestown in 1607 noted that the much larger Susan Constant, Godspeed,  and Discovery tied up to trees on James Island's shore.) It is unlikely that large quantities of early 17th-century artifacts can be recovered, due to the brief occupation period of the town, but a future clearing excavation strategy to the north, east, and west of the Smallpox Bay ruin has the strong potential of revealing the posthole footprints of several timber-frame structures, providing additional examples of the construction techniques of Bermuda's earliest domestic architecture.

Smallpox Bay Site Supervisor Leigh Koszarsky

Oven Site Summary

The Oven Site excavations provide insights into the first century of Bermuda’s settlement. The site appears on both the 1616/17 and 1662/63 Norwood Surveys.[1] Previous seasons revealed that the house dimensions were approximately ten feet by twenty-four feet and that the house evolved in two phases. This year’s research focused on exposing the majority of the Period I (thought to date c. 1615 to 1640?) phase of Oven Site in order to determine its dimensions and construction techniques, refine its dating, and shed light on how and when the Period II expansion occurred. Additionally, having defined the front of the house, we hypothesize the presence of a sheet refuse scatter in the yard to the north of the house that can yield further information about a succession of Oven Site occupants.

With these research foci in mind, we excavated a five-meter by one-meter trench extending north and perpendicular to the 2013-2014 central east-west trench (N4-9, E5), and also excavated five meter square units (N5, E 2-5; N6 E4) covering the predicted front wall of the Period I house.

The Northern Trench and Test Pits

Outside the boundaries of the house footprint, the new trench was very shallow (10-16 cm) and yielded very few artifacts, a mix of modern, 18^th-century quarry-associated, and 17^th-century finds. This surprising result led us to suspect that the sharp slope northward of the topography in this area may have produced a natural northerly/downslope migration of deposits over time, but the presence of the relocated 1970s yellow metal tank and backfill pile made extension of the trench difficult. To assess the possible distribution of the typical/expected domestic scatter, we excavated seven 50cm-square test pits in an array to the northeast, north, and northwest of the North Trench, ranging from ten to fifty-five feet away. All test pits yielded artifacts and varied in depth from 35 to 55 cm.

Miraculously, two of the test pits came down squarely on features. Test Pit Four completely exposed a 30 cm-diameter posthole extending 35cm into natural bedrock. Associated artifacts included bone, charcoal, and a tobacco pipe rim, but no datable finds.

Test Pit Five bisected a distinctive feature exhibiting a linear vertical cut into bedrock, which was treated with mortar and plaster. Close examination of a fragment of this plaster/mortar face led us to believe it related to a water cistern, since it very closely resembled an exposed tarris tank discovered at Smiths Island’s western end in 2014. Remarkably, the test pit exposed the sloping, smooth northwest corner of this hypothesized tank. The feature fill below the grade of natural bedrock was left unexcavated, pending the extension of the Oven Site’s formal grid to these units for more systematic excavation.

 

 The Cistern feature fell into units N10-11, E7-8. Excavation revealed the entire western face of the tank  and a portion of the north wall, which has been disturbed by two superimposed postholes dug at a later date that damaged the plaster wall. Three additional post holes were identified just to the west of the tank’s western wall, likely associated with a wooden structure covering the tank .

 Numerous fill layers were artifact rich. Upper layers (Master Context 155) contained a mix of 18^th- and 17^th-century ceramics and tobacco pipes, including white salt-glazed stoneware and Creamware as well as coarse earthenwares typical of Oven Site’s interior layers, while the cistern’s lower fill layers (Master Contexts 166, 170, 171) had consistently 17^th-century ceramics and a large concentration of large mammal, fish, and bird bones. Excavation at the Cistern Site continued through July 4 and removed nearly 60 cm of fill, but was suspended due to time constraints prior to completion; a final probe reveals that the bottom of the tank still remains at least 30 cm below the stopping point.

Our preliminary interpretation is that the cistern was probably abandoned prior to the general abandonment of Oven Site in circa 1712, given the dating of artifacts in much of the excavated fill, but that the hole remained open to receive additional deposits through the early- to mid-18^th century, when quarrying  was being conducted to the south of the site. Dateable material from the deepest strata of the Cistern fill will help establish when the cistern fell into disuse, and to which set of Oven House occupants we can associate the faunal and cultural materials within the fill.

Within Oven Site itself, excavations along the northern portion of the Period I house followed the same stratigraphic progression found in 2014: a thin quarry detritus layer (Master Context 018) underlying the surface layer, followed by a 19^th-century agricultural layer (MCxt 003) and a thick, dense earlier quarry detritus layer (MCxt 005) that seals the Oven Site Period II earthen floor layers (MCxt 006 and 009).

 Artifacts from the earth floor were consistent with past seasons and included several dozen additional worked stone chert flakes, which corresponds with the presence of Native American slaves during Boaz Sharpe’s occupation of the site (c. 1683-1705). A stone scatter abutting the northern front wall and in line with the revised rectangular orientation of the Period II house appears to have supported the footing for a wooden sill that anchored the front wall of the expanded Oven Site structure, since there is no evidence of earth-fast construction (i.e. sunken postholes) for this period.

Oven Site North wall, looking west. The 17th-century earthen floor (MCxt 006) is to the east of a linear stone scatter, which directly overlies the Period I northern wall cut.

As was found in 2014, the floor layers sealed a densely packed limestone rubble layer (MCxt 089) that we now interpret to be associated with the quarrying out of the hillside to the west to create the Period II house expansion and bringing the deeper Period I floor level up to the Period II level. This layer seals a greasy black charcoal-rich layer (MCxt 091) with brick and mortar fragments formerly associated with the salvaging of bricks from the original Period I hearth-less brick oven prior to the creation of the current Period II hearth and ovens in the site’s northwest corner. The density of charcoal and the presence of char in underlying postholes, however, now raises the possibility that the Period I house experienced a partial fire as responsible for the architectural transition. If this occurred, it does appear that in the rebuilding transition, the occupants removed evidence of earlier domestic deposits, since these layers are very sparse in terms of artifacts. In terms of datable ceramics, all recovered examples below the MCxt 009 floor layer are consistent with types found on the Sea Venture: plain North Devonware, Metropolitan/London coarse earthenware, and tin-glazed earthenware. The complete absence of bottle glass is consistent with pre-1640s sites. The absence of nails, scarcity of iron architectural artifacts of any kind, and absence of window glass in these early layers all point to an entirely timber-framed building held together with treenails and with open windows – material culture elements entirely consistent with daily life as described by the tenants of the Earl of Warwick and Sir Nathaniel Rich in the 1610s and 1620s in Vernon Ives’s The Rich Papers (Toronto, 1985).

Oven Site Period I footprint with MCxt 091 layer
and stone concentration, looking north.

The iron plate found at the end of the 2014 season in unit N5 E5 was completely exposed with the excavation of N4 E5. Despite the use of a microcrystalline wax consolidator as a field conservation effort, the plate was found to be entirely a rust stain lacking any intact iron and could not be removed intact. Unfortunately, the exposed shape was too ambiguous to attribute a use or function.

Perhaps the most significant discovery with regard to early Bermudian architecture emerged at the end of excavation as MCxt 089 and 091 were removed. Both layers abutted a stack of flat stones along the north wall cut, which were initially interpreted as a partial flagstone floor. As they were removed and put into a vertical position, however, it became apparent that they were actually fragments of the Period I house wall. Besides their consistent thickness, many stones exhibited evidence of smoothing or plastering and the impressions of wattle or wooden lathes. The porous character of the back portion of several of the stones reveals that they are reconstituted mortar fill rather than natural cut bedrock. The importance of this discovery is significant in establishing that at least at this early site, Bermudian builders were using limestone from the start in a local adaptation of traditional English wattle-and-daub timber frame construction, with a Bermuda stone and mortar slurry substituting for the clay daub that other English settlers used in England and Virginia. The fact that these stone examples are sealed by the earthen floor layers (MCxt 006 and 009) clearly dates them to Oven Site Phase I and the 1610s (that is, pre-1617 Norwood)

The final revelation of the 2015 season at Oven Site relates to the numerous postholes cut into the Period I floor and adjoining the northern wall cut. A second large posthole matching the one discovered in 2013 next to the early oven was found midway along the west wall cut. Period I Oven Site was supported by two huge posts carrying a roof ridge, which apparently anchored a series of cruck beams set onto bedrock but not anchored in postholes.  A single deep posthole was identified outside the wall cut at the midpoint of the north wall, which possibly anchored a wall sill; the lack of a similar corresponding posthole at either corner makes this particular feature enigmatic. A similar semi- subterranean early house was excavated by James Deetz at Flowerdew Hundred, Virginia, dating to the late 1610s or early 1620s and also apparently had cruck timber framing, so this building style was clearly within early English colonists' construction repertoires.

With dimensions of eight feet by fourteen feet, the Period I Oven Site was small and cramped, suggesting a very early shelter driven by immediate housing needs; even by the late 1610s, Richard Norwood reported that Bermudian colonists had embraced larger, more complex labour and resource- intensive housing for their families, shifting from living in "tents or cabbins" to "substantial houses."[2] The Period II expansion more than doubled the living area, increasing it from approximately 112 square feet to 240 square feet.

The discovery of the cistern and a large posthole to the north of Oven Site this season raises the distinct possibility that the building may in fact be a detached kitchen or service building for the main residence located nearby. This scenario fits well with the reality that Boaz Sharp and his nine Native American slaves would have been hard pressed to comfortably occupy a 240-square foot house, especially when surrounded by ample land for expansion. The proximity of water to the north, with access to St. George's just across the harbour and to Smiths Fort - which the master of the house was duty-bound to man and manage, strongly suggests that a conjectured but currently unlocated main house could be situated near to but north of Oven Site. Future research should include a comprehensive Phase I testing program, as well as more concentrated investigation of the area immediately adjoining the cistern feature - which would have had to have received water from some nearby roof. Rather than completing Oven Site, this season's fieldwork has recast the site's interpretation and importance.

---

[1] The original 1616/17 Norwood survey does not survive, but the author has argued the case that the 1626 printed Speed map reflects the state of Bermuda when Norwood departed in 1617, especially with reference to Southampton’s post-1617 settlement  patterns, as documented in Vernon Ives, ed., The Rich Papers. (Toronto, 1985).

[2] Richard Norwood, Insularum De La Bermuda Detectio, c. 1622, f. 9; John Carter Brown Library.