Prehistoric site 41MI96 in Mills County, Texas was subjected to archeological data recovery excavations by staff archeologists from the Archeological Studies Program of the Texas Department of Transportation (TxDOT) in May 1999. This work followed an initial environmental review by TxDOT personnel that concluded that a proposed bridge replacement and associated realignment of a county road (CSJ: 0923- 23-011) had a high probability to impact previously unrecorded archeological sites. Subsequently, an archeological impact evaluation was conducted by TxDOT staff archeologists, under the direction of Dr. G. Lain Ellis. TxDOT investigations were conducted under Texas Antiquities Committee Permit No. 2193 to perform data recovery efforts at 41MI96 prior to development impacts. In 2012, TRC Environmental Corporation (TRC) of Austin was contracted by the Environmental Affairs Division of TxDOT through Work Authorization 57-109SA003 to conduct analysis on the recovered remains and complete a technical report of TxDOT’s field investigations and TRC’s laboratory findings in fulfillment of TxDOTs’ Antiquities permit.

Data recovery excavations consisted of the excavation of four mechanical trenches across two creek terraces (T1 and T2) and hand-excavations in two small blocks (Blocks 1 and 2) within the TxDOT right-of-way on the northwestern side of the project area. Hand-excavations in both blocks were initiated to target newly discovered burned rock concentrations encountered in the bottom of backhoe scrapings. A total 5.5 m3 of manualexcavation was completed, which was comprised of 16 total 1-by-1 m units, 11 in Block 1, and 5 in Block 2. Cultural materials were dominated by ca. 602 burned rocks and 2,846 pieces of lithic debitage, 89 informal and formal tools, but lacked diagnostic artifacts and faunal material. Six small, intact burned rock features were identified in Block 1 and were the focus of laboratory analyses. The scattered burned rocks and debitage from Block 2 were only tabulated and discussed in a general way, as TxDOT personnel believed they were in mixed context.

The six small burned rock features ranged in size from 33 to 100 cm in diameter and represented four intact heating elements (two with basins and two without), plus two small burned rock discard piles. Radiocarbon dating of organic residues in nine burned rocks from five intact features indicates multiple occupations over a span of roughly 700 years from 820 to 1450 B.P. (cal A.D. 560 to 1270). The lack of recorded depth measurements for cultural materials, combined with limited sediment deposition between the successive occupations, prevented isolation of individual occupational episodes. The lack of discernible vertical separation in the prehistoric occupations reflects slow soil aggregation during this period, likely lengthy surface exposure and possible erosion between events, and soil conditions which may also account for a near absence of charcoal and other organic materials such as vertebrate remains.

Four technical analyses (radiocarbon dating, starch grain, lipid residue, and high-powered usewear) focused on a limited suite of chipped stone tools, associated lithic debitage, and burned rocks collected from five of the six intact features in Block 1 in the T2 terrace. Starch grain analysis on fragments of 20 burned rocks from five features and 20 chipped stone tools from around the features in Block 1 yielded positive results from 47.5 percent of the specimens. Of considerable interest is the documentation, in addition to multiple grass species, of grains of the tropical cultigen maize (Zea mays) on two burned rocks each, from Features 2 and 3, plus on two edge-modified tools in the vicinity of those two features. One specific burned rock with a gelatinized maize starch grain on it was directly AMS dated to 980 ± 30 B.P. or cal A.D. 1020 to 1150. Some identified maize starch grains had been damaged through grinding, heating, and/or boiling, evidence of processing as a food resource. This indicates use of maize as a food resource in central Texas a number of centuries earlier than previously suspected. Lipid residue analysis on portions of the same 20 burned rocks from those five features yielded residues in 100 percent of the samples. The results indicate that both plant and animal products were present on all the rocks, with large herbivore lipids (likely bison or deer) present on at least one rock, and oily seed lipids present on at least three rocks. Residues from conifer wood products, here likely juniper trees, were present on 60 percent of the rocks, and indicate at least one specific wood species used to heat the rocks. High-powered microscopic use-wear analyses on 15 chert tools (11 edge-modified flakes, 2 biface fragments, and 2 complete choppers) revealed their use in processing wood, plants, bone, and hide as well as unspecified soft and hard materials.

The sparse frequency of formal chipped stone tools likely reflects the limited area investigated and also the possibility that these occupations reflect low-intensity and short-term camps that focused on preparing and cooking a few food resources in heating facilities and the manipulation of other perishable resources.

The lipid and starch analyses of the burned rocks provides important information concerning the resources cooked by the rocks in these small burned rock features, most significantly the presence of maize and wild native grasses. These resources would have gone unidentified without these specialized analyses. Continued use of these two analytical techniques on suites of burned rocks from other features/sites in and around central Texas will provide an empirical basis for identifying changes in subsistence patterns over time and across geographical space. It is also notable that direct radiocarbon dating of organic residues contained within the porous sandstone burned rocks here has succeeded in providing satisfactory chronological control for the features and site, strongly indicating that this technique can be beneficially employed in the future in cases where other organics such as wood charcoal, charred seeds/nuts, and/or bone are unavailable for absolute dating.

In 1999 the Texas Historical Commission accepted TxDOT’s field investigations as sufficient and concurred with TxDOT’s recommendation that no further work was necessary under Texas Antiquities Committee Permit No. 2193. Parts of site 41MI96 outside the current TxDOT right-of-way have not been fully evaluated. Based on the present findings and the excavated intact features in Block 1, it appears that potentially eligible deposits may be present beyond the current right-of-way. If TxDOT further expands this county road, it is recommended that those areas at 41MI96 be evaluated prior to surface modifications related to that project.

Licensing Statement

This is a work for hire produced for the Texas Department of Transportation (TxDOT), which owns all rights, title, and interest in and to all data and other information developed for this project under its contract with the report producer. The report may be cited and brief passages from this publication may be reproduced without permission provided that credit is given to TxDOT and the firm that produced it. Permission to reprint an entire chapter, section, figures or tables must be obtained in advance from the Supervisor of the Archeological Studies Branch, Environmental Affairs Division, Texas Department of Transportation, 125 East 11th Street, Austin, Texas, 78701.


Submission Location


Tell us how this article helped you.


To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.