From Trash Pit to Pit House

In my last post, I described the discovery of Feature 09-04, a large concentration of charcoal and ash that produced a distinct magnetic signature. Upon excavation, this feature produced rather large quantities of artifacts and food remains such as animal bones. But late this past week, we realized that this feature is not just an over-sized garbage dump, but something else entirely. As we quarter-sectioned this big stain, we identified a concentration of ash deposits and burned soil near the center, a classic hearth feature. Then, at the outside edge of the feature stain, we discovered a ring of regularly-spaced post molds which most likely represent a superstructure of some kind.

Now, why would the prehistoric inhabitants of this site construct a roof over their trash pit? Of course they would not, but they would build a roof over a dwelling. Based on these clues, we concluded that Feature 09-04 is a pit house, a small structure used by just a few people.

Similar pit houses have been found at Late Woodland sites in Illinois and Pennsylvania. Our own excavations at the White Fort site in 2002, exposed six such structures. All have the distinctive "key-hole" plan form which consists of a round to oval basin with an elongated entryway. The image of Feature 09-04 below shows this characteristic form.

The quarter section profiles reveal a rather complex stratigraphy in the fill of this pit. As shown in the image below, most of the fill consists of lenses of trash deposits and sandy soil which were dumped into the pit after the dwelling was abandoned. Beneath this fill are several thin strata of carbon-rich soil which represent the successive floor levels of the dwelling. A distinct floor layer is indicated by the red arrow in the image below.

The central hearth feature also exhibits distinct layering of ash and burned sand which indicates that successive fires were constructed during each occupation. The distinctive entryway points to the southeast and may have been constructed to prevent large quantities of warm air from escaping the interior. This would have been an important consideration if the dwelling was inhabited during the winter months, a likely conclusion. The few decorated rim sherds recovered from the pit house indicate that it was inhabited during the fourteenth or fifteenth centuries A.D.

Finding the First Features

As I mentioned in my last post, our shovel-test survey discovered several prehistoric features. Interestingly, two of these features were not identified by the geophysical surveys conducted by Jarrod Burks. That does not mean that Jarrod did a poor survey, it just indicates that not all the features at the site contain material that is sufficiently magnetic to be detected by his instruments.

One of our previously unknown pits is Feature 09-02 which was excavated by Glen Boatman's crew. The cross-section of this feature (shown below) revealed it to be a rather deep pit with several layers of fill. It contained a nice sample of grit-tempered pottery with rather fine cordmarking. Such pottery is called Esch Cordmarked and appears to be the typical ceramic made by the Middle Woodland inhabitants of the site. In some respects, this pottery resembles McGraw Cordmarked which is the most common type of ceramic found on Ohio Hopewell sites in southern Ohio.


In my last post, I mentioned the shovel-test with a projectile point in the wall. Beneath the point, Jim Bower's crew found another apparent Middle Woodland feature (Fea. 09-03). They have just begun the profile of this feature but already have found one complete Flint Ridge bladelet and several sherds of Esch Cordmarked pottery. Once Jim's crew troweled the floor, we noticed a distinct line of post molds running across the feature and into the southeast corner of the unit. This line extended across the dark fill of the feature and showed up clearly in cross-section as shown in the image below. The observation that these post molds penetrate Fea. 09-03 demonstrates that the posts were set sometime after Feature 09-03 was filled. We will try to trace this post line to see if it is part of a structural outline (house?) or perhaps part of a stockade barrier.


In order to test the accuracy of the magnetic survey, we placed a 3m by 3m unit over Anomaly 100 identified by Jarrod Burks. Once the plow zone was removed, a very large pit feature was revealed, as shown below.


We have just begun the cross-sectioning of this pit, called Fea. 09-04, but have already exposed distinct lenses of ash and charcoal as well as bits of fish and mammal bone, flint debitage, and fire-cracked rock. One small pottery rim sherd bears an incised decoration typical of Late Prehistoric period pottery made by the Sandusky Tradition inhabitants of the region. Thus, it seems that this feature relates to the late period village occupation of the site, even though this feature lies outside (to the south) of the oval-shaped enclosure detected in the magnetic survey. We are quite excited about the subsistence evidence and other information that will come from this apparent trash pit.

So, all-in-all, this was a very productive week thanks to the hard work of our field staff and first-week field school crew.

Field School Begins at Heckleman site

This past Friday, we completed our first week of field school excavations at the Heckleman site. On Monday we began a shovel-test survey of the eastern end of the site within the large enclosure. This type of survey involves the excavation of 50 cm by 50 cm squares along east-west transects. The goal is to sample the contents of the plow zone in order to identify the spatial distribution of artifacts such as pottery, flint debitage, animal bone, fire-cracked rock, and more recent historic material. By end of the day on Tuesday our crew had excavated 16 shovel-tests and recovered good samples of artifacts. The counts of artifacts derived from this survey will be used to construct maps of artifact densities across the site. We are looking for patterns in these distributions which may help identify what activities were carried out by the prehistoric inhabitants and where these took place. In the image below, you can see small groups of students working on shovel-tests along the 490N line.


Below is a close-up view of a shovel-test in progress.



Our testing paid some immediate benefits when we uncovered several possible pit features and post molds. The small shovel-test units at these locations were expanded into larger units to exposed the suspected features and permit their excavation. I will discuss some of these discoveries in subsequent posts. In one excavation we uncovered the base of a Middle Woodland (Lowe cluster) projectile point at the base of the plow zone and just above a feature. The point can be seen in the lower right corner of the profile wall shown below.