Shumla’s Chemistry Lab Part II: Radiocarbon Dating Eagle Cave Rock Art

**This the second of two blog posts describing the Shumla Chemistry Laboratory. This post is based on the poster presented by Karen Steelman at the 2017 Texas Archeological Society Annual Meeting in Grapevine, Texas, titled: “Radiocarbon Dating Rock Paintings: New Plasma Oxidation Laboratory at Shumla.”**

By Karen Steelman

In the previous Shumla Chemistry Laboratory blog post, I gave an introduction of radiocarbon dating and plasma oxidation. For this post I wanted to shift gears and describe how we are using plasma oxidation and radiocarbon dating to address ongoing research questions. During The Alexandria Project, we are collecting data about the different iconographic images or symbols that are found throughout Lower Pecos rock art. With the Shumla Chemistry Lab I am interested in pursuing similar research questions dealing with determining the age of styles and repeated motifs across the landscape.

We know from previous research that some attributes within Pecos River Style rock art are spatially limited in their distribution (e.g., Harrison 2004). Rather than just obtaining radiocarbon dates on any Pecos River Style figure, one of our goals is to target specific attributes to better understand how the production of rock art imagery may have changed through time. This type of research has drastic implications for understanding not only the chronology of Lower Pecos rock art, but also provides insight into the visual language used by the hunter-gatherers who produced the pictographs.

Radiocarbon Research in Action: Dating Rabbit Ears

One of the Pecos River Style attributes we are interested in dating is what we call “rabbit ears.” Rabbit ears is a descriptive term we use to define one, or often 2, ovoid protrusions that extend upward from the heads of Pecos River Style anthropomorphs. Rabbit ears occur most frequently at sites west of the Pecos River. We do not know what rabbit ears represented within the culture of Lower Pecos hunter-gatherers, but it is a repeated attribute that we are interested in learning more about.
Pecos River Style anthropomorph at Eagle Cave with one black and one yellow rabbit ear. Image is enhanced with DStretch LDS channel.

Dating Rabbit Ears at Eagle Cave

In 2014 we were fortunate enough to be able to collect three paint samples from Pecos River Style anthropomorphs in Eagle Cave. One of our research questions was specifically dating the rabbit ears at Eagle Cave, but we also wanted to date the Eagle Cave pictographs to chronologically link the rock paintings to the physical deposits that were being sampled at the time by the ASWT Project at Texas State University.

Eagle Cave. Image courtesy ASWT Project.
Eagle Cave during excavations. The pictographs are on the shelter wall in the background. Image courtesy ASWT Project.
The main pictograph panel at Eagle Cave is on the downstream end of the shelter, and contains several excellent examples of Pecos River Style anthropomorphs with rabbit ears. The rock art at Eagle was fully documented by Shumla during the 2013 Field School, so we already had identification numbers for each of the figures we sampled (A007, A011, and A013).
The main pictograph panel at Eagle Cave in real-color (top) and DStretch YBK enhancement (bottom). The figures sampled for radiocarbon dating are highlighted.




A007 is a small, black, Pecos River Style anthropomorph located beneath the arm of the largest rabbit eared figure at Eagle Cave. These two photographs show before and after the sample was collected.
A013 is an elaborate Pecos River Style anthropomorph with wings, tail, and speech/breath extending upward on the panel. The sample location is highlighted in yellow, and the photos of the sample location before and after collection are shown on the right.
A011 is a Pecos River style anthropomorph with a single black rabbit ear. The sample location is highlighted in yellow, and the photos on the right show the sample location before and after.
The sample collected from A007 was a single spall that was about to fall off the shelter wall.
On each anthropomorph we targeted locations that were about to spall away so that we would do as little damage to the pictographs as possible. We mapped in each sample with a total data station, and did extensive photo documentation of each location before and after the samples were collected.

Plasma Oxidation of Eagle Cave Samples

Back in 2014 I was still teaching at the University of Central Arkansas. So after we collected the Eagle Cave samples, I headed back to my lab to begin the process of extracting carbon from the paint for dating. During sample processing, my students and I observed calcium oxalate accretions coating and underlying the paint layer within each sample.

We do not know exactly how calcium oxalate forms, but one of the theories is that microbes and lichen growing on the wall in the past excreted oxalic acid that precipitated with the calcium on the wall to form calcium oxalate (Hess 2008; Russ et al. 1996; Russ et al. 2000). Regardless of how the oxalate formed, we do know that the carbon within calcium oxalate is contemporary with the formation of the accretion. This was very exciting for the Eagle Cave samples because it would allow us to not only date the paint, but also the surface above and beneath the paint (see references cited in Russ et al. 2017). Dating a paint layer and dating oxalate accretions has been done before, but only twice that I know of on the same painting (Steelman et al. 2002; Rowe & Steelman 2003; Russ et al. 2017). In this study, for the first time we get minimum, maximum, and paint layer dates!
This thin paint flake was embedded in resin, sliced, and polished so that we can see the paint layer in profile or cross-section. You can see the thin, vibrant paint layer with an accretion coating. Oftentimes, when we look at painted shelter walls, the paintings appear “faded”, but the mineral pigment is as vibrant as it always has been, it is just obscured by this mineral accretion.
This microscopic photograph of red and black paint within A011 at 50x magnification shows the undulating and rugged calcium oxalate that covers the paint.

Preparing the Samples

Because we needed to isolate the oxalate layers from the paint layer, the first step in the process was using sterile scalpel blades to scrape each layer (top oxalate layer, paint layer, and bottom oxalate) into sterile test tubes. We then treated each sample to remove carbonates and any organic contamination.
First, the overlying and underlying accretion layers are analyzed with Fourier Transform Infrared Spectroscopy (FTIR) to confirm the presence of calcium carbonate (blue circles) and calcium oxalate (green squares) as you see here.
We then use phosphoric acid to treat the samples until all of the calcium carbonate limestone is removed. We then use FTIR to confirm that the carbonate is removed and the calcium oxalate (green squares) remains.
A former University of Central Arkansas chemistry student is using a scalpel to remove oxalate from a paint sample.
Once each sample was isolated, purified, and prepared, they were placed one at a time into the plasma oxidation combustion chamber following the plasma oxidation steps in the previous blog post. Once the carbon dioxide was collected from each sample, they were sent for accelerator mass spectrometry (AMS) C-14 measurement.

Eagle Cave Results

When we submitted the samples for AMS C-14 measurement we were expecting the lowest oxalate to be oldest, the paint sample to be in the middle, and the top oxalate to be the youngest age. When we got results back from the AMS lab we could not have been more excited because the three ages lined up perfectly!
Uncalibrated radiocarbon results for paint and oxalate layers. The x-axis is labeled as years BP or before present.
I was very happy and excited!
Unfortunately we were not able to get dates on the top oxalate, paint, and bottom oxalate for all three figures. The dates we did get indicate the bottom oxalate formed between 5000-6500 years BP, the paintings were applied between 3000-3700 years BP, and the top oxalate formed between 1800-2800 years BP.

Implications of Eagle Cave Dates

All three direct paint dates from Eagle Cave statistically overlap with each other, and the images were all painted around 3280 BP. Since all 3 figures have similar ages, this further supports the hypothesis that Pecos River Style images are part of murals that were likely painted at the same time, as opposed to a random collection of images.
Radiocarbon results are analyzes using the OxCal computer program to produce calibrated age ranges. The weighted average for all 3 direct paint dates is 3280 +/- 70 years BP, which calibrates to 1740 to 1420 cal BC.
Interestingly, we have a previous age for Pecos River Style imagery from Jackrabbit Shelter, where there are also anthropomorphs with rabbit ears. The age of the figure from Jackrabbit is approximately 3400 BP (Bates et al. 2015), which agrees very well with the dates at Eagle Cave. Does this mean that other sites with rabbit ear imagery like Rattlesnake Canyon were also produced around 3200 – 3400 years BP?  We are not sure, but it does bring up the necessity of further work.  As we continue with The Alexandria Project, we are developing a research design for radiocarbon dating repeated motifs across the landscape. We have a lot of work ahead of us!
Pecos River Style anthropomorph at Eagle Cave with rabbit ears. Image is enhanced with DStretch LDS channel.
Pecos River Anthropomorph at Jackrabbit Shelter with rabbit ears. Image has been enhanced with DStretch LDS enhancement.
Pecos River Anthropomorph at Rattlesnake Canyon with rabbit ears. Image has been enhanced with DStretch LDS enhancement.


This Eagle Cave project was a collaboration between Shumla, University of Central Arkansas, and the ASWT Project at Texas State University. In addition to continuing work in the Lower Pecos, Shumla is seeking to collaborate with archaeologists and rock art researchers around the world.  We are currently working with a team in Australia exploring chronology and spatial distribution patterns across the landscape of the SW Gulf of Carpentaria islands. With collaborations, we can all learn from each other to improve rock art research world-wide.


Bates LN, Castañeda AM, Boyd CE, Steelman KL. 2015 A Black Deer At Black Cave: New Pictograph Radiocarbon Date for the Lower Pecos, TX. Journal of Texas Archeology and History 2: 45-57.

Harrison, James B. III 2004 Rock Art Boundaries: Considering Geographically Limited Elements Within the Pecos River Style. Master’s Thesis, Deparatment of Anthropology, Texas A&M University, College Station.

Hess, Darren, Dana Jo Coker, Jeanette Loutsch & Jon Russ 2008 Production of Oxalates In Vitro by Microbes Isolated from Rock Surfaces with Prehistoric Paints in the Lower Pecos region, Texas. Geoarchaeology 23, 3-11.

Rowe MW & Steelman KL. 2003 Comment on “Some evidence of a date of first humans to arrive in Brazil”. Journal of Archaeological Science 30:1349-1351.

Russ J, Pohl MD, von Nagy CL, Steelman KL, Hurst H, Ashby L, Schmidt P, Padilla Gutiérrez EF, Rowe MW. 2017 Strategies for C-14 Dating the Oxtotitlán Cave Paintings, Guerrero, Mexico. Advances in Archaeological Practice 5(2): 170-183.

Russ J, Loyd DH, Boutton TW. 2000. A paleoclimate reconstruction for southwest Texas using oxalate residue from lichen as a paleoclimate proxy. Quaternary International 67:29-36.

Russ J, Palma RL, Loyd DH, Boutton TW, Coy MA. 1996. Origin of the whewellite-rich crust in the Lower Pecos region of southwest Texas and its significance to paleoclimate reconstructions. Quaternary Research 46(1):27-36.

Steelman KL, Rickman R, Rowe MW, Boutton TW, Russ J, & Guidon N. 2002 “AMS Radiocarbon Ages of an Oxalate Accretion and Rock Paintings at Toca do Serrote da Bastiana, Brazil.” In Archaeological Chemistry VI. Edited by K Jakes, pp. 22-35. Washington, DC: ACS Symposium Series.

1 Comment

  1. Great job Karen, I enjoyed your article very much. After being in Eagle cave this spring it was interesting to hear your findings in regards to the rabbit ears. Keep up the good work!


Submit a Comment

Your email address will not be published. Required fields are marked *