Spanish ceramics and wall trenches

Today our largest excavation unit to date (2 x 2 m.) was brought down on top of the wall-trench that we intersected with the 50 x 50 cm. shovel test last week. Ground penetrating radar survey in this area indicated that the wall-trench might extend both east and west from the original unit, and this proved to be accurate, as can be seen in the photo to the right (the trench contains slightly lighter-colored fill along with yellow-gray clay subsoil). The trench appears to have been excavated precisely east-west, and is roughly half a meter in width. At this stage, we will need to take particular care in excavating the trench and surrounding sediments, so that we can learn as much as possible about the structure it belonged to (such as which side was inside and which was outside, unless it turns out to be an interior partition wall).

In this same unit we were also pleased to find not one but three varieties of 18th-century Spanish ceramics, including two decorated types of tin-glazed majolica made in Mexico (Abo Polychrome and Puebla Blue on White, shown to left) as well as a sherd of lead-glazed El Morro ware. All three of these types overlap during the first half of the 18th century, precisely during the period of Mission Escambe's initial occupation. Perhaps even more than many of our earlier finds, these items provide sound confirmation of both the date and the clear Spanish association of the Native American village we have discovered. Indeed, the high frequency of European items we are finding in direct association with the wall-trench structure we are exploring makes it likely that it was one of the primary mission buildings, such as the church or friary, or perhaps a residence for members of the cavalry garrison stationed there about 1760. In any case, it places resident Spaniards and Native Americans in the same village, and exactly in the predicted location for Mission San Joseph de Escambe along its namesake river, the Escambia.

Still more

Today one of our crews opened a 2 meter by 2 meter unit around the shovel test with the wall-trench structure corner, with the goal of exploring this architectural feature in greater detail, and obtaining a larger sample of the artifacts associated with it. Due to its larger size, the unit is proceeding slowly using the flat-shoveling technique bring it down in even, 10 cm. levels (photo to right). By the end of the day, only the first level had been completed, bringing the unit down through the upper root mat and humus, but we have already begun to find more evidence of activities at the Escambe mission.

In addition to a good number of Native American sherds, including several of the type Ocmulgee Fields Incised, a white glass seed bead was found, virtually identical with the two found in a shovel test not far away during mid-June. More significantly, however, we recovered a stamped lead bale seal (photo to left, uncleaned) with the letter "K" above a line, and the number "653" below the line, and other letters around the outside on the top half (largely truncated by the edge of the seal). Such seals are thought to have been used to seal bales or bundles of cloth or other trade goods, and they are not uncommon on 18th-century sites in Southeastern North America, particularly on those associated with Native American trading. This one is a particularly well-preserved example, and might provide enough information to allow it to be dated more precisely, or associated with a particular merchant or manufacturer.

Other crews worked in additional tests along the margins of the site today, but we plan to focus on the mission village itself in the final three weeks of field school. We have only just begun our more extensive excavations into the newly-identified mission deposits, so we anticipate more to report in future days.

More finds

We had another exciting day today, with several unexpected discoveries, and further confirmation of the wall-trench structure first identified yesterday. In a shovel test just 20 meters from that structure's corner, what first appeared to be a simple concentration of charred wood (which might have been anything from a burned tree stump to a charred post) ultimately turned out to be what is known as a "cob pit," filled with large numbers of charred corn cobs, several of which were nearly intact (see photo to right). These pits are very common on Spanish missions across northern Florida, and appear to have been used as smudge pits to generate smoke for mosquito control in and around houses in the village. The cobs were placed in small subsurface pits with constricted mouths, where they could smolder slowly and generate enough smoke to keep bugs away. Importantly, the cobs found at our site is of the 8-row variety, consistent with maize grown by North American Indians during this period (and much, much smaller than modern hybrid varieties). Further study should tell us more details about the corn grown at the Escambe mission.

In the same unit, and at about the same level, students found a remarkably well-made ground-stone discoidal made from greenstone, a type of stone probably originating in the Alabama Piedmont region, well over 100 miles away. This type of object, generally thought to be a Native American gaming piece, might date to the late prehistoric Pensacola culture, or it might instead be an Apalachee item from the mission period. Since people were living at this site during both periods, either option is possible, though further exploration at the site may tell us more about the context of the object, and what period it belongs to. The same unit produced a tiny sherd of plain Spanish majolica.

In the adjacent shovel test, students continued to explore the wall-trench structure discovered yesterday, which now appears to be a post-on-sill construction type that was commonly used on French colonial sites, though it has also been documented in Pensacola's Spanish presidios. An intriguing feature of the structure we have identified at the Escambe mission is the apparent presence of the remains of some sort of floor structure on the inside of the building corner. A number of wrought-iron nails were found lying in place in association with this floor, and careful excavation of this complex set of features is still proceeding. We definitely plan to conduct additional excavations in this area in order to learn more about the identity of this structure and what role it played in the mission community.

Other students continued to work on the larger excavation unit opened above the radar anomaly described in earlier posts, where they were able to learn how to shovel-shave, and how to draw larger plan-view maps using rulers and a plumb-bob (see photo). While some Native American ceramics and tiny flakes of chert have appeared in this unit, the presence of sheet metal and other modern artifacts may indicate that the anomaly could be related to the 20th-century dairy barn in this vicinity. In any case, only further excavation will tell the tale.

And sometimes, we get very lucky

In recent days we've been working on filling in some of the gaps in our shovel test survey in the wooded area of the survey area, and since we're digging our shovel tests on a 20-meter interval, on any given day there is a fairly high priority that we'll place a shovel test right in-between anything important. Today however, we got lucky.

In one of our shovel tests we detected some odd soil colorations with linear boundaries, so we decided to excavate them carefully, and have discovered what appears to be the corner of a wall trench structure. While it is somewhat difficult to see in the photo the area of lighter colored soil in the center of the pit is the wall trench, and this lighter soil extends into the southeastern corner of the shovel test. On the western edge is a semi-circular darker area which is likely the corner post, and along the southern edge of the unit is a somewhat more dense area of soil which is likely the interior of the structure and may be related to a dirt floor.

As we excavated this we turned up a very nice square hand-wrought nail (above), which strengthened our hypothesis that these stains in the soil were structural remnants, and furthermore that the structure was almost certainly constructed using European-style construction techniques (as opposed to Native American). Based on these initial results, it appears we may have gotten very lucky in placing this shovel test precisely in the corner of one of the buildings of the mission. Many times, archaeologists search for days and weeks to find wall lines and trace them out to find building corners, but we appear to have found a corner right at the start. We'll be exploring this structure further in coming days.

Additionally, we found some more pieces of Native American ceramic fragments which also support our belief that we have likely located the remains of Mission Escambe. The rim fragment below is a type known as Ocmulgee Fields which was being produced by the Apalachee Indians around the time the mission was occupied.

Bigger and better

With less than a month left in this summer's field school, we're making good progress in confirming and defining the mission-period occupation that we first began to discover in shovel tests late in June. Four shovel tests have been excavated so far in the transects that were cleared through the woods last week, and all but one produced clear evidence of Native American ceramics dating to the right time period. The most recent, excavated today, produced a sherd of a type of ceramics commonly associated with Creek Indians, commonly known as Chattahoochee Brushed (see photo), decorated with bundles of pine straw. This type only begins to appear in the Pensacola area during the eighteenth century in association with other mission-era ceramics, and may well have been made by Apalachee Indians who grew up in the Creek country before 1718, or by Creeks who married into the Apalachee community from Creek settlements north along the Escambia.

In addition, this same excavation produced what may be our best indicator for the mission yet: a fragment of what may be a crude Native-made candlestick, called Colono Ware by archaeologists in recognition of Native American manufacture in European-inspired forms (see photo). These wares are generally thought to have been made by mission Indians for the use of resident Europeans, which is consistent with the presence of a Franciscan missionary and as many as 16 Spanish cavalry soldiers at Escambe.

Using the Ground Penetrating Radar data obtained last week, we have also set in a new, larger excavation unit measuring 1 by 2 meters in the area of a subsurface disturbance that might be related to the mission-period occupation. When this unit is brought down to the level of this anomaly, we hope to discover whether it is a cultural feature such as a firepit or hearth, or perhaps just some natural disturbance. In either case, we should obtain a larger sample of mission pottery from this unit, which should give us better evidence about the time period and cultures involved at this site.

In sum, we now have a total of 17 contiguous shovel tests that have produced evidence for Native American ceramics in the area we are currently working in, making a site measuring at least 120 meters by 80 meters, and occupying the highest original ground surface along the river margin of the terrace we have been surveying. Most of the ceramics appear to date to the mission period, and are consistent with Apalachee pottery from this period. At this point, evidence is rapidly building that we may well have discovered the site of Mission San Joseph de Escambe. Further testing over the next few weeks will be designed to confirm this tentative identification, and begin the archaeological exploration of this site.

Zeroing in?

Over the past week, the crew has made progress on several fronts, though we lost a full day and a half of work to the rainy weather on Monday and Tuesday. As noted in our last post, we were intrigued by several underground anomalies discovered using Ground Penetrating Radar (GPR) equipment last Thursday, and since these anomalies were very close to a shovel test with a good amount of mission-period Native American ceramics, we decided to conduct additional remote sensing over this area. Dr. Thompson returned this week and conducted two sets of tests in a 20 meter by 20 meter area, using not just GPR but also soil resistivity equipment within a carefully-arranged grid pattern (see photo to right, and video below). Once the data is processed, both these surveys should provide more detailed information about the soil characteristics below ground in the 20 x 20 m. block around the positive shovel test, including information about subsurface pits, layers, and larger objects (revealed using radar data) as well as the relative resistivity of the soil to electrical currents, based on factors such as moisture and organic content. Based on these results, we hope to "ground-truth" one or more anomalies in order to see what they are. With any luck, we may encounter soil disturbances associated with mission-village structures or activity areas, though many may turn out to be natural.

At the same time, one crew has been continuing to excavate shovel tests farther south along the river bluff in an area where the river begins to veer away from the high ground we have been focusing our search on, while other crews have continued to test in the area where we have been finding quite a bit of mission-period Native American ceramics. The total absence of any evidence of Native American occupation to the south is in stark contrast to our continued testing in the area of our previous discoveries, where we now have a total of 8 contiguous shovel tests with traces of what we believe is probably mission-era occupation. Just today we excavated one of the most productive shovel tests in this area, with more than a few potsherds bearing characteristics in common with that documented mid-18th-century Apalachee Indians in the area of Pensacola and Mobile Bays.

Since our positive shovel tests are all clustered along the edge of a patch of woods that we have not yet surveyed (and which lies directly between both concentrations of mission-period Native American pottery we have discovered), students have also spent much of this week clearing transects through the brush in order to lay in new shovel tests. This is slow and tedious work, and the combination of heat and poison ivy has been a challenge. Nevertheless, tomorrow we should be able to begin testing in this area, which may turn out to be just as productive as the field to the south.

Applying some modern technology

Today we were fortunate to have a visit by Dr. Victor Thompson, also of the UWF Anthropology Department, who gave a demonstration of Ground Penetrating Radar (GPR) equipment at our field school site, as well as an overview of a range of shallow geophysical techniques including soil resistivity and gradiometer survey. We were able to conduct a few limited transect surveys in the area where we have been finding the most mission-period artifacts on the bluff summit, and we were pleased to discover several areas of disturbed subsurface soil, at least one of which is fairly broad (see image below). It's impossible to say what the cause of these radar "anomalies" are until we check them directly, but at the very least we know something is creating a fairly distinctive signature underground in this area.

The students also worked on more brush clearing after finishing up three open shovel tests for the longer holiday weekend.