Graham-Applegate Rancheria

Ancient Houses of Central Texas

For many decades, archeologists were pessimistic about ever finding remains of prehistoric houses in central Texas. The mobile bands of hunter-gatherers who lived in this part of the state were thought to have constructed very flimsy huts or skin-covered tents that would leave little trace for archeologists to find, especially after hundreds or thousands of years. This changed in the early 1960s when unusual stone circles were discovered eroding out of the sandy shoreline of Lake Buchanan, a large artificial reservoir 60 miles northwest of Austin. There was speculation at the time that these stone circles were the remains of ancient structures, although other explanations also were proposed. Some thought they might be the remains of specialized cooking facilities. Excavations of similar circles at the Lion Creek site in Burnet County in the mid 1970s and the Turkey Bend Ranch site in Concho County in the late 1980s provided additional evidence that these features were probably the foundations of houses built by Archaic and Late Prehistoric peoples.

The remains of the five houses excavated so far at Graham-Applegate are very similar to the stone circles discovered earlier elsewhere. As at the other sites, all that has survived of the structures are a central stone pavement where the hearth fires burned and a ring of stone clusters surrounding this pavement that helped support the walls. Exactly how the outer ring of stones of these houses supported the walls in unclear. Some stones may have been used to hold down the hides or whatever material was covering the structures, while others braced the bases of the thin saplings that are believed to have supported the walls. At the Graham-Applegate rancheria, the stones likely were used primarily for the latter reason, that is, to stabilize the posts in the loose, gravelly soils around the site. While no traces of post holes (or woodpost stains) can be seen in the soil, the elevation of these rocks in relationship to the floor surface indicates that, in most cases, the posts were placed in shallow holes dug by the aboriginal builders. The rocks would have been placed around the base to prevent the posts from shifting or coming out of the ground during strong winds. It is also possible that a shallow trench was dug completely around the wall area, in part to create a trap for rain water, but also for the wall posts.

We can only speculate as to what the superstructure may have looked like based on materials and technologies available to the people during the Late Prehistoric. The brief descriptions left by early explorers of North America and the more-detailed descriptions of recent houses built by people with a similar level of technology also aid us in reconstructing the rancheria houses. Most likely a dome-like frame was constructed by arranging a ring of green sapling posts upright in the ground, and then bending upper ends inward and tying them together. Small branches would then be attached horizontally to these posts to form a framework to support the covering. Many covering materials would have been available, and early Spanish accounts of Indian houses in what is now south Texas mention grass, woven mats, and animal hides.

There is a curious aspect to these houses that sets them apart from other hunter-gatherer dwellings. Much of the interior space was devoted to a large, central paved area or hearth, leaving only a relatively narrow (a meter or less) ring of living area between the hearth and walls (note the spatial arrangement in the artist's scene of hut interior). With some of the houses, it is possible that the living space extended onto the stone pavement, while in others the pavement seems to have been raised somewhat above the floor. One theory is that the pavement served as a "work bench" of sorts, where tasks could be performed above or away from the floor and close to the fire for light and warmth. For obvious reasons, a large fire would not have been built on these hearths inside a thatch-covered structure, and the archeological evidence supports this. Very little charcoal staining was noted in the hearth fill of House 1—fires were built there but not intense ones such as those in the earth oven. Some Indian groups in the American West would heat rocks outside their dwellings and then bring the hot stones inside for heating without the hazards of smoke and flames. The presence of these large central hearths has led to speculation that this type of structure was used as a sweat lodge, but evidence from Graham-Applegate seems to support the view that these were places where people lived and carried out domestic activities. Several of these structures are in very close proximity to one another, which would be unusual for sweat lodges. Three of the structures have small hearths outside their walls, along with food remains, more in keeping with a small hut-like dwelling where cooking was done outdoors whenever possible.

The Graham-Applegate Houses

House 1 is the largest and the best preserved of the five houses so far excavated at Graham-Applegate. The rocks used to stabilize the walls (wall supports) form a nearly complete and well-defined ring. The large gap in this ring on the south side is a clear indication that the doorway faced in that direction, away from the prevailing winds of winter cold fronts or "northers." The diameter of the entire structure is 4.25 m (14.2 ft) and has a floor area (including central hearth) of around 12.5 square meters. A large stone pavement or central hearth covers the middle of the floor area. It provided a platform for the small fires that warmed the house and perhaps also served as low table or work bench. A cross-section of the hearth reveals that a shallow basin was first dug and then filled with stones of different sizes to create a level surface. This house also contains many more artifacts both on the floor area and outside it than the other houses and has at least five small cooking hearths outside its walls. The clutter about this structure might mean it was occupied (and perhaps reoccupied and rebuilt) over a long period, allowing more refuse to build up around it. Or perhaps House 1—with its larger size and close proximity to the communal earth oven—was the focal point for activities of the entire rancheria.

House 2 is a much smaller structure than House 1 and is similar in size to the three houses that were discovered later. The outer diameter is a little over three meters (nearly 10 feet) and has a floor area with the central hearth of about seven square meters. Perhaps this is the "normal" size of houses for the rancheria. The central hearth in this house, though of a slightly smaller diameter, is much like the one in House 1. Circular in plan, it contains large, nearly boulder-sized rocks as well as smaller ones. The surface is slightly uneven which could have resulted from later stone robbing. The rocks that supported the posts and wall structure of this house do not form a continuous ring. Bedrock lies close to the surface in areas where there are no rocks, and therefore digging pits in which to place posts would have been next to impossible. Still, something had to support the wall in those areas, and stones might have simply been placed on the ground around the posts without first digging a hole. The exposed rocks would then have been vulnerable to later stone robbing after the house was abandoned.

Houses 3, 4, and 5 were laid out in a closely spaced linear arrangement on what is now the edge of a densely wooded area that extends north of the site. The three houses share many construction details, including unusual triangular—rather than round—central hearths composed of cobble-sized rocks. It seems obvious that the structures were built at the same time by closely related households: they are set apart both by distance and in the way they were constructed from Houses 1 and 2. What all this means is not yet clear. Unfortunately, we are not exactly sure when these houses were built. The projectile points found with them (or next to them) cover a period of many thousands of years (from Late Paleoindian to Late Archaic) and no charcoal for radiocarbon dating has yet been recovered. The unexcavated central hearth of House 5 contains a distinct carbon stain in the fill around the rocks and eventually may yield charcoal.

A singular pile of quartz rocks two meters long connects Houses 3 and 4, stacked there by the rancheria folk for some unknown purpose. Quartz was not a very useful rock for these people. It tends to break rapidly into many small, sharp shards when exposed to heat, making it a poor choice for cooking stones. Unlike quartz crystal, the quartz found around the site area does not flake well and this rock was rarely used to make chipped stone tools. Perhaps they were stacking the stone there simply to get it out of the way.