Archaeology Curation History and Heritage Uncategorized

Amanda Franklin, Texas Historical Commission – Archaeology Division

I chose to complete my internship at the Texas Historical Commission to learn as much about the curation process as possible and try to do a little bit of everything. I learned from hands on experience about the steps necessary in the field to facilitate curation and how to create artifact inventories, site forms, and database entry. Before this internship I never realized how many collections are in a sort of curation limbo that never reach the museum level.

The Fort St. Louis/Keeran Ranch site was the main focus of my internship. I performed ceramic, faunal bone, and shell identification and basic level analysis, created a skeletal inventory for human remains from the site, labeled and resorted artifacts based on provenience, and learned how to fill out an atlas Tex-site form for an individual collection, and completed a lot of artifact database entry.

The Ft. St. Louis Site, located in Victoria, Texas, was the area where La Salle attempted to establish a French colony in the mid 1680’s. The original mission given to La Salle was to establish a French colony at the mouth of the Mississippi River to gain control of the river and the Spanish silver mines, but instead he landed along the Gulf Coast of Texas in Matagorda Bay. After ships running aground or sinking with most of the colonies supplies still onboard, then ongoing Indian attacks, the colony failed and La Salle was killed by his own men while attempting to walk to the nearest French colony for help against the Spanish and the natives.

Ceramic Analysis

The excavations of La Salle’s former fort recovered 53, 142 sherds of ceramics produced in Europe, Asia, and Mexico. The collection needed to be relabeled, checked for consistency, and reorganized by type. The French assemblage included faience fine ware, Beauvaisis and Bass Normandie stonewares, Saintonge, and red paste utility wares all manufactured in France or Germany and brought by La Salle’s expedition. The Spanish and Mexican assemblage, from the later 1720 presidio, dominates the total collection and consisted of Olive Jar ceramics made in southwestern Spain, Mexican majolica fine wares, and utility wares made by Indians and Mestizo in Mexico. Finally, a good portion of Chinese porcelain imported through the Philippines into Acapulco appeared in association with the Spanish assemblage.

Each different type of ceramic has unique attributes that help to identify it. The fine wares were created uniquely with different technology to mimic Chinese porcelain. For example, French faience usually has a soft earthen buff or light pink paste covered by an opaque lead and tin oxide glaze. Vessels were decorated by dipping them into the glaze and hand painting designs on top before firing them. The decoration on the French faience was plain and undecorated, blue on white, blue and orange on white, blue on blue, green on white, and unglazed.
French earthenware was composed of a dark fine grained red paste which a variety of different lead glazed finishes. Some of the earthenware was punctuated along the exterior or stamped. The French also had earthenware composed of a fine-grained white paste. The wares produced in Spain were limited to the olive jars. The olive jars were made with a light orange to gray colored paste that usually had air bubbles with various dark mineral inclusions, and the exterior were entirely bisque or had a light green glazed interior. Finally the Chinese porcelain was the most technologically advanced ware of the period, and easy to distinguish from other mimicked attempts. The paste was hard, composed of kaolin and feldspar, highly refined, and had a glass like appearance.


Figure 1. Hooked Mussel. This photograph is from the 41VT131 collection at THC, and shows an example of the coloration and ribbing pattern of the American Hooked Mussel.

Thanks to several ecological factors, shell analysis from an archaeological site can provide important insight into absolute dating of sites, peoples’ subsistence patterns, trade/migration patterns, and alluvial/saltwater levels. First, certain species of mollusks like clams, scallops, mussels, and oysters are found in different habitats. Oysters are found in higher salinity waters, while clams and mussels can be found in more brackish freshwater areas like rivers and streams. Second, different species of mollusk are more commonly found during certain seasons, and this could provide an idea of when people collected that type of mollusk. Besides the seasonality of mollusk availability, most species of mollusk have a certain geographical range. The geographical range can be very broad to encompass the coast of Virginia to Mexico, or highly specific to the Gulf Coast of Texas.

Mollusk shell can be identified by its internal/external coloration, shape, texture, umbo (hinge), and tooth morphology. The weight of the umbos and fragment can provide an idea for the amount of meat obtained, and some show indication of discoloration by burning that can indicate either trash burning or cooking. A general calculation for the minimum number of individual mollusks (MNI) present is to count the number of intact umbos and dividing by 2. It is also possible to determine which side of the shell is present by the muscle scars left on the interior of the shell. The top/left portion of the shell tends to be shallower than the bottom/right side containing the muscle scar.

Two shovel test survey sites associated with Ft St. Louis needed curation and analysis. The sites were VT131 and VT132 near Garcitas Creek in Victoria. There were 3 main species found in the Keeran Ranch shovel test survey assemblages, including: American oyster (Crassostrea virginica), Atlantic brown Rangia Clam (Rangia cuneata), Southern Ribbed Mussel (Geukensia granosissima), and the American Hooked Mussel (Ischadium recurvum). The Rangia clams were easily identifiable by their symmetrical rounded shell shape and brown exterior that often eroded away. The American oyster was also easily identifiable by its irregularly shaped and layered shell appearance. The American Hooked mussel has an elongate to oval shaped shell with highly ribbed purple lines along the exterior and a very iridescent purple to black colored interior. The Southern Ribbed Mussel was also elongate to oval shaped, but the exterior was generally lightly ribbed, white, and had an iridescent light pink to blue interior.

The majority of shell found in the Keeran Ranch shovel tests were Rangia clam and Oyster by weight, but the mussels appeared greater in abundance. The greater abundance of mussel could be attributed to their more fragile thinner shells, and the amount of meat found inside is usually much smaller than clam and oyster.

Human Remains and Bone Analysis

Figure 2. TARL Comparative Collection. This photograph shows one of the comparative collections at TARL. The collection shown is the metatarsals or many different animals.

Human remains are an important part of many excavation sites, and processing the remains in the lab becomes even more important to preserve and maintain them. After initial analysis and isotope testing, the remains are stored at a research facility. Some archaeological sites are located on private property, so the methods involved in preservation or repatriation can be difficult.

The Ft. St. Louis Site is located on private property, and the family donated all of the archaeological material to the Museum of the Coastal Bend in Victoria, Texas. The property owners and relatives of the excavated individuals ultimately want the reburial of the remains once analysis is completed. Before reburial, the museum wanted to have a facial reconstruction performed from the cranial material found in the burials of several individuals.
All of the skeletal remains were still in the possession of the THC, so they were brought to the lab to create a specimen inventory before permanently returning to the hands of the Museum of the Coastal Bend. Each bone, tooth, and element had to be assigned an arbitrary lot number (ID). Other information recorded in the inventory was provenience (if any), a brief description, box number (which one it is in), quantity, and type (human bone fragment, human tooth, human cranium).

After the form was filled out for each item, a new artifact label was placed with the item in its own individual bag. Most of the remains came loosely wrapped in tissue paper, open Ziplocs, or pill bottles that broke when opened. A lot of information about individual bones was lost this way and fragments mixed together, or the fragments were further fractured into many new pieces. In the end, new archival quality Ziploc bag and acid-free labels in a protective polymer sleeve were made for the fragments to help with preservation and protection of the historic remains.

Another important part of bone analysis in the lab is sorting through fragments into more specific and informational categories. A large amount of bone appeared in the VT131 and VT132 collections. The bone was sorted from the shell, organized into smaller more defining categories, and curated.

The Texas Archaeological Research Lab (TARL) has an excellent paleontology division that is exceedingly helpful with multiple comparative collections, fully articulated skeletons, and countless boxes of bones organized at the species level to look at. One of the biggest key features to analyzing faunal remains is the basic concept “form follows function” to understand how an animal moves to understand how their bones would appear. Animal bones are especially easily confused with those of juvenile or fetal human bones. The best way to distinguish between the two is looking for fused sutural lines and ends of long bones that could only occur in fully developed “adult” individuals. Once nonhuman bone is identified, the best way to further distinguish the genus and species of animal is from guides, comparative analysis, and help from a more experientially qualified individual.

The smaller Ft. St. Louis shovel test surveys were sorted and analyzed at the most basic level by separating the bones and fragments into broad genus/order categories of medium to large sized Mammalia, small sized Mammalia, Rodentia or Lagomorph (rodent, rabbit/hare), Atigonia (fish), and Chelonia (turtle).
Teeth size and shape are very distinct and informational, because they are based upon diet. Herbivores have very different shaped teeth than omnivores or carnivores, and the overall size of the animal’s mandible determines the size of the tooth. Most of the animal teeth found in the Keeran Ranch shovel test surveys that were identifiable on a basic level were large ungulate most likely belonging to deer, small rodents’ incisors, and rabbit incisors and molars.

Although some bones are unique and easily identifiable, most fragments are extremely hard to identify using pictures with limited hands on experience. Quite often, the assistance of a willing osteology instructor, Maggie McClain, from Texas State aided in the identification of some questionable whole bones. The overall assemblages of faunal remains identified were consistent with more extensive prior faunal analysis excavated in the same area.


I really learned an extensive amount of information during my internship at the Texas Historical Commission. The staff was very helpful and full of information, not only about the individual collections or artifacts but also their experience in graduate school and career advice. I never realized the full importance and the large amount of work that is involved once artifacts leave the excavation site, and detailed notes involving an artifact’s context and the site in general greatly facilitate curation in a laboratory often much later in time. The full hands on experience with all the artifacts from different collections was amazing, and though the work can be a little tedious at times I felt like it was important work that needed to be done and make my mark in history, in a tiny anonymous way of course.

About the author

The Internship Coordinator

Official Texas State University Disclaimer
Comments on the contents of this site should be directed to Adam Clark, Mary Gibson, Megan McSwain, or Neill Hadder.