Final Report on the 2009 Excavations at the Rutherford B. H. Yates House Site

Transcription

1 Final Report on the 2009 Excavations at the Rutherford B. H. Yates House Site by the Rice University Archaeological Field Techniques class (Anthropology 362/562) Dr. Jeffrey Fleisher (Rice University) Dr. Carol McDavid (CARI) Mr. Robert Marcom (CARI) Edited by Jeffrey Fleisher

2 Acknowledgements We are grateful to the Board of Directors and Advisory Board, Rutherford B. H. Yates Museum, Inc. (RBHY), the Yates Community Archaeology Project (YCAP), and the Community Archaeological Research Institute (CARI), for their support of excavations at the Yates House by the Rice University Archaeological Field Techniques class. CARI/YCAP co-directors, Carol McDavid and Robert Marcom gave generously of their time and energy during class time and while excavations were ongoing. In the field, thanks are due especially to Robert Marcom who helped oversee the excavations as well as the geochemical sampling at 1512 Wilson St. Thanks also to Megan Quinn, who helped orient the students to the Yates cataloging system and provided important lab support during the excavations. Finally, we gratefully acknowledge the financial support of the Center for Civic Engagement at Rice University which made our collaboration with the Yates Museum possible. Stephanie Post, Executive Director of the CEC, has been extremely supportive of this research, and helped ensure that funding was available. We are pleased by another successful season at the Yates House, and look forward to future, productive collaborations with the Rutherford B. H. Yates Museum and CARI. Jeffrey Fleisher Assistant Professor of Anthropology Rice University

3 Table of Contents 1. Introduction 1 Jeffrey Fleisher 2. Excavation Methods and Excavated Levels 5 Elizabeth Johnson 3. Ceramics 15 Babatunde Babalola 4. Glass Artifacts 39 Courtney Ng 5. Metal Artifacts (Non-Nail) 93 Ben Dwyer 6. Nails 109 Dustin Tannahill 7. Faunal Remains 127 Kylie Klein 8. Artifacts of Personal Adornment 141 Brittany Wise 9. Miscellaneous Artifacts and Building Materials 181 Nadine Boudreaux 10. Geochemical Sampling at 1512 Wilson St. 205 Sarah Nouri 11. Conclusion 239 Jeffrey Fleisher

4 Introduction Jeffrey Fleisher Background and Goals During the spring 2009 semester Rice University students and faculty conducted archaeological excavations at the Rutherford B. H. Yates Museum site (41HR980), as part of the University s course entitled Archaeological Field Techniques (ANTH 362). Students were supervised by Rice faculty member Dr. Jeffrey Fleisher as well as Dr. Carol McDavid and Mr. Robert Marcom, both affiliated with the Yates Community Archaeology Project (YCAP) and the Community Archaeological Research Institute (CARI). Students were guided through a hands-on course of excavation during four weekends in late January and February, followed by artifact analysis and interpretation at the Rice University Archaeology Laboratory. The aims of the excavation reflected objectives of both the Archaeological Field Techniques course and the Yates Community Archaeology Program. In this, the second year of collaboration between the Yates Museum and Rice University s Anthropology Department, the Yates site provided, yet again, a unique and important opportunity for Rice students to learn archaeological excavation and analysis techniques. Excavations at the Yates House have proven to be rich in historical artifacts (see McIntosh and Clark 2008) and the research provides students ample opportunity to examine how archaeological materials can connect with historical documentation and community interests. From a technical standpoint, the Yates property offers an important setting for the field techniques class in that it is generally possible to excavate, over the course of one month, to sterile soil, providing a full stratigraphic accounting for a single trench. YCAP (and now CARI) have a commitment to explore the many facets of Freedman s Town in Houston s Fourth Ward. The work they carry out involves historical and archaeological research, as well as many forms of community involvement and action (McDavid and Bruner 2004; McDavid, Bruner and Marcom 2008; see also Feit and Jones 2007). Their research interests focus on the economic and social lives of the residents of Freedman s Town, with an aim to connect this local community to more 1

5 national concerns. As one of the few surviving communities that were built by formerly enslaved peoples, Freedman s Town provides a unique archaeological, historical and community research context. Through archaeological research, YCAP/CARI hope to use the abundant historic artifacts recovered in excavations to understand consumer choices that were made in the past; these will help to understand how this community became integrated socially and economically with Houston while remaining a largely independent community. The community aspects of this research have involved the rapid gentrification of the Fourth Ward, and how that has impacted long-standing residents and the historic fabric of the community. This work has involved the collection of oral histories and historical research, in the hopes of documenting a rapidly vanishing historic community. A more complete history of research can be found in the 2008 report on the Yates/Rice University collaboration (McIntosh and Clark 2008:4-14). In 2009, excavations were carried out in the back lot of the Rutherford B. H. Yates museum, with a single 2 x 2 meter unit (YC). Decisions about unit placement and excavation techniques were carried out in consultation with YCAP and, as in the previous season, we followed the excavation and analysis protocols established by them. The unit was situated in an area that was believed to have contained an outbuilding related to the original Yates house. Based on insurance plats, we attempted to locate the unit close to the southeast corner of the structure, in an effort to expose deposits in and outside the structure. It was also hoped that this unit will increase information about the two historically-noted building episodes at the site, and the more recent uses of the current house s back lot. The Report As part of their coursework, each student was assigned a particular class of archaeological data, and asked to analyze, record and interpret patterns within these data. The goal is the creation of an excavation report that summarizes thoroughly the material recovered during the research. The results are found in the following chapters. Two undergraduate students Brittany Wise and Sarah Nouri took the class in fulfillment of their capstone course requirement at Rice, and thus carried out more extensive research projects. Brittany examined body adornments from the current excavations but also from 2

6 previous work at the Yates house site. Sarah carried out a soil chemistry project on another of the Yates Museum s historic properties, 1512 Wilson Street. Both of these projects are included as chapters of this report, in their entirety. References Feit, R, and B.M. Jones 2007 "A Lotta People Have Histories Here ": History and Archaeology in Houston's Vanishing Freedmen's Town. Archaeology Report no. 184, Hicks & Company, Houston. Unpublished report. McDavid, Carol and David Bruner 2004 The Yates Community Archaeology Program: Proposal, with Detailed Operating Plan, Phase One Investigation. unpublished. McDavid, Carol, David Bruner, and Robert Marcom 2008 Urban Archeology and the Pressures of Gentrification: Claiming, Naming, and Negotiating Freedom in Freedmen s Town, Houston. Bulletin of the Texas Archaeological Society 79:

7 4

8 Excavation Methods and Excavated Levels Elizabeth Johnson The Rice University Archaeological Field Techniques class (ANTH 362) began excavating at the Rutherford B. H. Yates Museum (site designation 41HR980) on January 31, The excavations took place over a course of four weekends, with the students excavating from 8 A.M. to 5 P.M. on Saturdays and from 12 P.M. to 4 P.M. on Sundays. The class excavated one 2 x 2m unit Unit YC in the backyard of the Yates House. Excavations focused on the area where a structure had been indicated on the 1924 Sanborn map; it was not clear whether this structure was an outbuilding or a privy. According to the Sanborn map, the outhouse was located in the center of the most northern boundary of the backyard. In locating the unit, we sought to avoid areas with excessive roots (near trees) and areas with fire ant mounds. The unit was ultimately located on what was believed to have been the eastern wall of the structure. We plotted this unit using rebar that marked previously-excavated units. Tapes were pulled from these points to locate the southern two points of unit YC; the northern two corner points were triangulated from these. String was then used to mark off the unit and then another set of stakes were placed outside of the designated unit. These stakes were then marked off with string as well in order to keep people from getting too close to the walls. The SE corner of Unit YC was the highest point of the unit and became the Datum Point, which served as the Point-of-Origin (P.O.). All depth measurements were taken from this P.O. The unit was excavated in arbitrary 5 cm level to a depth of 65 cm, at which point only one artifact was found. Excavation Methodology Excavation began with the removal of the grass and root layer over the entire unit area. Since the unit was on an incline, parts of it were already below 5 cm at this point and thus already into Level 2. Only the higher, eastern half of the unit was thus excavated as Level 1. After Level 1 was excavated, the unit was divided into four subunits (1 NW, 2 NE, 3 SW, 4 SE). All subunits were maintained until Level 6, 5

9 at which point only Subunits 1 and 2 were excavated to sterile soil; Subunits 3 and 4 remained unexcavated. The decision to stop excavating the southern Subunits was made by Dr. Fleisher, in the interests of time. We continued to excavate Subunits 1 and 2 in arbitrary 5 cm levels through Level 12; Level, 11, however, was excavated as a 10 cm level. When work was occurring in all four subunits, four or five students excavated one per subunit while one other student supervised. The supervisor recorded all of the data from the unit on excavation forms. The other students screened the deposits. After Level 1, when subunits were established, most of the recording was done on Subunit forms, and at the end of each level the supervisor would summarize the subunit information onto one comprehensive unit form. The forms included the opening and closing depth measurements of the subunit or unit (measured in each corner and in the center), the soil color (Munsell Color chart), artifacts found, and the excavation technique, as well as any observations noted by excavators or supervisor. On the unit forms scales maps were drawn to include roots, changes in soil, and artifacts of interest. Each level was approximately 5 cm deep. Flat-bottomed shovels were used to remove the grass and root mat in a process referred to as schnitting, and then trowels were used for excavating the levels. When an animal skeleton was found in Subunit 1, bamboo sticks, toothbrushes and spoons were used. Since Unit YC was located next to a tree, roots were often a problem. The roots were often small enough to be cut with clippers, but some were cut out with a handsaw. The excavated deposits were collected in buckets, and then screened through ¼ screens. The screeners generally focused on soils from one subunit in order to avoid mixing up the deposits. Initially artifacts were placed in Ziploc bags and labeled with the site, the level, the depth of the level, the date, and the excavator. However, as the soil became wetter at deeper depths, the artifacts were placed in paper bags so they might dry. The artifact bags were assigned Lot numbers and then washed and catalogued in the Rice University Archaeology Laboratory. The categories of artifacts included glass, ceramics, nail metal, non-nail metal, bone, personal adornments, and miscellaneous. The miscellaneous category includes building materials, foil, paper materials, and other modern items. Photographs of the unit were taken at the beginning and end of each level. 6

10 When the four weekends came to a close on February 22, 2009, the site was cleaned but the profiles had not been drawn. A group of three students came back out to the site on March 5, 2009 to begin the drawings. The group returned on March 16, 2009 to finish the profile drawings. Trowels were used to clean and cut the walls. To draw the strata, two measuring tapes were used. One was placed along the top of the wall and the other used to measure down to the different strata. The measurements were taken every 20 cm and plotted onto graph paper. The artifacts left in the walls were drawn into the profiles to scale. A P.O. depth measurement was then taken for each profile drawing. The weather was sunny and nice every day except on February 21, On that afternoon it rained and the buckets full of excavation deposits had to be quickly moved underneath the porch to keep them from getting wet. The unit was covered with a tarp before the rain came to prevent water from getting inside. When the rain subsided, excavating continued. Excavated Levels The stratigraphy for Unit YC is described in this section, including the soil and artifact content. Table 1 provides a summary of Munsell soil colors and artifacts located in each Level. There are also four section drawings, which show the natural layers of the soil for the four walls of the unit, as well as, in dashed lines, the excavated arbitrary levels (Figs. 1 and 2). These figures are followed by a key describing the types of soil found in each natural level. The soil in Level 1 was dark gray. It began under the root mat, which was removed by flat-bottomed shovels. The deposits graded downward slightly to the west, so Level 1 comprised only the eastern half of Unit YC. The artifacts located in this level include modern human trash, rusted metal (including nails, stakes, and nuts), concrete, oyster shell, glass, as well as a single glass bead. Oyster shell was very dense in the northeast and southeast corners of the unit. The artifacts were unearthed using trowels. This level clearly contains mostly modern materials. Because of the slight western slant of ground surface at Unit YC the western part of Level 2 had already been exposed from the grass and root removal. Therefore, many roots remained from the grass in this level. The western half of the unit contained a 7

11 sandy clay loam. In Subunit 1, the soil was very dark gray in color but it was much darker in Subunit 3, where it was close to black. The color of the soil was consistently very dark grayish brown throughout the eastern half of the unit. The soils were a sandy clay loam, predominately made up of sand that crumbled very easily. There was a significant increase in the number of artifacts found in Level 2 over Level 1. Artifacts recovered included metal (nails, and a bottle cap); glass shards (clear, amber, blue, some pieces colored green on one side and white on the other); oyster shell (mainly on the eastern side); brick fragment (one handmade fragment almost fully intact); and bone (a cut piece as well as bone fragments). Other artifacts include wall plaster, a small car part with the Ford logo, a rubber band, roofing tar material, a large sewer-piping fragment, a carbon rod, a shingle fragment, chert, and charcoal. The soil in Level 3 was highly variable, with a different Munsell soil color for each subunit. Subunit 1 was very dark grayish brown in color, Subunit 2 was dark gray, Subunit 3 was dark brown and Subunit 4 was dark grayish brown. Though there were slight color differences throughout the level the soil texture remained consistent throughout the entire level, a loose sandy soil (although some clay was found in the soil matrix in Subunit 3). Rootlets were still found throughout all of Level 3. Large bricks appeared between Subunits 1 and 3 jutting out from the western wall of the unit. There continued to be a high concentration of oyster shell and building material found in Subunits 2 and 4. The artifacts found in this level were located throughout the subunits. These include glass and ceramics (terra cotta and white earthenware); metal (mainly nails); bone; asphalt; plastic; and a glass bottleneck. A U.S. penny found in subunit 3 dates this level to A decaying baseball was located in subunit 3. There was an increased amount of building material found along the eastern wall of the unit. As in Level 3, the soil found throughout Level 4 varied in both color and texture. The soil in Subunit 1 was very dark grayish brown in color. It was a dry, crumbly sandy loam containing rootlets, but far fewer than in previous levels. A silty loam occupied Subunit 2. It was grayish brown in color and contained light, sandy clay patches. The soil in this Subunit was noticeably wetter than the soil found in Subunit 1. The soil in Subunit 3 was dark grayish brown in color and consisted of a sandy silty loam. Light sandy clay was found in the northwest corner and dark loam was found along the 8

12 southside of the subunit. Light yellowish brown clay was found in Subunit 4. The clay was thick and contained some sand. A large flowerpot fragment was found in Subunit 1 and larger-sized building material larger than in previous levels was found. Other artifacts include large fragments of glass laying flat in the NW corner of Subunit 1 as well as a large volume of eggshell. Nails, unidentified metal objects, etched and curved glass, a mother of pearl buckle, a metal button, yellow string, a typeset spacer, a zipper, a plastic toy belt buckle, a bottle cap, a baseball, and metal wire were found throughout the entire unit. The soil in Level 5 was a very dark grayish brown with lighter patches found in Subunit 2. However, the soil from Subunit 4 was a light yellowish brown. The soil in Subunits 1 through 3 were a sandy silty loam but in Subunit 4 it was a thick sandy clay resting on top of a darker clay. The artifacts in this level remained consistent with types and amounts in previous other levels. They include: glass and ceramic fragments, rusted metal, charcoal, nails, a Buffalo nickel, the bottom of a milk bottle, a mason jar lid, half of a marble, pearl buttons, a yellow ball, linoleum, bone fragments, a razor blade, and a metal hinge. In Subunit 2 a beige woman s slip with lace was recovered. The soil in Level 6 was consistent throughout: a very dark gray to a very dark grayish brown. It was very moist clay in all the subunits. Artifacts found throughout the entire unit consisted of charcoal and a large amount of bone; fragments of shell, common in earlier levels, decreased significantly. In Subunit 1 a hanger and a small metal badge which read Melvin Purvis Junior G-Man Corps were found. The remaining artifacts include nails, glass, ceramic pieces, a pen cap, linoleum fragments, a mother of pearl cufflink, iron, and a large piece of rubber. From Level 7 to Level 12 only subunits 1 and 2 were excavated. The soil found in Level 7 was very dark gray clay loam. Large fragments of charcoal were found in the SE corner of Subunit 1 along with iron pieces, unmodified coal, and a button. In Subunit 2 many nails were located, along with a miscellaneous iron artifact, porcelain fragments, a whiteware handle, pieces of eggshell, bone, half of a light blue glass bead, coal and a safety pin. The soil in Level 8 was a wet sandy clay loam; some of the wetness likely came from rain on the previous day. It was very dark gray in color. The artifacts found in 9

13 Subunit 1 include wire, nails, a milk glass shard, a pipe fragment, charcoal, glass, a button, ceramic shards (one piece with the maker s mark), and bone. The bone was exposed with the use of a trowel but the method of excavation was switched to bamboo sticks and brushes when an animal skeleton began to be exposed. The bones contained a curved vertebra, ribs, and two legs, which were photographed. The artifacts unearthed in Subunit 2 consisted of metal (including a white piece), glass, more bone fragments, charcoal, rusted nails, and a child s badge. Some materials from Subunit 2 were accidentally placed in a Subunit 1 artifact bag, but these were removed. The soil in Level 9 was consistent between both subunits; it was a very dark gray clay loam with sand. The animal skeleton located in Level 8 was still visible in Subunit 1 in this level, and the soil around the skeleton was kept for flotation. Artifacts found in Level 9 included bone, nails, ceramic, glass, charcoal, some rusted metal, a metal spoon, a tooth, a rusted cast iron hinge, and a piece of a white cup. Two flat bricks were found below the skeleton in Subunit 1. The soil in Level 10 was clay with a higher sand content then the levels above it. The soil was very dark grayish brown in color. Artifacts included rusted metal (including rusted nails), glass, two turn-of-the-century molded bricks, and cut bone. The number of artifacts in Level 10 was significantly less than in previous levels, suggesting that the excavations were nearing the bottom of cultural deposits. Initially the soil found in Level 11 was very clayey with a high sand content, and very dark gray in color. As Level 11 came to a close, the soil was entirely clay with a hint of green. This soil consistency was affected somewhat by rain on the day of excavation. Very few artifacts were found in Level 11: some rusted metal (mostly nails but some larger pieces as well), along with pieces of bone, glass, and brick fragments. Soil in Level 12 was a silty clay, very dark gray in color. The soil color was green with what appear to be rust streaks, even though few metal artifacts were found. There were very few artifacts found at all in this level. Those that were included charcoal and some metal fragments in Subunit 1. In Subunit 2 a railroad spike was unearthed at 67 cm. When the spike was found on what was thought to be sterile soil, a test pit, 25 x 25 cm was dug in the NW corner of Subunit 2 by Robert Marcom. Eggshell fragments were 10

14 discovered in the test pit between 78 and 83cm. Because this level was largely sterile, excavations were ended. In general, the first few levels consisted of a large number of small artifacts, including a large volume of oyster shell and building debris. Rootlets once common in the upper levels disappeared after the sandy levels ended in Level 4. Subunits 1 and 2 were the only ones excavated after Level 6 due to time constraints. The volume of artifacts remained high until Level 9 when the larger artifacts were encountered, yet fewer in number. Relation of Excavated Levels to Stratigraphy North Profile East Profile 11

15 West Profile Cement Brick Tree root Figure 1: Drawings of natural strata and excavated levels for the North, East and West walls of Unit C. P.O. is the point of origin. The horizontal lines for excavated levels (dashed) are extended for clarity. The numbers on the right side are the excavated level numbers, while the letters represent the types of soil (see below for key). The numbers on the top and left are distance in centimeters. On the profiles the hatched areas are areas that were unexcavated. The area is large on the East profile because these were the subunits that were left unexcavated after level 6. On the North profile the area dug into the hatching is the test pit, which was dug in subunit 2. On the West profile the vertical dashed line in the hatched area represents the line between the subunits, which could not be excavated because of the cement block. Key for the different types of deposits in the North, East and West profiles Label Description A Very dark gray; sandy clay loam, a substantial number of artifacts B Very dark brown; sandy clay loam; the number of artifacts remains consistent C Brown sand (North Profile); yellowish brown sandy clay loam (East Profile); Oyster shell level in North Profile; many artifacts D Very dark grayish brown; sandy loam; many artifacts E Dark yellowish brown (North Profile); very dark gray (East Profile); sandy clay loam; many artifacts F G H Black silty clay (North Profile); very dark gray; silty clay; artifacts Very dark gray; silty clay; fewer artifacts Very dark grayish brown; sandy clay; areas of compacted clay; railroad spike found 12

16 South Profile Figure 2: Drawings of both natural strata and excavated levels for the South wall of Unit C. P.O. is the point of origin. The horizontal lines for excavated levels (dashed) are extended out for clarity. The numbers on the right side are the excavated level numbers, while the letters represent the types of soil (see below for key). The numbers on the top and left are distance in centimeters. Label A B C D E F G Key for the different types of deposits in the South profile Description Black; sandy clay loam; large number of artifacts Very dark grayish brown; sandy clay loam; some artifacts Brown; sandy loam; many artifacts Very dark grayish brown; sandy clay; few artifacts Dark yellowish brown; sandy clay loam; few artifacts Yellowish brown; sandy loam; few artifacts Black; clay loam; many artifacts 13

17 Table 1: Depth measurements; Munsell values, artifacts, and general observations for each level. L is level, O.D. is opening depth and C.D is closing depth. Opening and closing depths are measured in centimeters from the point of origin. L O.D. C.D. Munsell Values YR 4/1 (dark gray) YR 3/2 (very dark grayish brown) YR 3/3 (dark brown) YR 4/2 (dark grayish brown) YR 4.2 (dark grayish brown) YR 3/2 (very dark grayish brown) YR 3/1 (very dark gray) YR 3/1 (very dark gray) YR 3/2 (very dark grayish brown) YR 3/2 (very dark grayish brown) YR 3/1 (very dark gray) Y 3/1 (very dark gray) Artifact Observations Very dense amounts of oyster shell, mostly recent human trash Mostly clear glass, shell (mainly on the east side) Plastic; ceramic fragments; metal; glass; nails A decaying baseball; mostly ceramics, glass, nails, and bricks Greater density of bone; low density of shell and building material Even greater density of bone; charcoal; decrease in shell Large chunks of charcoal; large amount of nails Skeleton is exposed in NW corner Skeleton is removed; two flat bricks found below skeleton Significantly less artifacts; some glass and nails Fewer artifacts; some glass, bone, and metal Very few artifacts; a rusted railroad spike is found General Observations Layer directly under the root mat Many roots; very rocky on the NE side Loose soft, sandy clay in SW wall; stretches into SE wall The soil in the SE corner is light yellowish brown and thick clay The soil in the SE corner is again a light yellowish brown and thick sandy clay Feature 1: 3 stacked cement blocks in Western wall The Southern half of the unit is no longer being excavated Sand is emerging in the NW corner The clay can t be sifted Rusted piece of metal began to emerge, appears to be in level 11 The soil is becoming sterile A test pit was dug after the railroad spike was found 14

18 Ceramics Abidemi Babatunde Babalola Introduction This chapter provides an analysis of ceramics excavated from Unit YC during the 2009 excavation at the Yates Museum. Ceramics from all levels and subunits were analyzed and weighed in sub-material categories and artifact classes. Additional quantitative and qualitative observations/variables (beyond the Yates cataloging system) are also included as part of the analysis (see Appendix 1). This chapter also attempts a brief interpretation of the ceramics, focusing on past consumption patterns and site chronology. Methods The ceramics analyzed here are part of the materials excavated from the Rutherford B. H. Yates Museum yard by the Department of Anthropology, Rice University in The ceramics came from Levels 1 11 of a 2 x 2 meter unit (YC). Subunits 3 and 4 were only dug to level 6 (i.e. 30 cm below the surface). All ceramics recovered from the excavation were catalogued using the Yates Museum cataloguing and coding system. Based on the Yates material-based ceramic categories, only five categories were present in the assemblage: porcelain, stoneware, earthenware, china, and fired clay. A major factor in delineating different ceramic classes is the intensity of heat during the production process. Porcelain endures the greatest heat: it is glassy in nature and translucent when put under daylight, as well as a fine and fully vitrified ceramic. China follows porcelain; china is not as glassy as porcelain and is less vitrified. Stoneware is also well fired, but not to the degree of porcelain and china. Stoneware has fine paste with a low degree of porosity. Stoneware is distinguished from earthenware by differences in porosity; earthenware has a high degree of porosity while stonewares are intermediate between porcelain and earthenware (Wise and Zimet 2008: 23). This difference is most noticeable when a fresh break on a ceramic is tested with a tongue, and the ceramic sticks to the tongue. This process helped in differentiating earthenwares from 15

19 stonewares. No specific criterion is used to identify fired clay since this category was comprised entirely of brick fragments. Earthenware was further classified into creamware, pearlware, and whiteware using Miller s (1980:2) classificatory model, which was based on ceramic paste and glaze. Chronologically, creamware is the earliest, dating to the second half of the 18 th century. Characteristically, creamware has a heavier body, off-white paste and yellow glaze. Chronologically, pearlware follows creamware, which results from the modification or improvement of creamware with the addition of cobalt blue glaze, which is most visible around the rim. According to Terance (n.d.), it is this cobalt blue glaze over a whitish body which is regarded as the single most important diagnostic feature in identifying pearlware. Whiteware is the temporally latest ceramic type. Although the line between pearlware and whiteware is somewhat unclear, a transformed paste and glaze result in a cleaner ceramic. The cobalt blue is greatly diminished on whiteware; however, a faint blue tint was apparent in areas where the glaze is thicker. The overall idea in the production of whiteware was to produce ceramics with whiter bodies and colorless glazes ( The visible characteristics of these three types of earthenware were taken into consideration during the analysis. Since most of the ceramics are earthenware, they were also grouped into vessel and non-vessel categories. This allows for an understanding of the vessel that the sherds represent. In doing this, I defined vessel as any ceramic made or used to contain substance, either for storage or consumption. Therefore, plate, bowl, cup, and flower pots were classified as vessels, while tiles, water/sewage pipes, bricks, and toys fell under non-vessel. Finally, decorative motif was also considered for the analysis. A number of the ceramics are decorated, most with transfer prints and glazes. Other decorative techniques such as over- and under-glaze were also identified. Data Presentation Ceramics occur in all Levels (1 11). Only five categories of sub-materials were represented throughout the unit. These are porcelain, earthenware, stoneware, china, and fired clay. Overall, Level 5 had the highest frequency of ceramics, closely followed by Level 6. The number of ceramics drops considerably in Level 7 compare to Levels 5 and 16

20 6. This is likely as a result of the fact that we stopped excavation in two subunits (three and four). Levels 1, 11, and 10 yielded the lowest counts of ceramics. In terms of submaterial frequencies, earthenware was the most prominent, with the highest frequencies in Levels 5 and 6, and followed by Levels 7 and 8. The frequency of ceramics in all the levels is not adequate to suggest a break in the sequence. In other words, no distinct break was observed in the occupation of the site (Figure 1). The occurrence of fewer ceramics in Levels 1 and 10 is similar to the situation in the ceramics assemblage excavated in Unit P in Like the Unit P excavations, the highest number of ceramics was found in Levels 5, 6, and 7. However, Level 7 in Unit P yielded considerably more ceramics compared to Level 7 in Unit YC. Also the lack of artifacts other than porcelain and earthenware in Level 6 in Unit P has been suggested to be a possible archaeological break (Wise and Zimet 2008), although this need further verification in comparison with other units. However, no break was noticed in the ceramic assemblage from Unit YC. As seen in Figure 2, among the categories of earthenware, whiteware was the most common class found, occurring in all levels (except Level 1). Creamware was found in Level 2, but occurred mostly in lower levels, from 5 to 10, although creamware counts decrease in earlier levels. Like creamware, pearlware was present in Level 2 and then again in Levels 6 to 11. The distribution of earthenware types from Unit YC is quite opposite the situation in Unit P. In the 2008 analysis, pearlware had the highest counts, occurring in almost all the levels except in Levels 1-3 and 10, where none of the three types were found. After pearlware, whiteware appeared in almost all levels except 5, but in very low numbers compared to pearlware. Some earthenwares in the assemblage are decorated, including print and glaze decorations, alone and in multiples, located both overglaze and underglaze. Decorated earthernwares are represented across the levels. Of interest is a plate fragment found in Level 8 with the print of a windmill in stippled blue, and one sherd with an image of two women in Chinese attire in Level 6. Others show flowers or simple lines in various colors (Figure 3). Varieties of ceramics with transfer prints were also recorded in Unit P (Wise and Zimet 2008:25-27). 17

21 Most earthenwares were from vessels, with only few from non-vessels. Nonvessel ceramics, as defined above, occurred in low numbers in Levels 2 through 8 only (Figure 4). The classification of ceramics into their vessel part served as a check as to whether the depositional processes at the site followed the expected pattern in archaeological assemblages. The analysis here shows that bodysherds were most prevalent, followed by rims, bases, and handles respectively (Figure 5). This suggests a normal deposition of ceramics, as one would expect to find larger numbers of body sherds than more diagnostic rims or bases. By looking at the presence of handles as well as determining rim orientation, body type, and base morphology, it was determined that most of the vessels were tablewares or teawares. Tablewares are items used in the preparation, serving, or consumption of food or for drinking at the table (e.g. bowls and plates, cups, saucers, dishes; Herskovitz 1978:96-107). Teawares, on the other hand, are ceramics used specifically for serving tea, including teapots, teacups, saucers, sugar bowls, slop dishes, cream jugs, and tea caddies ( Fragments of some tablewares are presented in Figure 6. Of particular interest among the tablewares are two fragments of shell-edged wares found in Levels 6 and 8. Shell-edged ware is important because it is a good chronological indicator (Figure 7). Three artifacts have makers marks. One has a broken part of a crown in blue transfer print at the base of the plate fragment and was found in Level 8; one earthenware sherd with a transfer print of a windmill on the inside has a mark on the outside of the plate. However, these marks are fragmented in such as a way that it is impossible to determine their makers. For example, the mark with the number 36 could indicate a variety of numbers (e.g. 1736, 1836, 1936 or No. 36). One other sherd was found with a floral print on the base; research on this figure did not reveal a particular maker or date (Figure 8). Discussion and Conclusion Ceramics are often the most common material category in archaeological assemblages. The survival of ceramics is due to their inorganic nature, and their ability to withstand physical or mechanical stress over a long period of time. The attributes of ceramics, such as form, paste, glaze, and other decorative motifs, have been the object of 18

22 studies by archaeologists seeking to understand not only the life of the people that used the artifacts but also the circumstances of their manufacture and how they have come to be accepted in the market (Collard 1979, Miller 1980). It is apparent from the analysis of the excavated ceramics that although other ceramics types such as stoneware, fired clay, china and porcelain were also used, earthenware was the most commonly used and discarded (frequencies have been presented in the section above, see also Appendix 1). The presence of the other ceramics, particularly fragments of water/sewage pipe and tile have their significance in the lifeways of the people. These suggest domestic repairs and perhaps household construction. This assumption could be validated by the early experience of the freed slave in the 19 th century. A related example might be that of brick manufacture and the construction of streets in Freedman s Town (McDavid et al. 2008:43). The presence of brick fragments could as well suggest repairs and constructions by the community at large. Feit and Jones (2007) argue that residents of Freemen s Town did not only engage in yard and household maintenance, but also collaborative community infrastructure maintenance. The idea of community development and collaboration of the ward s residents is also echoed in historical literature and biographies of area residents (e.g. Yates and Yates 1985). Herskovitz s (1978:96) studies at Fort Bowie have classified housewares into kitchenware and tableware. The majority of the analyzed ceramics from Unit YC were housewares, and most of these were tablewares. These ceramics were used in the preparation and serving of food. Further classification into kitchen or teaware was not possible because of the fragmentary nature of most of the ceramics, although it is certain that many the ceramics represented tablewares. Another form of ceramic vessels found in the assemblage, flower pots (Figure 9), suggests specific aesthetic sensibility (Feit and Joane 2007:165). Flowerpots are more common in African American households than in those of white residents (Feit and Jones 2007:166). Apart from the fragments of flowerpot vessels found, some of it was also painted. Previous excavations in Freemen s Town have yielded similar flower pot fragments and these were interpreted to mean that residents at the site may have grown plants not only for decoration as mentioned above, but also for food. The possibility of 19

23 household boundaries and courtyard space was also suggested (Feit and Jones 2007:166). This explanation for flowerpots might not be obvious in the Unit YC assemblage, due to the fragmentary nature of the artifacts. However, the presence of flowerpots sherds may indicate some of the possible activities just described. One interesting artifact was a lamp found resting at the base of Level 6 (Figure 10). The electric lamp was made of multicolored earthenware ceramic. Beyond its functional use, this lamp might be reflective of value attached to aesthetic objects for beautification or decoration, similar to decorated flower pots. It has been demonstrated that creamware is the oldest earthenware type represented at the site, followed by pearlware and whiteware (Hughes 1961, Collard 1979, Miller 1980). Shell-edged ware was first produced during the late 18 th century and became popular after the second half of the 19 th centuries. Shell-edging was generally found on plates (Hunter and Miller 1994). Shell-edged wares are characterized by molded/shell-like or scallop rim motifs with blue or green underglaze ( Those found in the assemblage were not molded rims but scalloped with blue underglaze. The presence of the shell-edged ware could mean that the residents (1) had a variety of tableware, and (2) bought cheap and affordable items. The issue of affordability is important in the reconstruction of lifeways of African-American communities in the 19 th through the early 20 th centuries as this enables us to understand the difference in livelihood among different groups of people. However, the fragmentary data from the ceramics cannot provide us with detail information in this regard. The abundance of whiteware in the excavations (and across the site) is a useful chronological indicator. The presence of whiteware throughout the levels suggests 19 th and 20 th century dates. Although some creamwares and pearlwares were present, their small numbers suggests they may have been residual and thus post-date their peak popularity. However, the contemporaneity of earlier ceramics (creameware) and the later ones (whiteware), and the known late 19 th to early 20 th century occupation of site, may suggest that residents were buying older housewares secondhand, or that the material was being curated. These ideas require larger data sets and more precise dating of the deposits. 20

24 In conclusion, the excavated ceramics from Unit YC represents an assemblage related to a post-civil war African-American urban community. In sum, more earthenware was present in the assemblage than other sub-material classes, suggesting that although these consumers were purchasing finished tablewares, they were buying the most affordable varieties. This can also be seen in the dominance of tablewares over teawares. References Collard, Elizabeth 1979 Nineteenth-Century Pottery and Porcelain in Canada. Montreal: McGill University Press Feit, R, and B.M. Jones 2007 "A Lotta People Have Histories Here ": History and Archaeology in Houston's Vanishing Freedmen's Town. Archaeology Report no. 184, Hicks & Company, Houston. Unpublished report. Herskovitz, Robert M Fort Bowie material Culture. Tucson: The University of Arizona Press. PostColonialCeramics.htm Hughes, Bernard 1961 English and Scottish Earthenware. London: Abbey Fine Arts. Hunter, Robert R., Jr. and George L. Miller 1994 English Shell-Edged Earthenwares. Antiques March: McDavid, Carol, David Bruner, and Robert Marcom 2008 Urban Archeology and the Pressures of Gentrification: Claiming, Naming, and Negotiating Freedom in Freedmen s Town, Houston. Bulletin of the Texas Archaeological Society 79: Miller, George L 1980 Classification and Economic Scaling of Nineteenth Century Ceramics. Historical Archaeology, 14:1-40. Trigger, Bruce G Artifacts and Ideas: Essays in Archaeology. New Jersey: Transaction Publisher. 21

25 Wise, Brittany and Zoe Zimet 2008 Ceramics. In McIntosh, S. K and B. Clarke (eds) Report of Archaeological Excavation in Yates Museum, Freemen s Town Houston. Pp Yates, R. B. H. and P. L. Yates 1985 The Life and Efforts of Jake Yates. Houston: Texas Southern University Press. 22

26 Figure 1: Frequencies Distribution of Sub-material Across the Levels Figure 2: Frequencies of the Categories of Earthenware in Levels

27 Figure 3: Whiteware Plate Fragment from level 8 with Transfer Print of Windmill on the inside Figure 4: Frequencies of Vessel and Non-Vessel Ceramics Across the Levels 24

28 Figure 5: Vessel Parts Frequencies Across all Levels Figure 6a: Fragments of Plate base. Note the ridge is the plate pedestal. 25

29 Figure 6b: Whiteware Bowl Fragment from level 8 26

30 Figure 6c: Whiteware Cup Fragments. A. Above three quarter of a cup found in Level 4; B. Almost two-third of a small cup found in level 9. 27

31 Figure 7: Pearlware/shell-edged ware from level 6. The arrow points to the depression of the scallop. Figure 8: The backside of the Ceramic with Windmill showing 36 print. 28

32 Figure 9: Flowerpot with yellow paint on the outside and drain hole on the bottom. Found in Level 5. 29

33 Figure 10: Ceramic Lamp from level 6 30

34 Appendix: Catalogue of Analyzed Glass Artifacts Level Lot Sub- Unit Materials Sub- Material Artifact Class Quality Weight (g) Comments I CH body; 1 rim I CB I CB Brown glaze on the outside and white on the inside I CB Fragment of water/sewage pipe I CC Creamware I CC I CC <1 White ware I CC Orange piece of potted Vessel I CC White ware I CC White glaze with multiple overlain color; Incised groove decoration and orange color on the inside I CC <1 Rim sherd; Pearl ware I CC I CC I CH Rim; potted Vessel, a bit curved with incised groove at the tip Multiple overglaze with strenght and wavy incision: orange on the inside I CH I CC I CC Base of flower pot? With hole at the center Multiple overglaze with strenght and wavy incision: orange on the inside I CC I CC White ware; Rim; Potted Vessel with angular curve I CC I CC Rim fragment I CF I CF I CF < I CH with incised multiple groove I CH <1 31

35 I CH I CC White ware: Potted vessel; Both Rim;One very large with angular curve I CC Large body fragment, 1 small base of potted vessel with Orange glaze I CC Whiteware I CC I CC Whiteware Multiple overglaze with strenght and wavy incision: orange on I CC the inside Multiple overglaze with strenght and wavy incision: orange on the inside I CC I CC Fragment of fool tile I CC Whiteware; half of teacup I CC White ware I CC Whitware; Base of a plate I CH Rectangular ridges I CA One rim; one body I CA <1 Shell-edged ware with red overglaze at the tip of the rim I CB Fragment of water/sewag pipe with bronw glaze on the inside Brown glaze on the inside and outside; water/sewage pipe; 16mm thick I CB I CC Creamware; Base of potted vessel I CC Dark brown paint on the outside 1 large rim with angular curve, 1 body fragment,1 large base with round hole at the center of the; the three are from same potted vessel: they fit together I CC I CC Rim sherd I CC I CC Cream ware I CC White ware I CC Large rim sherd of potted vessel white yellow paint on the 32

36 outside I CC Creamware with brown and overglaze; plat fragment I CC Multiple overglaze with strenght and wavy incision: orange on the inside I CC Rim sherd; white glaze with black line on the tip I CC <1 Rime sherd; white glaze with rusted stain on the outside I CC <1 White ware white blue-black glaze on the inside I CC Whiteware; Rimsherd I CC Cream ware with raised surface Creamware with brown glaze on both the inside and the I CC I CC I CC Creamware; rimsherd outside Multiple overglaze with strenght and wavy incision: orange on the inside Multiple overglaze with strenght and wavy incision: orange on the inside I CC I CC Whiteware I CC Creamware with raised body I CC Potted Vessel; Painted yellow on the outside I CC I CC I CC Whiteware Multiple overglaze with strenght and wavy incision: orange on the inside Multiple overglaze with strenght and wavy incision: orange on the inside I CC I CC Whiteware I CC Creamware; Rim fragment of teacup I CC <1 Whiteware I CC <1 Dark gray glaze on the inside and light gray on the outside I CF <1 Whiteware with flower pattern on the inside I CH I CH

37 I CH Fragment of water/sewage pipe with dark brown glaze on both the inside and the outside I CH < I CA I CA <1 Rim with brown print on the tip I CA Image of two women in Chinese tire on the inside I CA Raised scallop on both I CB Water/sewage pipe with black stain on the inside I CB I CB Creamware I CB I CB <1 Dark Brown glaze I CB fragment of water/sewage pipe with dark brown glaze on the inside and outside I CC Rim sherd with tellow paint on the outside I CC <1 Pearlware; Rimsherd I CC White ware; Two are rim sherds I CC I CC White ware; All rim sherds I CC thick rim fragment with white glaze I CC Pearl/shelledged ware; Rimsherd I CC Whiteware with brown flower pattern on the inside Whiteware with brown decoration pattern on the inside; I CC Rimsherd I CC Creamware I CC Whiteware I CC <1 Creamware; handle/stem? I CC <1 Pearlware I CC White glaze; One tiny rim sherd I CC White ware I CC Multiple overglaze with strenght and wavy incision: orange on the inside 34

38 I CC Multiple overglaze with strenght and wavy incision: orange on the inside I CC Multiple overglaze with strenght and wavy incision: orange on the inside I CC Pearlware I CC Large fragement of potted vessel with Multi-color print on the outside; incise groove and diamond on the outside I CC Whiteware I CC <1 Cream ware; gcreamw glaze on the outside and black on the inside I CC Green glaze on the both sides I CC <1 White glaze; base of a plate I CC Creamware with dark brown glaze on the outside I CC Whiteware I CF Whiteware with green print on the tip of the rim I CF Globular body I CA Base of a plate I CA <1 Tiny fragement of cup handle with white glaze I CA <1 Base of a plate with flower decoration on the outside, probably I CA makers' mark I CA Glaze orange on the outside and white on the inside Rim sherd; brown line on the tip with brown and yellow radiating pattern on the inside I CA I CA Base of aplate I CB Creamware with raised body I CB Creamware with raised body I CB Rim sherd; pearl ware with blue glaze I CB Brown glaze; fragement of water/sewage pipe I CB Fragment of water/sewag pipe; dark brown glaze I CB Base of a plate I CC Rim sherd with round body; white glaze 35

39 I CC Whiteware I CC Whiteware; Rimsherd fragment I CC Creamware; Base of plate I CC <1 Pearlware; Rimsherd I CC Rim sherd; brown glaze I CC Whiteware; handle of teacup I CC Brown Glaze I CC White ware; White glaze: Plate base I CC Whiteware I CF <1 Rim sherd I CH I CA I CB <1 Curved gblobular shape I CB Whiteware I CC I CC Whiteware; Rimsherd I CC Pearlware I CC White glaze with drawing of crown on the inside (the crown is broken half way) I CC Rim sherd; wihte glaze I CC I CC water/sewage fragment Whiteware, white glaze with raised decoration and groove line on the inside Large fragment of suquare/rectangular serving plate; White glaze with brown line and scallop on the inside I CC I CC Whiteware I CC Rim sherd; White Ware; White glaze I CC <1 Pearware I CC Whiteware I CC Whiteware; Rimsherd I CC Creamware; base of aplate with print of wind mil on the inside, part of the makers' on the outside has broke of, only figure 36 36

40 is left (the three pieces fit into one another) I CC White/shell-edged ware with blue glaze on the inside I CF <1 Whiteware I CF Rim fragment of a plate I CA Rim sherd I CA <1 Globular body with stamped circle decorations on the outside I CA <1 9 Nil 2 I CA Nil 2 I CA Handle fragment Blue glaze; one with hollow surface and tiny handle like (Toy?) 9 Nil 2 I CA I CB Base of potted vessel with gray glaze 9 Nil 2 I CB I CC I CC Rim sherd Cream ware; Rim sherd with dark brown glaze on the inside and cream on the outside I CC Nil 2 I CC White ware 9 Nil 2 I CC Rim sherds; White ware 9 Nil 2 I CC Whiteware 9 Nil 2 I CC Handle fragment 9 Nil 2 I CC White ware with tiny blueish decoration almost break of 9 Nil 2 I CC Whiteware; base of a plate I CC <1 Whiteware; Rimsherd fragment I CC Whiteware; handle 9 Nil 2 I CC Whiteware; half of small teacup 9 Nil 2 I CF Rimsherd with gold print on the tip I CF Nil 2 I CF Toy? I CA Nil 1 I CC Whiteware 37

41 10 Nil 1 I CC Cream ware; Light brown glaze 10 Nil 1 I CC <1 Whiteware I CC Rim sherd I CC Base of serving bowl white white glaze: White ware I CF I CF Pearl ware with raised decoration on the outside I CB Orange glaze on the outside I CC <1 11 Nil 1 I CC Whiteware 11 Nil 1 I CC <1 Pearlware; Rimsherd I CC <1 Pearlware; Rimsherd I CF < I CF Table 1: Cataloguing table: The table shows the qualitative and the quantitative attributes of the artifacts analyzed 38

42 Glass Artifacts Courtney Ng Introduction The excavations which occurred at site 41HR980, the historic home of Rutherford B.H. Yates in Freedmen s Town, Houston, yielded a large amount of glass artifacts. These artifacts ranged from flat plate glass to curved bottle glass to tableware rims. The vast majority of the artifacts were fragmented and therefore largely unidentifiable. In addition, formation processes and human activity make a concrete dating of the levels through an analysis of glass artifacts alone nearly impossible. Some artifacts, however, could be identified and dated (using maker s marks or other features), and these dates are discussed to give a general sense of dating of Unit YC. Although the purpose of this report is to offer some insight into the use of material culture in daily life at the Rutherford B. H. Yates site, it is dangerous to assume that artifacts which we date and identify were used around the time of their production and for the purpose for which they were originally produced. This is why we must treat terminal dates with care. We can always have some indication of a starting date for a technique if we can find who first put the idea into practice. But any technique, once developed, can be used right up to the present (Toulouse 1969b) The same idea persists for the function of an artifact. A soda bottle, for example, may have been emptied and used for another purpose, which we may never know based on the artifact s qualities alone. Given all this, it is important to recognize the value in searching for trends, patterns, and anomalies in the data. This kind of information can, along with historical documents, give us a clearer sense of what life was like at the Yates home in the early 20 th century. Through the analysis of glass artifacts, I hope to make some sense of the ways in which these artifacts reinforce or challenge perceptions about the Yates home or its inhabitants. 39

43 History of Glass Production As the site of the current Yates museum was occupied starting in 1913 by Rutherford B. H. Yates and his family, the brief history of trends in glass production contained in this section will look at production techniques as they evolved in the 20 th century. To begin, it is important to note that the second half of the 19 th century through the first quarter of the 20 th century was a time of revolutionary changes in bottle production. This period saw the phasing out of mouth-blown bottles and the introduction, in 1905, of the Owens Automatic Bottle Machine, which made mouth-blown bottles essentially obsolete by 1920 (Lindsey 2009). Therefore, during this period, the most identifiable differences can be seen between mouth-blown and machine-made bottles. Mouth-blown bottles are easily distinguished by two key characteristics: the presence of a pontil scar and the nature of the bottle s seam. A pontil scar is a circular irregularity on the base of a bottle that is a result of the bottle being held on a metal rod (or sometimes the blowpipe itself) as it was molded (Figures 1a and 1b). When the molding process was complete, the bottle would be broken off of the rod, leaving behind a scar (Toulouse 1968). When bottles were produced via machine, there was no longer a need to puncture the bottle s base, therefore eliminating pontil scars in machine-made bottles. Another kind of scar known as a suction scar (Figure 1c) can often be confused with pontil scars, but are the result of the Owens Automatic Bottle Machine and other blow-and-blow machines which funnel glass through molds using a suction process (Lindsey 2009). The second identifying feature, the seam, is also a result of the methods by which the bottle was produced. Bottles are left with seams when tiny spaces exist between the two (or in some cases, three) parts of a mold (Figure 1d). Mouth-blown bottles, however, have a distinctive kind of seam that never reaches the very top of the bottle rim. This is caused by the finishing process involved with hand-made bottles, which uses a tool to mold the mouth of the bottle that results in the removal of the seam (Toulouse 1969:533). Machine-made bottles do not require such a tool, therefore the seam created by the mold remains intact from the base of the bottle all the way to the top of its rim. Another characteristic that was altered by the invention of new production techniques was the color of glass itself. While glass color on the whole is not a 40

44 particularly straightforward variable for dating bottles, some distinctions may help in the general dating of levels. The color of glass is determined by the purity of the sand (silica) used in its production, and the amount of iron present in the sand. In sum, the higher the silica content, the clearer the glass. Most glass, though, is not entirely colorless. Glass generally tends to have a tinge which can be any color from purple to yellow. Straw tinged glass is a result of mixing selenium or arsenic into the sand, which was a common practice between 1915 and The precursor to clear tinged glass was aqua-colored glass, which was a result of the natural iron present in sand. Aqua-colored glass was common throughout the 19 th century up until about 1920 in the US, though it was still used in the production of fruit jars throughout the 1930s (Lindsey 2009). Purple, or amethyst, tinged glass is a result of the presence of manganese dioxide that darkens the glass with exposure to UV light. This kind of glass was produced before World War I (Lindsey 2009). There are many more identifying features of glass, but the ones highlighted in this section appear to be most beneficial in helping us understand bottles features and their period of production. Methodology of Cataloguing and Identifying Glass Artifacts The Yates Community Archaeology Project Cataloguing System was used to catalogue all of the glass artifacts uncovered in Unit YC. Each bag of artifacts was catalogued separately, with all the similar fragments weighed together and described in detail. The first variable recorded was glass color. For this analysis, it was important to distinguish between clear glass that had a different color tinge (i.e. green or blue) and glass pieces that were actually green or blue. Next, shape was determined, distinguishing between flat or curved, decorated or plain, a part of tableware or kitchenware. Flat pieces of glass are likely window glass or remnants of flat objects like mirrors. Curved pieces of glass are usually parts of broken bottles or jars. While it was easiest to determine whether a piece was flat or curved, defining the other terms was less clear and straightforward. For instance, pieces of glass that were patinated as a result of the natural chemical processes in the ground may have been confused with pieces of glass that were patinated as part of their production. In this case, all pieces of glass listed as patinated are catalogued as plain and not decorated. 41

45 Decorated pieces of glass are generally any kind of molded glass with designs or writing on them. Determining whether glass pieces were tableware or kitchenware was equally as challenging. As all of the artifacts that were catalogued appeared to be part of plates or large serving bowls, they were listed as tableware. It is possible, however, that they were used in the kitchen as well. Without a maker s mark present to identify those pieces, it is very hard to determine exactly where they were most likely used. For this reason, the distinction between tableware and kitchenware was not particularly useful in classifying these artifacts. In attempting to date the artifacts, it is easiest to start by identifying marks such as writing or maker s marks. As there was a good amount of this kind of evidence present in the artifact assemblage, identification required extensive review of documents detailing maker s marks and distinctive writing on bottles. When maker s marks or diagnostic designs were not present, other clues such as glass color, shape, and marks from production techniques were used in the dating process. While these clues cannot give us exact dates for fragments, they can help us narrow the window of possible periods during which the glass was produced. Level Analysis A total of 12 levels were excavated in Unit YC. However, Subunits 3 and 4 were only excavated down to Level 6 due to time constraints. Exactly 2160 glass artifacts (approx grams) were discovered, supporting the idea that Unit YC contained part of a trash pit where bottles, jars, and other household glass artifacts were deposited. The distribution of glass (as shown in Figure 2) demonstrates a spike in artifacts around Level 4 that decreases, beginning at Level 6. As the majority of glass artifacts were flat and clear, we can assume that several window panes were broken in Unit YC, possibly in the construction or demolition of a building, or through the active life of it. Other types of glass, such as curved and colored glass, make up a greater percentage of the artifacts found in earlier levels (6 through 12) and were remnants of broken bottles. Fragmentation of the artifacts was also considered (Figure 3). By finding the total weight of all the artifacts in each level and dividing it by the quantity of artifacts, we are able to tell the 42

46 average size of a fragment in each level. Knowing this information gives us a sense of whether the artifacts were deposited and broken apart over time, or deposited all at once and covered over (as in a trash pit), allowing them to remain intact. The following levelby-level analysis of glass found in Unit YC offers an in-depth description of these trends. Level 1 was not excavated by subunit, as we attempted to compensate for differences in elevation by removing the top layer, consisting primarily of grass and roots. The first level also had a comparably low density of glass artifacts (c. 160 grams). The vast majority of this glass was flat and clear, with a green tinge. Besides a few pieces of curved amber glass most likely from a bottle, there was one square clear piece of glass that might have been part of a bottle base. The numbers 2 and 10 are embossed on the glass, but this does not seem to provide enough information for us to identify where it actually came from. The average weight of a glass fragment in Level 1 was 1 gram, demonstrating that the fragments were extremely small in this level. There was a significant increase in both the number of glass artifacts found in Level 2 (c. 920 grams) and the size of the average fragment (2 grams). However, the majority of this glass is still clear, flat windowpane glass. When we look at the distribution of plate glass across this level within the subunits, we find that most of the plate glass found in Level 2 was distributed across Subunits 2 and 4, the eastern side of Unit YC (Figure 4a). There is also large amount of curved glass in this level, including two large pieces of molded glass with ridges. These two pieces seem to be from the same bottle and one of the pieces has the remnants of an applied label with the letter S printed in green on the bottle. We know that applied labels were not used in bottle making until 1934, so this bottle definitely post-dates this year. Exactly what kind of bottle it is, though, is indeterminable due to the lack of a base or maker s mark. Another piece of curved glass, probably from a jar, has the words uniformity, use, and bottle, applied in white print on the exterior, but the piece is still unidentifiable. Finally, a piece of clear flat glass has the letters ASO embossed on it, but this information is not enough to tell us where the glass originated from. Level 3 had approximately 600 grams of glass artifacts, almost all of which were clear. The average fragment size was similar to that of level 2, about 2 grams. Most of the glass was also flat, though about a third of it was curved. There was a larger amount of 43

47 bottle pieces (bases, rims and/or necks) in this level than in either of the previous two levels. One of these pieces is a bottle neck, which has seams going up all the way to the top of its rim, indicating that it dates to after 1905, when bottles began to be produced by machine (Lindsey 2009). Several fragments of flat decorated glass with a flower pattern were also found in this level. They may be part of the same piece of tableware. In addition, Level 3 saw the appearance of melted glass and milk glass, which appear more frequently in lower levels. The glass artifact frequency increased again in Level 4, where there were 1300 grams of glass present. Most of the plate glass in this level was concentrated in Subunits 1 and 3, which is a result of the lower elevation of the left side of the unit, where Levels 3 and 4 were at the same elevation as Level 2 of Subunits 2 and 4 (Figure 4b). However, more than 100 grams of the glass in level 4 was curved green glass, demonstrating the increased presence of bottles in the unit. There were also about 35 grams of white and green glass, which is the start of the appearance of fragments from Vessel 1, possibly a serving dish (see Figures 5a and 5b). A green machine-made bottleneck was found, as well as several pieces of molded glass. One piece has the letters ER, DEM, and SO, embossed on its exterior, while another reads MAS and appears to be close to the base of a bottle. In general, the number of pieces of molded glass and the average size of the fragments (3.4 grams) increased dramatically in Level 4, which seems to be the precursor to the heart of a trash pit which appears in Level 5. There were almost 1700 grams of glass found in Level 5, more than in any other level. The diversity of artifacts found was also much greater: more than a third of the glass found was colored glass (see Figure 6). The majority of the remnants of Vessel 1, which first appeared in Level 4, were found in this level. Several bottle pieces were also found, including one intact bottle and at least 3 bases with maker s marks. The intact bottle has a diamond-shaped base with an unidentifiable maker s mark (see Figure 7). One intact bottle neck (Figure 8) and a few partial bottlenecks were present, all with seams to the rim, indicating that they were machine made (Lindsey 2009). Another interesting find was a flat piece of highly patinated glass with metal attached to one side. This piece was most likely part of a small mirror (Figures 9a and 9b). At least three large pieces of molded glass appear to be part of a drinking glass. The base is also present, but 44

48 there are no marks to identify its maker. The wealth of artifacts in this level, along with the fact that many of these artifacts were large pieces of glass (average size 3.9 grams) and were found in a vertical orientation is an indication that this level contained a trash pit where artifacts were deposited and then covered over before they could be trampled into smaller pieces. Level 6 also contained part of the trash pit, as there were several intact pieces found in this level as well (average fragment weight was 3.9 grams). Even though there was a drop in the amount of artifacts found (c. 650 grams), the diversity of the artifacts remained high. Almost half of the glass found was likely bottle glass because it was curved, and two intact bottles were discovered (which accounted for a large portion of the glass weight of level 6, as shown in Figure 10). One of these bottles had an identifiable maker s mark, and is discussed in detail in the next section. The other bottle has interesting triangular shape (Figure 11), but is lacking a maker s mark that would help us determine much more about the bottle. Based on its shape, this bottle was most likely either a medicine or perfume bottle. Another bottle base of a small rectangular bottle, possibly a medicine bottle, was found with a marker s mark, but the factory behind the mark has yet to be identified. A good deal of molded glass was also found in this level, included pieces of glass with bubbles embedded in them. Although air pockets within glass are usually considered an irregularity, in these pieces the bubbles are so large and take up the entire fragment, so we might consider them to be an intended pattern for the glass. Another interesting find was a piece of thick curved glass which has magnifying qualities (see Figure 12). The exact use of this glass, though, is undetermined. The downward trend in artifact density and average size (3.3 grams) continued in Level 7, where there were approximately 260 grams of glass present. Almost all of the glass in this level was from bottles, though there were several unique finds. Another milk glass insert was found, probably from the zinc cap of a mason jar. It was marked Genuine Porcelain, even though it has a transparent quality and is obviously a glass artifact. One piece of clear glass with a blue tinge has an al embossed in script on it, which could match the design of Ball mason jars that used zinc caps with milk glass inserts. A thin piece of glass was found with what appears to be measurement marks that read 80 and 100. This piece of glass could have been part of a measuring cup. One 45

49 small bottle neck was also found; it may be part of a machine-made medicine bottle. Another partial bottle neck looks like it might be from a clear beer bottle. Level 8 contained close to 200 grams of glass, almost all of which was curved glass. Five of the pieces of clear molded glass fit together into what appears to be a small bottle or glass. The exact use of this vessel is unclear. Several other bottle pieces, including rims, bases, and corners of square bottles, were discovered, but their origins are not discernable. This is primarily because the pieces are more fragmented than in previous levels (average weight of 2.5 grams), providing us with fewer identifying clues. Approximately 240 grams of glass (average weight 3.3 grams) were excavated from Level 9. Again, most of this glass was curved. A highly patinated blue-tinged bottle base was found without a maker s mark, as well as some pieces of white and green glass from vessel 1, which was first seen in level 4. An interesting find in this level is a piece of glass that was likely part of a liquor bottle. Discussion of this fragment is located in the next section. Two small bottlenecks were also found, as well as a large piece of bluetinged glass that has the word REGULATOR embossed on it (see Figure 13). It is possible that this piece of glass was part of a liquor bottle. There were 130 grams of glass in Level 10. Most of this glass was colored and curved. Overall, the glass was also much thicker (average weight 3.7 grams), and several pieces had what appears to be white paint on their exteriors. Two bottle rims were found, one with a very large diameter, and the other probably from a beer bottle. Another interesting note is the presence of purple-tinged glass in Level 10, which is indicative of a mouth-blown bottle made before World War 1, or (Lindsey 2009). Of the 64 grams of glass (average weight 3.2 grams) found in Level 11, almost all was curved glass. The amount of molded glass in Levels 10 and 11 were much lower than in higher levels. Colored and tinted glass was also much more common, including purpletinged glass. Level 12 contained only one small piece of curved clear glass, so very little is known about the nature of the level based on glass artifacts. Identifiable Glass Artifacts There were a number of identifiable glass artifacts discovered in Unit YC. These artifacts provide us with valuable knowledge about the material culture of the inhabitants 46

50 of the Yates house. Being able to narrow the time frame in which the artifact was used, as well as how it was used, may give us a clearer sense of how those who used the artifacts were constructing and enforcing their identities through their possessions. At the very least, it tells us about the daily life of those who used the artifacts. The first piece of identifiable glass found in the unit was a fragment of a glass bottle found in Level 2 and was embossed with part of the Coca Cola script logo, as well as the volume of the bottle, 10 fl oz (see Figure 14a). Contoured bottles of these kind were first produced in 1915, but this fragment appears to be from a version that existed starting in 1961 (Figure 14b). A similar fragment was found in Level 4. One of the most easily identifiable fragments was the base of a bottle which was embossed with the writing KERR GLASS MFC CO SAND SPRINGS, OKLA AUG 31, PAT (Figure 15). Comparable bases demonstrate that this was part of a fruit or mason jar. It is likely that this jar was used for a while after it was produced and purchased, as the fragment was discovered in Level 5, which is stratigraphically higher than the levels containing other artifacts that date to the 1930s. This disorder could also be a result of formation processes that pushed this shard up from a deeper level. A zinc cap with a milk glass insert that reads FOR BALL MASON JARS GENUINE ZINC CAP was also discovered in this level (Figure 16a). It is unlikely that this cap came from the bottle with the Kerr base, simply because Ball mason jar refers to a separate company known as Ball (Figure 16b). This doesn t rule out that the cap was associated with the Kerr jar at some point, but we can be sure that the mason jar cap and base found were not originally intended to be used together. Also discovered in Level 5 was one half of an amber beer bottle with a clear maker s mark on its base (see Figure 17). The maker s mark can be traced to the Owens- Illinois Glass factory in Streator, Illinois, which opened in 1930 and is still open today (Lockhart 2004:27). However, the maker s mark itself was only in use from (Toulouse 1971). An analysis of the mark reveals that the number to the right of the mark represents the year the bottle was produced. Since the number to the right of our maker s mark was an 8, we can assume that the bottle was either produced in 1938 or 1948, which fall within the time frame the mark was in use. Further research shows that after 1946, a period was placed after this number (Lockhart 2004:25). As there is no period 47

51 after the 8 on our bottle, we can narrow down the year of the bottle s production to This deduction is dependent upon the accuracy of the changes discussed by Lockhart and Toulouse. Another bottle, found in Level 6, can be dated to an exact year as well. An intact, clear cylinder jar with a screw cap was discovered with two maker s marks, a K in a keystone shape, and a J in a circle (Figures 18a and 18b). After researching the first maker s mark, it was discovered that the mark represents the Knox Glass Bottle Company of Knox, Pennsylvania. The Knox Bottle Company was founded by Roy Underwood in 1917, and the factory in Knox was opened in The second maker s mark on this bottle is interesting, which stands for the Knox Glass Bottle Company of Mississippi. In 1932, Roy Underwood s brother, Chester, opened the Mississippi factory in order to expand the company into the south. In the intervening time before Knox of Jackson, MI came up with a maker s mark of its own, (the J ) it used the molds with the K from the Knox, PA factory. Those bottles were later marked with the Jackson factory s maker s mark. For this reason, bottles with both maker s marks are identified as having been produced in 1932 only (Lockhart 2008:7). In Level 9, a clear glass fragment with the words FEDERAL LAW PROHIBITS RE-USE OF THIS BOTTLE was found (Figure 19a). Though it is rare to be able to identify a fragment, unique conditions allow us to get a general understanding of where this fragment came from. According to the Society for Historical Archaeology, liquor bottles produced after 1935 were required to be labeled with this message (see Figure 19b), as Prohibition had been repealed in 1933 and the government wished to discourage the re-use of bottles by bootleggers and moonshiners (Lindsey 2009). The message was written on liquor bottles until the mid-1960s, thereby allowing us to roughly date the bottle as having been produced between 1935 and the 1960s. Comparison to Unit P Unit P was excavated at the Yates house in the spring of 2008 by the Archaeological Field Techniques class. Nine levels were dug in all four subunits of Unit P. This automatically distinguishes this Unit from this year s unit, since we dug 12 levels in Subunits 1 and 2 and only 6 levels in Subunits 3 and 4. The finds from Unit P were 48

52 also very different from what we discovered in Unit YC. The total weight of all the glass found in Unit P was approximately 2000 grams, about a third of what was discovered in Unit YC. The glass artifact density in Unit P was greatest in Level 7, with the second largest concentration in Level 4. Unit YC, in contrast, had a unimodal distribution, with a peak of glass artifact density at Level 5. The decreased artifact density in Unit P is mostly due to the lack of plate glass which was abundant in Unit YC. The majority of glass found in Unit P was curved bottle glass, which grew more diverse in style and color with depth. The same is true of Unit YC, except this diversity included more intact and identifiable pieces. In fact, none of the artifacts in Unit P could be dated to an exact year or place of production, as opposed to the multiple pieces that were identified in Unit YC. This highlights the main difference between Units YC and P: Unit P lacked the trash pit that was present in Levels 4 through 6 of Unit YC. The trash pit was an invaluable source of identifiable artifacts that mark a cultural layer which was barely visible in Unit P. While there was an increase in artifacts with writing or other identifying features in Level 6 of Unit P, the sample size of such artifacts was significantly smaller than that of Unit YC. Unit P also claims to demonstrate a clear divide between the production of handmade and machine-made bottles, which occurred in the early 20 th century. The divide is evidenced by changes in the color and seam styles of bottles. There is no definitive evidence of this shift in production techniques in Unit YC. Conclusion The analysis of glass artifacts in Unit YC yielded many interesting finds. First and foremost, the abundance of plate glass in Levels 1 through 5 points to the presence of structure which was at one point torn down leaving broken window glass strewn over the area. As a result of differences in elevation across the unit, we find that Subunits 2 and 4 contained the highest concentration of plate glass in Level 2, while the bulk of the plate glass found in Subunits 1 and 3 was not unearthed until later in the excavations, around Level 4. Based on the presence of a Coca Cola bottle that was first distributed in 1961, it is likely that the structure which once existed within or adjacent to Unit YC was torn down in the 1960s. 49

53 Plate glass nearly disappears in Levels 6 through 12, where there is mostly curved and molded glass, presumably from bottles. Three intact vessels were found, two of which could be identified by their maker s marks. Several other artifacts could also be identified and dated using maker s marks or identifying characteristics such as embossed writing. The presence of so many identifiable artifacts, especially concentrated in Levels 4 through 6, provides strong evidence for the presence of a trash pit in those levels which was later covered over. This theory is further strengthened by the relative intactness of the artifacts in these levels, as well as their orientation within the ground (vertical as opposed to horizontal). These factors indicate that the artifacts were not trampled upon by humans, which makes sense if the area was a trash pit for some time. The identifiable artifacts in Levels 5 and 6 dated to the 1930s, indicating that the trash pit was probably in use during that decade. One factor that was considered in dating levels with unidentifiable fragments was the color of tinged-glass. As noted in the History of Glass Production section, the presence of purple, straw, or aqua-tinged glass can sometimes help to determine the time frame in which the glass could have been produced. It was discovered that purple-tinged glass appeared in Levels 10 and 11 most frequently, which is a sign that these two levels date to before Straw-tinged glass, which is supposed to identify glass that was produced between 1915 and 1950, appears consistently in Levels 5 to level 10. While this does not necessarily help us to date the individual levels, it supports the idea that Levels 5 and 6 could have been deposited in the 1930s. Overall, however, these types of glass do not offer a clear diagnostic tool for dating levels, and are often convoluted by the presence of multiple glass colors across levels. Other clues can further complicate these dates and lead us to believe that there were disturbances in the levels over time. For example, Vessel 1, a white and green glass piece of tableware, had fragments present in Levels 4 through 9. The fact that most of the fragments were discovered in Level 5 seems to indicate that that was where the tableware had been deposited. However, the fact that pieces of the vessel are present even as far down as Level 9 demonstrates how formation processes can shift artifacts throughout levels. Therefore, the trends we witnessed, while valuable to consider, should always be viewed with a critical understanding that what enters the ground through human activity 50

54 does not always come out of the ground in the same place, or same shape, as it was deposited. In closing the chapter, it is apparent that there are many questions left unanswered by the data thus far. Putting aside even the multitudes of unidentifiable artifacts excavated from Unit YC, there is also room to dig even deeper into the data in searching for trends and patterns. Studying characteristics such as bottle shapes and tinges could create more certainty in dating the levels. In addition, comparing the kinds of glass artifacts used by the Yates family to larger trends in glass use throughout this time period would offer opportunities to broaden our understanding of how this family differed from other families in their community. References: Jones, Olive 1971 Glass Bottle Push-Ups and Pontil Marks. Lindsey, William 2009 Bottle/Glass Colors. Electronic document, accessed April 14, Lindsey, William 2009 Bottle Body Characteristics and Mold Seams. Electronic document, accessed April 14, Lindsey, William 2009 Bottle Bases: Pontil Marks or Scars. Electronic document, accessed April 14, Lockhart, William 2009 The Dating Game, Electronic document, accessed April 14, Lockhart, William 2009 Knox Glass and the Marks Toulouse Missed. Electronic document, accessed April 12, Lockhart, William, Peter Schulz, Carol Serr, and William Lindsey 2009 The Knox Glass Bottle Co. Electronic document, accessed April 12,

55 Lorrain, Dessamae 1968 An Archaeologist s Guide to Nineteenth Century American Glass. Toulouse, Julian 1968 Empontilling: A History. The Glass Industry, 49. Toulouse, Julian 1969 A Primer on Mold Seams. Electronic document, accessed April 12, Whitten, David 2009 Glass Factory Marks on Bottles. Electronic document, accessed April 12,

56 Figure 1a: The above illustration demonstrates the different ways that a pontil scar can be created in hand-made bottles. Source: Figure 1b: A bottle base with pontil scar. Source: 53

57 Figure 1c: A bottle base with suction scar, often confused with a pontil scar. Source: Figure 1d: A side mold seam that travels up and over the rim of a bottle indicates the bottle was machine made. Source: 54

58 Figure 2: The distribution of glass artifacts across the levels reveals that the greatest density of artifacts existed in Levels 4 and 5. Levels 7 through 12 had low artifact density, but that could simply be a reflection of the fact that only two subunits were being dug down to those levels. Figure 3: The average size of the artifacts in each level gives us a sense of how the artifacts were deposited. The relative intactness of fragments in Levels 5 and 6 are further indication that a trash pit existed in these levels, which protected the artifacts from being broken apart by human activity. The very small size of glass fragments in Level 12 is a reflection of an extremely small sample size in this level. 55

59 Figure 4a: Subunits 2 and 4 with the greatest concentration of plate glass within Level 2. 56

60 Figure 4b: Subunits 1 and 3 had the most plate glass in Level 4 (with large amounts also in Level 3), because the ground slopes upward from right to left across the unit. 57

61 Figure 5a: Exterior view of a base and rim fragment from Vessel 1, a piece of tableware. These fragments were found in Level 5, Subunits 1 and 2, but more fragments of this piece were discovered in Levels 4 9. Figure 5b: Interior view of base and rim fragments from Vessel 1, found in Level 5 Subunits 1 and 2. 58

62 Figure 6: This graph shows that clear glass was the dominant submaterial found, but that there was more colored glass in Levels 7 through 12, and that Level 5 showed the greatest diversity in glass colors. However, the lack of a definitive trend in glass color prevents us from dating the levels with any certainty. 59

63 Figure 7: Intact jar found in Level 5, Subunit 2. The maker s mark is unidentifiable, but it was most likely a food jar. Figure 8: Intact bottleneck found in Level 5, Subunit 2. Note the seam, which continues over the rim of the neck, indicating that this bottle was machine-made. 60

64 Figure 9a: Front view of patinated glass found in Level 5, Subunit 3. This artifact is probably part of a small mirror. Figure 9b: Back view of patinated glass shown above. Attached metal is evidence that this fragment is part of a mirror. 61

65 Figure 10: The artifact class distribution demonstrates that although there was a large amount of flat glass in Unit YC, curved glass was dominated Levels 7 through 12. The whole bottles found in Levels 5 and 6 may appear to skew the data; there were in fact only three intact bottles excavated, but they were large and weighed more than most of the fragments. 62

66 Figure 11: Possible perfume or medicine bottle found in Level 6, Subunit 3. Figure 12: Convex glass artifact which has magnifying qualities, found in Level 6, Subunit 3. 63

67 Figure 13: Highly patinated blue-tinged glass fragment with REGULATOR embossed on exterior. This fragment was found in Level 9, Subunit 1, and is most likely part of a liquor bottle. 64

68 Figure 14a: Two fragments from Coca Cola bottles. The one on the left has 10 FL OZ embossed on the exterior, and was found in Level 2, Subunit 3. The fragment on the right was found in Level 4, Subunit 3. Figure 14b: Coca Cola bottle designs have changed drastically throughout the 20 th century. The two fragments about most likely came from the design that was introduced in Source: nbusinessabroad/bottles.jpg 65

69 Figure 15: Base of a jar found in Level 5, Subunit 1. The embossed writing reads: KERR GLASS MFC SAND SPRING OKL AUG 31, 1915 PAT 3. Further research on the Kerr Glass Company indicates that this base was likely from a fruit or mason jar. 66

70 Figure 16a: Zinc mason jar cap with milk glass insert discovered in Level 5, Subunit 2. The glass has GENUINE ZINC CAP FOR BALL MASON JARS embossed on the exterior. Figure 16b: An example of a Ball mason jar. The zinc cap is very similar to the cap found in Level 5, Subunit 2. Source: 67

71 Figure 17: Base of amber bottle found in Level 5, Subunit 4. The maker s mark informsus that this bottle was produced by the Owens-Illinois Glass Company in

72 Figure 18a: The base of an intact bottle found in Level 6, Subunit 1. The presence of two maker s marks tells us that this jar was produced by the Knox Glass Company of Jackson, MI in Figure 18b: Side view of jar found in Level 6, Subunit 1. This jar most likely contained a condiment such as salad dressing. 69

73 Figure 19a: Glass fragment from Level 9, Subunit 2 with embossed writing that contains part of the message FEDERAL LAW PROHIBITS RE-USE OF THIS BOTTLE. Figure 19b: Example of a liquor bottle with the federal warning embossed on exterior. This message wasembossed on liquor bottles following the end of prohibition in 1935 until the 1960s. Source: 70

74 APPENDIX: Catalogue of Analyzed Glass Artifacts Cataloguing Key: II. Glass Sub-Material: GA. Clear GB. Red GC. Green GD. Blue GE. Amber GF. Brown GG. Black GH. White GI. Violet GJ. Yellow GK. Other Artifact Class: Flat, no backing: plain Flat, no backing: decorated Flat, mirrored backing: plain Flat, mirrored backing: decorated Curved: plain Curved: decorated Bottle Rim: plain Bottle Rim: decorated Bottle Base: plain Bottle Base: decorated Tableware Rim: plain Tableware Rim: decorated Tableware Base: plain Tableware Base: decorated Kitchenware Rim: plain Kitchenware Rim: decorated Kitchenware Base: plain Kitchenware Base: decorated Whole Bottle Melted Fragment Bottle Neck: plain Bottle Neck: decorated 71

75 Lot Level Subunit Submaterial Artifact Class Quantity Weight Comments N/A GA green tinge N/A GA patinated; peach paint? N/A GA N/A GC N/A GF N/A GJ N/A GA molded glass; ridges N/A GA GR GA N/A GD N/A GF N/A GA molded glass; one possible bottle base with square impression N/A GF molded glass; ridges, bumps GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge; patinated GA green tinge; patinated GA green tinge; patinated GA green tinge; patinated GC patinated GD patinated GE patinated GA patinated; green tinge GA GA GA

76 GA GA GA GC GC GI GI GA patinated peach and orange GA green tinge GA green tinge GC one piece darker, with smooth grooves (possibly manmade?) GA one piece- white paint on exterior GA patinated GA red paint on exterior GA GA GA GA GA GA GC GC GE GE GF GF GF GA green tinge GA molded glass (flower design); possible part of tableware; jar with grooves; bumps GA molded glass (screw cap of jar), one piece patinated GA molded glass, 73

77 writing (ASO, Coca Cola bottle 10 fl oz) GA molded glass; flower pattern; white paint on exterior GA molded glass; one with white print on exterior GA red backing GK white interior; green exterior (Vessel 1) GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge; patinated GA patinated; green tinge GA patinated; green tinge GA patinated; green tinge GA patinated; green tinge GA GA GA GA GA GA GC GA molded glass GA molded glass (flower design); possible part of tableware GA blue tinge; patinated GA green tinge 74

78 GA green tinge; all part of one bottle GH milk glass GA patinated GA GA GA GA GA GC GC GH GA molded glass GA molded glass GA molded glass; one piece similar to flowered glass in GA molded glass; one w/ wave pattern; another possible bottle top; another with writing "82 W 2" GA rough texture on exterior; warped; bit of concrete attached GE GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge; patinated GA green tinge; patinated GA patinated GA patinated 75

79 GA patinated GA patinated GA patinated GD patinated GI patinated GA patinated; green tinge GA patinated; green tinge GH rough texture on exterior GK white interior; green exterior (Vessel 1) GA GA GA GA GA GA GA GA GA GA GA GC GC GD GD GE GE GJ molded glass GA molded glass GA blue tinge GA green tinge GA green tinge GA patinated GA patinated GC patinated GD patinated; white paint on ext GK white interior; green exterior 76

80 (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GA GA GA GA GA GA GA GA GA GC GC GC GE GE GF GA green tinge; possibly part of bottle GA molded glass GA molded glass GA molded glass GA molded glass GA molded glass GC molded glass GE molded glass (bumps on exterior) GA molded glass (one piece part of a coca cola bottle - COC) GA molded glass (rippled) GA molded glass (wave pattern) GA molded glass, writing (LE, MAS, 77

81 white print "40", ER DEM SON, RKRE) GA molded glass; green tinge GA molded glass; green tinge GA ridges on inside GA some fragments part of dish GA two pieces part of same bottle GA GA GA molded glass GK white interior; green exterior (Vessel 1) GC patinated; molded glass GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge GA green tinge; patinated GA metal connected GA one very thick piece GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GA GA GA GA

82 GA GA GA GA GA GA GC GD GE GH mason jar lid; metal attached; writing ("GENUINE ZINC CAP, FOR BALL MASON JARS") GA molded glass GH molded glass (see lot 1431) GH writing ("I6") GA patinated; metal attached (likely a mirror) GA blue tinge GA blue tinge GA green tinge GA green tinge GA green tinge; patinated GC part of bottle GE part of bottle in GE part of bottle; seam going up one piece; slightly patinated GA patinated GA patinated GE patinated GA rough texture on exterior GC very thin GK white interior; green exterior (Vessel 1) 79

83 GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GA GA GA GA GA GA GA GA GA GA GA GA GA GC GC GC GC GE GF GF GH GH GA green tinge; corner of glass bottle GA green tinge; molded glass (ridges) GC mass of pockmarked glass GC molded glass GA molded glass 80

84 (ridges) GA molded glass (ridges, "AMERY") GA molded glass (ridges, bubbles) GA molded glass (ridges, grooves, possible part of bottle base) GA molded glass (ridges, possibly part of cup) GA molded glass (ridges, star imprint) GA molded glass (wave pattern) GH part of mason jar cap GA purple tinge; patinated; molded glass GA square shape GA GA GA GA GH GA green tinge GF partial rim GA GF GA green tinge; partial maker's mark (unclear) GA green tinge GA GA GH GE almost a whole bottle/glass; writing "G11"; maker's mark 81

85 present; seam on side; slightly patinated GH maker's mark (13) GA molded glass (maker's mark and writing "KERR GLASS MFC SAND SPRING OKL AUG 31, 1915") GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GA molded glass (wave pattern) GA diamond base shape; maker's mark present GA rough texture GA one partial and one whole bottleneck; seam goes all the way up GA GA GC green tinge; patinated; two pieces fit together GA blue tinge GA blue tinge; white paint on exterior GA green tinge GA green tinge GA green tinge GA green tinge 82

86 GA green tinge GA patinated green GA GA GA GA GA GA GC GE GH jar lid pieces? Writing "cap of" GA molded glass GA blue paint on exterior of one piece GA blue tinge GA blue tinge GA blue tinge GA green tinge GA green tinge; patinated GA green tinge; patinated GA green, red patination (possibly paint) GA magnifying glass? Very thick GA one piece looks like part of thermometer GA purple tinge GK white interior; green exterior (Vessel 1) GA yellow tinge; bubbles GA GA GA GA GA

87 GA GA GA GA GC GE GE GF GH GA blue tinge GA blue tinge; writing (ER) GA green tinge; molded glass (ridges) GA molded glass GE molded glass GA molded glass (indentation, ridges) GH molded glass (possibly bottom of jar) GA molded glass (ridges, repeating square pattern) GA molded glass(writing: "HES," "NE") GH molded glass; possibly bottom of jar GA patinated GA ridges; grooves GA GA GA small rectangular bottle; maker's mark present (L) GK white interior; green exterior (Vessel 1) GA intact jar; seam goes all the way 84

88 up; makers mark present; writing (G08 3/8, 14) GA patinated GA green tinge GA molded glass, writing (0) GA bubbles GA green tinge GA green tinge GA green tinge; patinated GA green tinge; patinated GA green tinge; patinated GA possible part of base of a bottle GA very thick molded glass; possibly bottom of bottle or jar GA GA GA GH GD bubbles GA patinated GA patinated GG patinated GA purple tinge; patinated GE white paint on exterior GA yellow tinge GA GA GA GC GH GA blue tinge GA blue tinge; molded glass 85

89 GA molded glass (bubbles; sharp curves; writing "HIS," "F12") GC molded glass (possible bottleneck) GA molded glass (writing); piece of measuring jar GH molded glass (writing: "genuine porcelain") GA molded glass; patinated; green tinge GC patinated; gold paint on exterior GA possible bottle base GA possible part of rim with ridges GA purple tinge; molded glass (writing) GA patinated GH GA GA GA blue tinge GA blue tinge; patinated GA green tinge GA green tinge; patinated GA purple tinge; patinated GA yellow tinge; patinated GA GA GA GA GA

90 GH writing ("ORCEL" porcelain?) GA blue tinge GA blue tinge; part of bottle base? GA blue tinge; patinated GA blue tinge; patinated GA green tinge GC highly patinated GC patinated GK pink GF white paint on exterior GA yellow tinge GA GA GA GA GC GC GC GA edge of square bottle GA four patinated molded glass pieces; fit together to make small bottle or shot glass GA green tinge GA molded glass; one piece has very rough texture on exterior GA molded glass; writing (EPH) GF possible partial bottle rim GA ridges; grooves GA square pattern GC partial rim GA GE mold number (2) 87

91 GA blue tinge GA green tinge GA green tinge XXX 9 2 GA green tinge? 9 2 GA patinated; green tinge GA GA blue tinge; highly patinated GA blue tinge; possibly jar base; highly patinated GF bubbles ; patinated? 9 2 GA patinated? 9 2 GC patinated GA patinated; green tinge GA purple tinge GK white interior; green exterior (Vessel 1) GA GA GA ? 9 2 GA XXX 9 2 GA GC GC ? 9 2 GC GD GH XXX 9 2 GA blue tinge; patinated GA molded glass? 9 2 GA molded glass GH molded glass (ridges)? 9 2 GA molded glass; patinated; brown backing? 9 2 GA molded glass; yellow tinge GA patinated 88

92 ? 9 2 GA patinated; green tinge? 9 2 GA patinated;blue tinge; white paint on int/ext GA white paint on int/ext; blue tinge; patinated; words on one piece "PECU ATOP," another "REGULATOR" GA writing ("federal law prohibits reuse of this bottle," ridges)? 9 2 GA molded glass GK white interior; green exterior (Vessel 1) GK white interior; green exterior (Vessel 1) GA patinated GA seam goes all the way up? 9 2 GA molded glass?? 10 1 GA green tinge GC patinated???? 10 1 GA yellow tinge?? 10 1 GA GA green tinge???? 10 1 GA blue tinge; patinated GA blue tinge; patinated; white paint on int/ext?? 10 1 GA green tinge???? 10 1 GA green tinge; patinated?? 10 1 GE part of bottle; seam going up one piece GE patinated?? 10 1 GA purple tinge?? 10 1 GF very thick 89

93 ?? 10 1 GA very thin???? 10 1 GA white paint???? 10 1 GF white paint???? 10 1 GJ white paint?? 10 1 GA yellow tinge???? 10 1 GA ???? 10 1 GC GA GD GE patinated GC partial bottleneck???? 10 1 GA partial bottlerim; green tinge; patinated GA blue tinge; patinated GA blue tinge; patinated; white paint on int/ext??? 11 1 GA green tinge GA patinated GA patinated GA purple tinge GE white paint on int/ext GA GF GA blue tinge; patinated; white paint on int/ext GA green tinge GA patinated??? 11 1 GE possibly part of jar rim GC very dark green glass (almost black) GA GC ??? 11 1 GC GA molded glass; patinated GA molded glass; 90

94 purple tinge; rough texture on exterior GA raised ridge GA white paint on int/ext GA green tinge; patinated GA ????? 6 to 8 1 GA WALLFALL; one w/ bubbles; writing (ARK, P) TOTAL

95 92

96 Metal Artifacts (Non-nail) Ben Dwyer Introduction Metal artifacts can be difficult to analyze and catalogue because of significant amounts of rust that usually develop on said items. In spite of this, metal objects can serve as useful indicators of time periods. Analysts can use the rust layers to date an object (provided they have the proper equipment), or they can examine the object in an attempt to try and discern its purpose and use. This requires a great deal of patience and focus, for rust can make the simplest of objects very difficult to identify. If an object is identified, the information can then be used to try and determine a time period for the artifact and in turn date the excavated context. Analysis Metallic artifacts constituted a large portion of the items recovered during the excavation of Unit YC. After being bagged according to their level and subunit of discovery, the metallic artifacts were later cleaned and sorted into two separate groups: nails and non-nail metal. After this point, work began on cataloging the non-nail metal artifacts (Figure 1). The cataloging process utilizes the Yates Community Archaeology Program Artifact Cataloging System to organize the artifacts based on identifiable details. For non-nail metal, the first round of sorting and separation focuses on the sub-material of the artifacts (iron/steel, aluminum, nickel, etc.). It is known that the artifacts are metallic, but it is not known what type of metal they are. Once non-nail metal artifacts have been sorted according sub-material type (Figure, they are then further divided into artifact classes, such as bottle caps, keys, coins, jewelry, utensils, etc. More often than not, though, the majority of the non-nail metal artifacts are flat metal flake fragments (3173) or indeterminate (3176, Figure 2). In any case, each of the non-nail metal artifacts, no matter how big or small, requires careful scrutiny to tease out clues in order to figure out its purpose or use. This usually involves a healthy dose of imagination and discernment, 93

97 for some of the items have rusted a great deal in their time underground. It is at this stage where some exciting discoveries can be made. Once the artifacts have been divided into classes, they are then weighed in groups and the data is recorded into a table. Of the identifiable and indeterminate non-nail metal artifacts, the overwhelming majority examined were either iron or steel (Figure 3) and aluminum (Figure 4). Copper, brass, and nickel artifacts were present among the non-nail metal artifacts cataloged, but their presence declined after Level 5 (see Appendix). Only one non-nail metal pewter/tin item was found (see below). Intact Artifacts of Interest Of the non-nail metal artifacts cataloged from Unit YC, a few were identifiable and are therefore noteworthy. Little is known about these artifacts beyond what is found with/on them. Level 4 Subunit 2: A metal zipper, inscribed with TALON (Figure 5). The company that manufactured this zipper (TALON) is still in business today, and has been manufacturing zippers since the 1920 s, mainly for jeans. Subunit 3: A bottle cap. It is flat on one side and cupped on the other, and it has crimped edges. Level 5 Subunit 2: A bike pedal (Figure 6). There is still degrading rubber attached to the pedal. Subunit 2: A key (Figure 7), as evidenced by the head and shaft. Subunit 3: A file fragment with a rectangular cross-section (Figure 8). One can clearly see the diagonal striations in the photograph. Level 6 Subunit 3: A pull chain, perhaps for a light bulb. Level 7 Subunit 1: Light bulb bottom. The tapered/pointed end is missing, but it may have rusted away or become separated from the bottom at some point in time. 94

98 Subunit 2: A fountain pen clip, inscribed with SHEAFFER-CLIP. Based on the style of the inscription and the clip, I was able to determine that this particular clip was only manufactured between 1912 and 1922 (Figures 9 and 10). Level 8 Subunit 2: A pewter/tin cup that has been smashed (Figure 11). I came to this conclusion based on the presence of a handle, and the possible presence of an opening on the top and a rim on the bottom. Level 12 Subunit 2: A wrought square spike. Based on the depth of its discovery, it has been theorized that the spike may be from the trolley line that runs in front of the Yates house. Conclusion It is difficult to come to any overarching conclusion concerning Unit YC based simply on these non-nail metal artifacts, but it can be said that there was a high presence of metallic artifacts throughout the excavation levels of Unit YC. Furthermore, there was a discernibly broad range of non-nail metallic artifacts found, ranging from clothing items to hardware parts to pieces of electronic appliances to luxury items. If nothing else, this indicates the importance that metallic objects played in the lives of those that lived in the vicinity of the Yates house, and the degree of integration that these objects had achieved in people s everyday lives. References Talon Zipper accessed April 27, Pentrace Article # Sheaffer s Flat-Top Pens accessed April 27,

99 Overall Quantity and Weight of Non-Nail Metal Artifacts Quantity + Weight (g) Level Quantity Weight (g) Figure 1: Overall quantity and weight of all non-nail metal artifacts by level. Unknown Sub-material Artifacts Quantity + Weight (g) Level Quantity (Indeterminate) Quantity (Determinate) Weight (Indeterminate) Weight (Determinate) Figure 2: Quantity and weight of determinate and indeterminate unknown sub-material artifacts by level. 96

100 Iron/Steel Artifacts Quantity + Weight (g) Level Quantity (Indeterminate) Quantity (Determinate) Weight (Indeterminate) Weight (Determinate) Figure 3: Quantity and weight of determinate/indeterminate iron/steel artifacts by level. Aluminum Artifacts Quantity + Weight (g) Level Quantity (Indeterminate) Quantity (Determinate) Weight (Indeterminate) Weight (Determinate) Figure 4: Quantity and weight of determinate/indeterminate aluminum artifacts by level. 97

101 Figure 5: Zipper, inscribed with TALON. 98

102 Figure 6: Bike pedal Figure 7: Key 99

103 Figure 8: File Figure 9: Fountain pen clip, inscribed with SHEAFFER-CLIP. This clip was manufactured between 1912 and Figure 10: Example of a SHEAFFER-CLIP fountain pen clip. 100

104 Figure 11: Pewter/tin cup 101

105 Appendix: Catalogue of Analyzed Metal Artifacts (Excluding Nails) Lev el Subunit Lot Numbe r Materi al Sub - mat. Art. Class Qu ant ity Wgt. (g) Comments Metal MA Indeterminate Metal MH Modern: indeterminate (can lid?) Metal MA Modern: wire, round (crosssection) Metal MI Modern: shell casing OR pen cap Metal MH Modern: pull tab Metal MH Modern: aluminum pan fragment Metal MA Flat metal fragments: "flakes" Metal MA Screw Metal MA Modern: staple, large Metal MA Indeterminate Metal MA Indeterminate; white band around upper middle Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Modern: square washer Metal MH Modern: pull tab Metal MA Modern: toggle switch Metal MA Modern: clothing rivet Metal MH Modern: small tube Metal MA Bolt Metal MA Flat metal fragment: "flake" Metal MI Flat metal fragment: "flake" Metal MH Modern: aluminum pan fragments Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Indeterminate Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: 102

106 "flakes" Metal MA Spoon (fragment) Metal MA Hardware: indeterminate Metal MA Flat metal fragment: "flake" Metal MJ Zipper, pull tab reads "TALON" Metal MA Flat metal fragments: "flakes" Metal MA Square spike: other (painted red) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Indeterminate Metal MA Nut Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Safety pin or Hook Metal MA Staple: small Metal MB Wire: round (cross-section) Metal MH Indeterminate Metal MH Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Bottle cap: crimped sides Metal MA Indeterminate Metal MA Bottle cap: crimped sides Metal MA Nut Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Hardware: indeterminate Metal MI Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Bottle cap: crimped sides (fragments) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Hardware: handle Metal MA Flat metal fragments: 103

107 9 "flakes" Metal MA Indeterminate Metal MA Hook Metal MA Hardware: hinge/hinge plate Metal MA Indeterminate Metal MA Hardware: key Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Hook Metal MA Flat metal fragments: "flakes" Metal MH Other: pencil ferrule Metal MB Hardware: indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Wire: round (cross-section) Metal MH Other: metal label Metal MA Hardware: indeterminate (coil) Metal MA Safety pin or Hook Metal MI Indeterminate Metal MA Rivet: clothing Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Hardware: indeterminate (coil) Metal MA Hardware: key Metal MA Flat metal fragments: "flakes" Metal MA Hardware: indeterminate (coil) Metal MA Modern: bike pedal Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Bottle cap: crimped sides Metal MA Wire: round (cross-section) Metal MA Indeterminate Metal MA Flat metal fragment: "flake" Metal MA Can: round Metal MH Indeterminate Metal MH Hardware: indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: 104

108 "flakes" Metal MA Indeterminate Metal MA Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Staple: small Metal MA Flat metal fragment: "flake" Metal MA Indeterminate Metal MH Pull tab Metal MA Indeterminate Metal MH Staple: large Metal MA File: rectangular (crosssection) Metal MA Indeterminate Metal MA Bottle cap: crimped sides Metal MI Hardware: indeterminate Metal MA Indeterminate Metal MA x nuts, 1x bolt, all connected Metal MA Indeterminate Metal MH Bottle cap: screw sides Metal MI Hardware: indeterminate Metal MI Flat metal fragment: "flake" Metal MA Indeterminate Metal MA Hardware: indeterminate Metal MA Hook or Latch Metal MA Indeterminate; threaded at one end Metal MA Flat metal fragment: "flake" Metal MA Wire: round (cross-section) Metal MA Indeterminate Metal MA Indeterminate Metal MA Wire: square (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Can: round Metal MA Flat metal fragments: "flakes" 105

109 Metal MA Bottle cap: crimped sides Metal MA Indeterminate Metal MI Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MH Other: pencil ferrule Metal MH Other: metal label Metal MA Flat metal fragments: "flakes" Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Indeterminate Metal MA Other: pull chain Metal MA Bottle cap: crimped sides Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Indeterminate Metal MA Button: plain (hollow) Metal MA Bolt and Nut Metal MA Wire: round (cross-section) Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MI Indeterminate Metal MA Indeterminate Metal MA Bottle cap: crimped sides Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Hardware: indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MH Modern: lightbulb bottom (screw) Metal MI Hook Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Thimble Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Indeterminate 106

110 Metal MA Indeterminate Metal MA Other: clip to pen; reads "SHEAFFER-CLIP" Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Bottle cap: crimped sides Metal MA Hardware: indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Hardware: indeterminate Metal MA Bottle cap: crimped sides Metal MA Staple: small Metal MA Wire: round (cross-section) Metal MA Indeterminate Metal MA Indeterminate Metal MI Wire: round (cross-section) Metal MA Staple: small Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MI Hardware: indeterminate Metal MI Indeterminate Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MF Complete vessel: cup Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate 9 2 #### Metal MA Indeterminate #### Metal MH Other: pencil ferrule 9 2 #### Metal MA Hardware: indeterminate (coil) 9 2 #### Metal MA Utensil: spoon 9 2 #### Metal MA Flat metal fragments: 107

111 "flakes" Metal MA Staple: small 10 1 #### Metal MA Indeterminate 10 1 #### Metal MA Flat metal fragments: "flakes" Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MH Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Chain: large links Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Wire: round (cross-section) Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Wire: round (cross-section) Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MI Indterminate 11 1 #### Metal MA Indeterminate 11 1 #### Metal MA Flat metal fragments: "flakes" Metal MA Flat metal fragments: "flakes" Metal MA Hardware: indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Indeterminate Metal MA Flat metal fragments: "flakes" Metal MA Indeterminate Metal MA Square spike: wrought 108

112 Nail Fragments Dustin Tannahill Introduction Nails have been separated from other metals into their own class because of the vast quantity of nails found in Unit YC. A total of 4,615 nail fragments (8,440.4 g) have been catalogued and analyzed, including those fragments with or without a head intact. By looking at the history of nail making, it is possible to determine approximate date ranges for the types of nails found at the site. The results of the data outlined in this chapter will then contribute to the overall chronology of Unit YC. History of Nail Manufacturing Prior to the late 1700 s and early 1800 s, nails were made by hand by a blacksmith or nailor. These hand wrought nails were heated and hammered out of a square iron rod. Handmade nails were used for sheathing, roofing, finish work, trim boards, and flooring. By the 1800 s nails were being mass manufactured on machines that chopped the nails off of an iron bar. These machines produced a tapered shank by bending the bar from side to side, and later a machine was invented to pound the head on the end. This type of nail became known as Type A cut nails and were produced until about the 1830 s. At this time new technology was developed that turned the iron bar over after each stroke of the hammer. The cutter was also set at an angle to produce a tapered cut. These Type B cut nail were widely produced throughout the rest of the 19 th century. Following new developments in steel processing, steel became the most popular material for nails in the 1890 s. These new nails were cut from a round steel wire, and became known as wire cut nails. By 1913, 90% of all nails were wire cut nails, although type B cut nails are still used today for hardwood flooring and historical restoration purposes (Visser 1996). Table 1 presents the date range for each type of nail. 109

113 Cataloguing Methods Nails were divided into three main classifications: round nails, square nails, and indeterminate nails. Round nails (Figure 1) are wire cut nails with a distinctly round shaft. Square nails (Figure 2) are most likely Type B cut nails and have a square shaft at the cross section of where the head meets the shaft. Indeterminate nails could not be placed into a category based on their level of rust or malformation. Round tacks (Figure 3) are small round nails, usually about one centimeter in length. Spikes (Figure 4) are either round or square nails that are at least 12 centimeters in length and have a thickness to the shaft. Additionally, screws were classified based on the thread around the shaft, and bolts had a clearly defined flat end instead of a point. One stake was classified based on its shape and length comparable to that of modern tent stakes. Analysis Looking at the overall quantity of nail fragments by level (Figure 5), we can see that nail fragments peaked in Level 6 with a considerable drop off after Level 7. This concentration of nails could be evidence of a structure or workplace that contained a large number of nails. When we look at the amounts of round and square nails by level (Figure 6), we can see that round nails peak at Level 6. Very few round nails were found deeper than Level 7, and none were found in Levels This data indicates that these levels were deposited before the production of wire cut nails. Square nails peaked at Level 7, but no square nails were found in Levels 1-4. Since the two peaks in nail type occur between Levels 6 and 7, this could possibly be a sign of the shift in nail production at the turn of the century. It should also be noted that the difference between the weights of nail types in Figure 7 is not as great as the difference in quantity of nail types in Figure 6. This is because square nails found at the site were much heavier and thicker than wire cut round nails. Indeterminate nails accounted for a large portion of nail fragments, mostly because of deterioration brought on by the humid Houston climate. Figure 8 and Table 2 give a breakdown of nail fragment quantity by each subunit. This information is important because we stopped excavating Subunits 3 and 4 after Level 6. Notice that reducing our subunits by half had very little effect on the total quantity of nail fragments 110

114 found between Levels 6-7. This could imply that a true peak in total quantity of nail fragments may have occurred in Level 7 if we had continued excavating Subunits 3 and 4. We also see a greater concentration of nail fragments in Subunit 2, which could indicate a more precise location of a structure or workplace. Looking at Table 3, we can see that almost the entire amount of round tacks found were contained in Levels 4 and 5. Conclusion In our analysis of the nail fragments in Unit YC, it is important to look at how the different types (round or square) impact the chronology of the site. The transition from Level 6 to Level 7 shows a sudden shift from a peak in round nails to a peak in square nails. This is a strong sign of the transition to wire cut manufacturing methods at the end of the 19 th century. Figure 6 shows that there were a large number of round nails in Level 7, and a small amount in Level 8. These could be explained by the approximately 15 year period before 1900 where wire cut nails were slowly gaining popularity across the United States. In 1892 wire cut nails made up over half the nails manufactured in the United States (Visser 1996). This is represented in Figure 6 which shows round nails already overtaking square nails in Level 7. In 1886 wire cut nails were only 10% of the nails manufactured in the United States (Visser 1996). These statistics are also reflected in Figure 6, which shows that Level 8 contains far more square nails than round nails. It should also be noted that finding round nails does not mean a level is dated before they became mass produced, especially in the mixed urban soil of Houston. Human activities and natural processes can always cause artifacts to shift in the ground. Our findings have certainly been able to back up conclusions found in the 2008 report, in which they too found evidence of Levels 6 and 7 dating to the turn of the century (Bahorich 2008:66). This previous work did not offer a firm conclusion based on the large amount of indeterminate nails found in those levels. Here, the amount of indeterminate nails does not comprise a majority of the fragments found, and in conjunction with the findings of the 2008 report we have a much better picture of where the year 1900 occurred stratigraphically at the Yates site. 111

115 References Visser, Thomas D Nails: Clue s to a Building s History. Electronic Document, accessed April 19, Bahorich, Ben 2008 Nails. In Final Report on the 2008 Excavations at the R.B.H. Yates House. S.K. McIntosh and B.T. Clark, eds. Pp Rice University. 112

116 Image Type, Years Produced Hand-wrought nail, before circa 1800 Type A cut nail, circa Type B cut nail, circa Wire nail, circa 1890-present Table 1: type and production of nails 113

117 Figure 1: round nails Figure 2: square nail 114

118 Figure 3: round tacks Figure 4: spike 115

119 Figure 5: total quantity of nail fragments by level Figure 6: quantity of indeterminate, round, and square nails by level 116

120 Figure 7: weight (g) of indeterminate, round, and square nails by level Figure 8: total quantity of nails by level and subunit 117

121 Level Subunit 1 Subunit 2 Subunit 3 Subunit 4 Subunit N/A Grand Total Grand Total Table 2: total quantity of nails (broken down by level and subunit) Level Square Spikes Round Spikes Square Tacks Round Tacks Screws Bolts Stakes Grand Total Grand Total Table 3: total quantity of miscellaneous fasteners (broken down by level) 118

122 Appendix of Nail Fragments Lot # Level Subunit Sub Artifact Material Class Quantity Weight (g) MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

123 MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

124 MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

125 MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

126 MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

127 MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA

128 MA MA MA MA

129 126

130 Faunal Remains Kylie Klein Introduction Faunal analysis of Unit YC can help reconstruct food consumption trends for the inhabitants of the Yates household. This can be accomplished through categorizing the different types of animals and specific cuts of meat consumed. Insight into the diet of the inhabitants can aid in understanding how socioeconomic status was reflected in food consumption. Other modifications to the bones, such as burning or bleaching, also add important information in reconstructing the activities that took place in the area around Unit YC. Methodology Analysis began by sorting the faunal remains in potentially identifiable or unidentifiable fragments. Potentially identifiable remains were ones that could be further classified to either species or type of bone. Unidentifiable remains were ones that could not be further analyzed because they had no distinguishing characteristics, and were therefore indeterminate, or could not be analyzed beyond their initial categorization into classes such as bird or mammal. As the bones were sorted, they were weighed and catalogued by level and subunit. Independent of the classification as identifiable or unidentifiable, bone that had been bleached, burnt, cut, or modified was weighed individually and classified respectively. Bones were identified by species and/or element through comparison with collections in the Archaeology Laboratory at Rice University. Artifact Frequency Including identifiable and unidentifiable faunal remains, Level 10 had the highest frequency of fragments. However, a significant portion of these bones can be attributed to the presence of a dog skeleton in this level. The dog remains could have been deposited at the depth of Level 10 or buried at a later date so they do not necessarily correlate with the other artifacts of Level 10. Without the 318 g of dog bones, Level

131 has the lowest amount of faunal materials. Taking this into account, Level 5 had the highest faunal density. Among the lowest densities were Levels 1 and 2. Figure 1 shoes that there is no clear trend through the different levels in the total amount of the faunal remains. The data for identifiable bones also fluctuates throughout the levels considerably. There are more unidentifiable fragments Levels 5 through 7, with another spike at Level 11. Species Identification Figure 2 summarizes the percentages of identified species in Unit YC, based on bone weight. Species density is greatest for dog, followed by cow. Chicken, deer, and pig make up smaller proportions. Fish and rat do not constitute a substantial proportion but are present. Bones identifiable to type but not to species are labeled as unclassified. Bird unclassified consists of remains that were not identified as chicken; some of these may be wild species. Mammal unclassified comprises bones that had significant indicators of type but were too fragmentary to identify to species. However, with some bones, cow could be eliminated as a potential species, creating the mammal unclassified, not cow category. Bones of smaller animals are subject to greater fragmentation due to their increased fragility, whereas larger bones are more durable and thus more capable of sustaining an identifiable size. Therefore, this data may be lacking percentages of remains belonging to smaller animals. The number of bones identified to species is outlined in Table 1. Chicken bones are the most prevalent. As the Minimum Number of Individuals (MNI) was not calculated, it is not clear how many individual chickens this represents. The quantity of identified chicken bones increases in Levels 4 and 5, reaches a maximum in Levels 6 and 7. There is a constant number of mammal unclassified fragments throughout the levels except for a large peak in Level 5. Other species with bones identified display low numbers randomly dispersed through the levels. The classification of bone to species, including any modifications, is summarized in Table 2. The large increase in the quantity of chicken bones in Levels 6 and 7 may have resulted mostly from one chicken since there is diversity in the representation of the 128

132 types of bones present. There are distinct left and right bones from different skeletal parts, leading to the possibility that they comprise a single chicken. Additionally, there is only one identified butchered chicken bone, in Level 5. These factors support two different hypotheses. The presence of different types of chicken bone within two levels may indicate that chickens lived and died in this area. Or, more likely, it validates the proposition that this area was used for trash disposal. Two cow teeth were located in Level 9. It may be that cows were raised on site or that unusual cuts of meat were consumed. If the residents were butchering their own meat, this would also explain the presence of bones not typically found in cuts of meat, although the evidence does seem to argue against this. Analysis of Modified Bone Figure 3 illustrates the distribution of bleached, burnt, and butchered bone by level in terms of total number of fragments. Bleached bone was only present in Level 7 and results from extensive exposure to sunlight. This level would not have been quickly covered but would have remained subjected to the sun. Burnt bone appears in five levels, with the greatest amount in Levels 5 through 7. The smoke blackened fragments do not result from being cooked but instead indicate direct burning from fire. An explanation for burnt bone would be the burning of trash in pits. However, there is not a significant proportion of burnt bone in these levels to necessarily demonstrate that the contents of the unit were intentionally burned. Butchered bone was found in Levels 2 through 11. A maximum was reached in Level 6, and a significant number was present in Level 5 when the fourth category of burnt and butchered is considered. The identified butchered bones are listed in Table 2 along with the type of cut determined through analysis. Butchered bones designated leg could not be further classified as hind or front so the specific cuts of meat could not be identified past round or chuck cuts (Figure 4). In cows, round cuts are of the hind legs where the meat is tougher and typically used to make round steaks or jerky. In pigs, round cuts are for ham. Chuck cuts are of the front legs and are one of the most common sources for roast and hamburger meat. Shank cuts are of the lower legs where the meat is so tough that it is generally used in stews (Figure 5). For cuts of meat, round and shank 129

133 are less expensive. There are also several rib cuts, mostly of pig. The cow shoulder chop would have been part of a steak or roast. The loin chop is from the backside of the middle section of a cow or pig. These cuts are characteristic of more expensively-priced cuts. The more expensive cuts of meat were found in Level 2, whereas the other cuts of meat appeared in Levels 3 through 10. Since there is a difference in price of meat between round and chuck cuts and in most cases there was an inability to distinguish between them, it is difficult to draw any general conclusions concerning trends in socioeconomic status. The marks on the identified butchered bones indicate a clean cut, meaning a band saw was used during butchery. Professional butchers use band saws, but it is also possible for individuals to own them. The butchered cow femur (Figure 6) is not a typical bone fragment found in particular cuts of meat that one would purchase. Instead, it implies that a larger section of meat was purchased, which may have been butchered at the site. There were sixteen identified butchered cuts, significantly less than the 48 from Unit P, excavated in A possible explanation for this is that Unit YP was excavated near the Yates house and was a primary trash pit. Due to its proximity to the house it would have been more convenient location to dispose of bone fragments after meals. Canine Remains of a dog skeleton were found in Level 10. As previously mentioned, the dog remains dominated the total faunal assemblage of Level 10. Further analysis revealed dog bones in Levels 3, 6, 7, 8, and 10. Remains in the higher levels (3, 6, 7) consisted of two phalanges and a rib fragment. The size of these fragments would have permitted them to easily migrate upward through the levels upon disturbance. Table 3 outlines the quantity and type of canine bone, per level for the unit. The canine bones belong to a single dog, and most of the skeleton is represented and intact. Figure 7 displays the general skeletal canine anatomy. Of the twenty-seven canine vertebrae only the first cervical vertebra is missing. No cranium fragments were recovered. Interestingly, the vertebra missing is the first of the vertebral column, which begins at the base of the cranium. Of the skull, only the right mandible (Figure 8) was present. 130

134 Analysis of the teeth indicates that the dog was older than one year, yet the teeth did not show excessive ware so the dog had not reached old age before death. Based on the skeletal size, the dog was small, and the presence of a baculum denotes male. The growth plates were fused further supporting that the dog was older than one year, and during its lifetime, the dog sustained a fracture to its left proximal fibula. These remains exemplify that diverse types of unbutchered bone from the same species may be evidence that the remains are of an animal that lived at the site. Overall Trends There seems to be a correlation between the different classifications of bone in Levels 5 through 7. In these levels, the amount of unidentifiable, bleached, burnt, and butchered bone is highest. Moreover, these levels contained lighter, sandier soils in comparison to the other levels. Levels 4 and 5 had the greatest composition of lighter, sandier soil. The change in composition can be attributed to flooding in the area or to purposeful addition of sand to even the area. The bottom of Feature 1, three stacked concrete blocks, extends into these levels as well. Construction of this supposed pillar may have disrupted some of the faunal remains in these levels. Any of these scenarios may account for the increase in the more fragmentary unidentifiable bone. Overall, there is increased faunal activity in Levels 5 through 7. Conclusion Due to the relatively low numbers of butchered bone and lack of obvious cuts of meat within Unit YC, little can be concluded in terms of socioeconomic status. The bones, however, do display insight into the species composition and their relative percentages. The identified bone helps construct which species may have lived at the site and provides a brief glimpse of the inhabitants diet. 131

135 Faunal Frequency Weight (grams) Unidentifiable Identifiable Level Figure 1: Weight of identifiable and unidentifiable faunal remains, in grams Species of Unit YC Chicken Cow 18% 2% <1% 6% 3% 4% 29% Deer Dog Fish Pig <1% 4% Rat 34% Bird unclassified Mammal unclassified Mammal unclassified, not cow Figure 2: Percentages of species represented in entire unit 132

136 Table 2: Classification of bone type, by species and level with modifications indicated Level Species Chicken Cow Deer Fish Pig Rat Bird unclassified Mammal unclassified Mammal unclassified, not cow Table 1: Number of bones identified by type, per species Level Species Type Grams Comments 2 Chicken 1 Femur 0.3 Burnt 2 Cow 1 Scapula 8.1 Shoulder Chop 2 Pig 1 Rib Rat 1 Right Mandible Mammal unclassified 1 Rib 2.9 Loin Chop 3 Cow 1 Femur 132 Cut, Young 3 Pig 1 Vertebra Mammal unclassified 1 Rib Mammal unclassified, not cow 1 Femur Chicken 1 Cervical Vertebra Chicken 1 Dorsal Vertebra Chicken 1 Phalange Chicken 1 Rib Chicken 1 Sacrum

137 4 Fish 1 Vertebra Pig 1 Rib 2.5 Cut 4 Pig 1 Tusk 25 4 Bird unclassified 1 Proximal Femur Mammal unclassified 1 Carpal or Tarsal Mammal unclassified 1 Joint Fragment Mammal unclassified, not cow 2 Tarsals 9.1 Shank Cut 4 Mammal unclassified, not cow 1 Ulna Chicken 1 Left Femur 1.7 Cut 5 Chicken 1 Radius Chicken 1 Right Humerus 1 5 Chicken 2 Tibia Chicken 3 Vertebrae Deer 1 Right Proximal Tibia 35 Shank Cut 5 Fish 1 Mandible Fish 2 Vertebrae Bird unclassified 1 Femur Mammal unclassified 6 Ribs Mammal unclassified 1 Rib 1.1 Burnt 5 Mammal unclassified 1 Tarsal 11.9 Shank Cut 5 Mammal unclassified 2 Vertebrae 88 6 Chicken 1 Carpo -Metacarpus Chicken 1 Cervical Vertebra Chicken 1 Dorsal Vertebra Chicken 1 Left Coracoid Chicken 1 Left Femur Chicken 1 Left Humerus

138 6 Chicken 1 Left Radius Chicken 1 Left Scapula Chicken 1 Left Tibia Chicken 1 Left Ulna Chicken 1 Radius Chicken 1 Rib Chicken 1 Sacrum Chicken 1 Vertebra Cow 1 Leg 22.2 Round or Chuck Cut 6 Fish 1 Vertebra Pig 2 Ribs 3.8 Cut Bird unclassified 1 Femur 3.9 Bird unclassified 2 Left Femur 1.4 Bird unclassified 1 Right Femur 0.5 Mammal unclassified 1 Vertebra Chicken 1 Coracoid Chicken 1 Left Femur 2 7 Chicken 1 Left Humerus 1 7 Chicken 3 Ribs Chicken 1 Right Carpo-Metacarpus Chicken 1 Right Tarso-Metatarsus Chicken 1 Sacrum 2 7 Chicken 1 Scapula Chicken 1 Tibia 2 7 Chicken 1 Ulna Cow 1 Femur 20 Round Cut 7 Cow 1 Rib 10.7 Cut 7 Pig 1 Head of Femur 3.1 Young 7 Pig 1 Vertebra Mammal unclassified 3 Ribs 9.4 Burnt 7 Mammal unclassified 1 Rib 3 Cut 135

139 8 Cow 2 Ribs Fish 1 Vertebra Mammal unclassified 1 Vertebra Cow 2 Leg 55.7 Round or Chuck Cut 9 Cow 1 Incisor Cow 1 Molar Cow 1 Vertebra Mammal unclassified 1 Leg 12.9 Round or Chuck Cut Mammal unclassified 1 Hind leg 8.7 Round or Chuck Cut Mammal unclassified, not cow 1 Vertebra 2.9 Mammal unclassified 1 Proximal Femur 5.2 Mammal unclassified, not cow 2 Joint Fragments 5 Young 136

140 Modified Bone Frequency Number of Fragments Bleached Burnt Butchered Burnt and Butchered Level Figure 3: Modified bone, by total number of fragments Figure 4: Examples of leg slices from either round or chuck cuts of unclassified species 137

141 Figure 5: Tibial shank cut of a deer Figure 6: Butchered femur of a young cow Table 3: Classification of canine bone type, by level Sub Lev Type Quantity Comments 3 3 Phalange Rib Phalange Caudal Vertebra 4 138

142 1 8 Lumbar Vertebrae 2 6th, 7th 1 8 Pelvis Sacrum Caudal Vertebra Cervical Vertebrae 6 2nd, 3rd, 4th, 5th, 6th, 7th 1 10 Left Accessory Carpal Left Metacarpals 3 2nd, 4th, 5th 1 10 Left Metatarsals 4 2nd 3rd, 4th, 5th 1 10 Left Pelvis Left Proximal Fibula 1 old fracture present 1 10 Left Radial Carpal Left Radial Carpal Left Scapula many 1 10 Left Tarsals 4 of 7 total 1 10 Left Ulnal Carpal Lumbar Vertebrae 5 1st, 2nd, 3rd, 4th, 5th 1 10 Baculum Pelvis Phalanges Rib many 1 10 Right and Left Femur 3 left femur in two pieces 1 10 Right and Left Humerus Right and Left Radius Right and Left Tibia Right and Left Ulna 3 one in two pieces 1 10 Right mandible Right Metatarsals 4 2nd, 3rd, 4th, 5th 1 10 Right Pelvis Right Scapula many 1 10 Right Tarsals 4 of 7 total 1 10 Right Tibial Tuberosity Sternum Thoracic Vertebrae 13 of 13 total 139

143 Figure 7: General canine skeletal structure ( Figure 8: Right mandible of canine 140

144 The Things They Wore: A Study of Personal Adornment at the R. B. H. Yates House Brittany Wise Introduction The inhabitants of the early 20 th -century Rutherford B. H. Yates house were not only consumers of mass-produced goods; they were conscientious shoppers who kept up with consumer trends while occupying a significant consumer space in the arena of exchange. One way of understanding how they constructed their identity within the African-American community in Houston, Texas is through an examination of the things they purchased. The act of shopping and consumption is now well recognized as a means to identity construction, and therefore a study focused on the choices these consumers made can shed light on how they attempted to create new meaning from mass-produced European-American material culture. The decisions they made about the clothing and accessories they wore provide direct evidence of identity construction, compared to the interpretive leaps that archaeologists sometimes have to make to see how other materials such as nails or glass can be indicative of identity construction. The repeated ritual of choosing what to wear and with what to accessorize are part of the process of the construction of individual identity (White 2005:6), and this act furthermore confirms socially-constructed ideas of the collective group. While there are numerous material culture studies of enslaved African-Americans, fewer have been undertaken that focus on free, land-holding African- Americans (Singleton 1999:4); this study aims to provoke discussion about one of these 141

145 diverse African-American communities and explore how material culture can indicate relationships between African-American identity and social class. It is rare to encounter whole articles of clothing during excavation due to the deterioration of their materials; however, the durable parts of garments are found in abundance (Heath 1999; Lee 2008; Mullins 1990: ; White 2005). These artifacts, such as buckles, beads, buttons, safety pins, rivets, and clothing hooks, combined with jewelry, pendants, and other decorative accessories form the personal adornment category that will be discussed in this essay. Unit YC at the Yates Museum provided too small a number of personal adornments to observe any significant trends, but nonetheless there were some valuable discoveries about some of the pieces that may speak to questions of identity and culture throughout the history of the Yates house and surrounding community. Methods Unit YC in the backyard of the Rutherford B. H. Yates house was excavated using hand trowels level by level. The closing of the unit for Subunits 3 and 4 at Level 7 may be a factor in the decrease in frequency of adornments found at later levels. When personal adornments were unearthed at the unit, they were bagged separately; depending on their fragility, they were sometimes stored in plastic containers and even wrapped in paper towel or aluminum foil to prevent breaking. One garment was discovered, and it was left in situ until the level was deep enough for it to be freed; at that point it was immediately bagged in a paper bag. 142

146 Washing the artifacts sometimes warranted special attention; depending on the material and condition of the adornment, it could be washed in water with a toothbrush (typical for most items), gently brushed with a dry brush to remove excess dirt (metal items), or dipped in water (fragile-looking items that were prone to flaking). Excess dirt was removed from the garment and small roots were picked or cut out of the decorative lace trim so that the garment could be straightened out and viewed in whole for effective analysis. After washing, personal adornments were sorted from the general artifacts found in Unit YC. These included beads, buckles, buttons, metallic clothing parts, pendants, safety pins, and the garment. These artifacts were cataloged into Microsoft Excel using the Yates Community Archaeology Program (YCAP) Artifact Cataloging System. The Digital Archaeological Archive of Comparative Slavery (DAACS) Cataloging Manual (Grillo et al. 2003) was also used in cataloging the buttons, since there was an abundance of them and further categorization proved to be advantageous for analysis in dating and detecting trends. The YCAP Artifact Cataloging System classifies the artifacts first by material, spanning a number of categories: ceramics, glass, metal, rubber (plastic), ecology, and other (textiles), then submaterial and artifact class are specified, giving a description of the artifact. It was determined that shell artifacts would fall in the sub-material category of shellfish (freshwater) in most instances, rather than ocean or land, due to its popularity in the United States during the period of the house s inhabitance. The freshwater pearl button industry boom first occurred in the United States around 1898; from that time forward, ocean pearl buttons from overseas decreased in circulation as various rivers in 143

147 the United States were used as sources of freshwater shells (Claasen 1994). It is difficult to distinguish between ocean and freshwater pearl buttons because few differences are uniformly clear (Hughes and Lester 1993: ). Since the color of the shell classes is variable when exposed to other elements such as detergent and general wear, slight color differences are not explored in this project. However, one clear difference between the two species of shellfish lies in the crystalline structure: saltwater species tend to exfoliate in layers, but freshwater species do not (Claasen 1994:74). This information made it possible better categorize the shell buttons; if a decayed button or button fragment was found to have layers flaking off, then the button was labeled shellfish (ocean). Finally, the shell buttons were compared to button guides to confirm their material composition (Hughes and Lester 1993: ). The DAACS Cataloging Manual classifies buttons in a descriptive manner, addressing the following aspects: completeness, manufacturing technique, type, shape, material, color, number of eyes, shank (style, material, condition), decoration (technique, material, color, and motif), and condition. The term eyes is synonymous with button holes, and shank is the loop sometimes attached to the backside of button that has no perforations, in order that the button can be sewn to a garment. The shank comes in many different variations whose popularity has waxed and waned, making it possible use this feature to date the buttons. In addition, measurements of weight, diameter, and height were taken using a scale that measured to the nearest tenth of a gram and calipers that measured in millimeters. The height to diameter ratio of each button was calculated in order to help determine the button type whether it is flat, domed, or semi-domed. A height to diameter ratio of 0.29 or less indicates a flat button, a semi-domed button falls 144

148 in the range of 0.3 and 0.45, and a domed button has a ratio of 0.46 or more (Grillo and Aultman 2003). Illustrations and descriptions in the DAACS Cataloging manual were used to help identify the buttons, especially with regards to shank description. The decoration of the sew-through buttons (buttons without shanks) was made easier to categorize by noting in the description motif category if the button fell into the any of the following categories of button style: fisheye (with or without ring), French bevel, English rim, and cup. Fisheye was a stylized, lenticularly-cut eye in which the two sewing holes were bored. The sometimes-accompanying ring is indicated by a raised rim encircling the button near the edge of the face. The French bevel style is characterized by the edge of the face sloping inward to an inner circle, where the button eyes sit. Finally, the English rim consists of an enlarged rim that occupies the outer third of the button face, leaving the perforations in the thinner-layered center (Claasen 1994:76-77). The china (or to be more accurate, Prosser) buttons were distinguished from others similar in appearance, such as glass, by examining the backside of the buttons. While glass buttons have a very smooth back, Prosser buttons have a well-known pebbly surface, comparable in appearance to an orange peel, which is theorized to be an imprint left by the surface the button rests on while being fired in the manufacturing process. Prosser buttons are also known to have a peripheral seam on the outer edge, but they are not unique in this so it is not considered a defining trait (Sprague 2002:111). Buttons made out of Prosser and glass are in fact so similar, their chemical composition is identical (Sprague 2002:119). The only sure way to determine the difference is to examine their crystalline structure. Prosser buttons that have broken will have uneven, jagged, grainy edges when viewed through a microscope, but glass buttons will break in 145

149 smooth-edged fragments (Sprague 2002:111). This method is effective but not practical, as it requires that a button be damaged in order to observe differences; since none of the buttons in question for Unit YC were broken, this method was not used. The clothing garment, a knee length cloth slip with lace trim, was dated using the microform series of Sears Roebuck and Company catalogs,. Based on the level where the garment was discovered, an estimate was made for the date of manufacture, and the Sears catalogs were sampled at five year intervals around the year estimated. The women s undergarments sections were explored for different years, until observable differences relating to the garment in this case, the use of lace were seen to take place that would indicate a terminus post quem.. Data Presentation In Unit YC, I examined 42 adornment artifacts in the following classes: beads, buckles, buttons, clothing eyes, safety pins, badges and garments. Three beads were found, as well as a series of five beads melted together and a rusted matrix containing numerous seed beads. All beads were made of glass, with the exception of an orange modern plastic bead, found in Level 1. Five buckles were recovered, four of which were intact; one was made of mother of pearl and resembled a stock buckle, one had a tan plastic frame with a metal tongue, and the rest were different metals, such as copper and brass. Compared to the other adornment types, a relatively large number of buttons (26) were unearthed: 20 shell/pearl, four metal, one plastic, and one glass (Figure 1). Other personal items consist of three clothing eyes, three fragments of steel safety pins, two badges, and an intact garment (Figure 2). Buttons were observed in all levels, with the 146

150 highest frequencies at Levels 4 and 5. With the exception of a modern plastic bead located at Level 1, beads were found in Levels 5 through 7 only. Three buckles were seen in Level 4, and two others were located in Levels 2 and 6 (Figure 3). Artifacts were distributed across the levels (Figure 4). Levels 1 and 2 each contained two artifacts each, followed by a break in Level 3, where no personal adornments were observed at all. Levels 4 and 5 contained the greatest frequencies, with ten artifacts each. Artifact numbers decreased to eight and six in Levels 6 and 7, respectively, succeeded by Level 8 with two findings. Levels 9 and 10 each yielded one artifact. No artifact classes were present in all levels, but metal as a material was represented in all but Levels 9 and 10 (not counting Level 3, with no artifacts). Rubber (hard plastic) materials were found in upper levels only, while glass materials were unearthed in Levels 6 and 7. Objects of shell were found in Levels 4 and below, where this material represented a sizable portion of personal adornments collected (Figure 4). The objects of shell (n=21) make up half of the total personal adornments analyzed, followed by metal (n=12). Rubber and glass were much less common, with only two adornments each of these materials (Figure 1). Analysis of Data Before any analysis of the personal adornments is performed, it must be noted that due to the small sample size, any apparent trends drawn from Unit YC are inconclusive and are stated to simply suggest possible trends and patterns. A larger sample would help to make trends more visible and findings statistically more significant. The relatively large number of buttons uncovered in Unit YC, along with button finds from previous 147

151 Yates backyard excavations (Units D and P), allowed me to study in greater detail the use of buttons as personal adornments, with a large sample size (n=104) of buttons (Table 1). Patterns observed in this larger sample can help to confirm suspected trends for Unit YC, so those figures constructed from a larger database will be included and referenced periodically. Stratigraphy In terms of the stratigraphic distribution of artifacts, there seems to be a break in artifacts at Level 3. This is interpreted to mean that, at this time, the use of the houselot changed significantly, contributing to a smaller number of personal adornments in the deposits. Alternatively, there could have been an abrupt occurrence where the area of the unit was covered by fill which did not contain artifacts deposited in the manner of previous decades. In the lower, earlier levels, there are relatively many more artifacts than the upper levels; the stark contrast in the artifacts numbers, as well as the absence of shell in material in upper levels, suggests that a distinct amount of time that passed between Levels 4 and 2. Thus, the former possibility offered is more likely. An examination of the data for the other two units indicates that it cannot be assumed that the backyard was used uniformly. While Level 3 in Unit YC is devoid of any personal adornments, Units D and P show their highest frequency of adornments in Level 3 (Figures 5 and 6). The only similar pattern detected is that there is somewhat of a break in artifact frequency in these units as well, but between Levels 5 and 6 and Levels 6 and 7. There are a number of reasons for these frequency differences between units: first, this might be a sample size issue, second the yard may have been used in various 148

152 ways and thus the patterns represent different patterns of deposition, and finally, it may be that the Level numbers do not correspond temporally. Artifacts Patterns and Consumer Choices From pearls to plastic The rising and falling frequencies of different materials can be compared to their introduction, availability, and popularity in order to determine information about the consumer choices of the Yates inhabitants. The freshwater pearl button industry experienced its first boom around 1897, around the Mississippi River Valley; prior to this date, ocean shell buttons were imported from overseas (Claasen 1994). The popularity of freshwater pearl buttons grew as shirts began to button down the front in the 1890s, and new resources of materials in the Arkansas River helped to replenish the dwindling supply of freshwater shell species. A slump occurred from 1909 to 1914 due to pearl buttons falling out of favor with consumers, but the First World War contributed to increased production of American buttons as fewer ocean shell buttons were imported. Another fall in production occurred only a year after a short-lived increase, from reasons due to World War I; these include a decline in consumer purchasing power, scarce labor, army preference for the duller vegetable ivory button, and the competing Japanese pearl button market (Claasen 1994). A widespread advertising campaign informed consumers of the merits of pearl buttons during the Great Depression, and this seems to be reflected in the data. The Yates family members were witnesses as well as active agents in these ongoing changes, as the decline in shell button frequency seen in Unit YC from Level 7 to Level 6 (Figure 4) could be a possible indication of the Yates decreased use of shell 149

153 buttons. The move in fashion away from small buttons and toward clothing that slipped over the head contributed to the decline of the pearl button industry, as new fasteners (such as the zipper, available starting in 1930) were invented that eliminated the need for a row of buttons. By 1962, the freshwater pearl industry had almost completely vanished as factories replaced their pearl line with plastic buttons (Claasen 1994). Data from Unit YC mirror these historical fashion trends (Figure 4); the break in Level 3 is preceded by an abundance of shell buttons, but in the later top two levels none were located. Similarly, in Unit D, Level 2 contained three plastic buttons but none of shell, while in Level 3 there were four shell buttons but none of plastic (the ecology material noted in Level 2 is a button carved from vegetable ivory, Figure 5). The earliest plastic buttons that is, those manufactured between 1920 and 1945, were creations more unique and original than the buttons manufactured in the second half of the 20 th century, which had more conventional styles, aimed at imitating buttons of natural materials. Art Deco buttons from the 1930s fit the colorful and inventive clothing style of the time, and are unique and flashy enough for button collectors to consider them worth searching for (Hughes and Lester 1993:66). Unit P contains four highly decorated plastic buttons in Levels 4, 5, and 6, three of which have a gemstone motif with a faceted face, and the remaining one was molded to imitate thread sewn in and out around the edge (Table 1). These buttons were clearly part of identity construction, worn with clothing of matching splendor, likely reflecting a sense of style and flair. Since clothing is rarely found in a preserved condition at archaeological sites, what we recover barely suggests its brilliance or critical role (Yentsch 1994:298). If buttons of wide variety and uniqueness both plastic and otherwise throughout all the sites are of any indication, 150

154 clothing and its accompanying adornments were used as a way to separate oneself in society: as a symbol of class, gender (see the discussion of the garment below), ethnicity, and even social issues (see the discussion of badges below). Keeping up with trends was a way for people living at the Yates house to construct social identity and convey to others not to mention themselves their self-identity. One trend that emerged clearly in the Yates data is that of shell button production. By examining the layering of shell buttons, it is possible to distinguish ocean and freshwater pearl buttons; this distinction allowed me to detect a shift from one to the other through time. The shift occurred from Levels 5 to 6 in Unit YC (Figure 7), and Units P and D offered evidence of similar trends, with freshwater shells found solely in the upper levels in the stratigraphy (Figures 8 and 9), as predicted by previous data and history patterns (Claasen 1994). The absence of any shells as outliers to the trend is helpful in narrowing the time frame for dating the levels, but the condition of the two industries as being simultaneously present for a considerable number of years prevents approximate dates from being provided with confidence. It seems likely, however, that consumers were presented with a choice of shell button types Evidenced by the earliest levels (8-10), the inhabitants of the Yates house were using ocean shell buttons. The artifact deposition reflects their usage shifting with the trend of the new freshwater pearls; from Level 7 to Level 4, there exists an increasing trend in the freshwater pearl buttons excavated, accompanied by a decreasing trend in ocean shell button count. If there was no consumer preference within the Yates household for a particular type of button, this trend would not likely be sustained over so many levels, for both materials. Indeed, an examination of shell material type over time yields evidence of the Yates 151

155 family as shoppers following trends of mass-consumers, using their economic power to purchase popular and stylish items to construct the identity they wished to present. A sharp decrease in consumption of shell buttons in Levels 1, 2, and 3 is mirrored by the extreme decrease in frequency of the artifacts (Figure 8, 9, and 7, respectively). This possibly shows an alternative way that economic power can be expressed this time, by the consumer withholding their capital from the market. Shoppers at the Yates house may have participated in the mass movement away from shell buttons, and this phenomenon exemplifies the way identity is dynamic: that many aspects of culture, especially fashion, undergo radical transformations. The role of the consumer as an active agent is to create new meaning from material that circulates in the society and the residents at the Yates house were no exceptions (Singleton 1999:5). Glass Ornamental Buttons There were three glass artifacts of personal adornment found in Unit YC: two in Level 6 and one in Level 7 (Table 1). Level 6 (Subunit 2) had two instances where multiple beads were melted together, one was a cluster of green seed beads in a matrix, and the other was a row of 5 purple beads lined up hole to hole. It seems significant to find the only two cases of melted glass in the same subunit of the same level; perhaps an activity occurred that found these two objects in the same proximity and thus affected similarly, such as a fire. The other glass object was half of a light blue bead, found in Level 7. There is a growing literature on blue beads in African American contexts, with blue beads seen as indicators of African heritage (Stine et al. 1996). However, due to the occurrence of only one bead, there is not enough data to speak to this issue. 152

156 In Units D and P there were five and two glass buttons, respectively (Table 1). No glass buttons were found in a Level later than 5 or earlier than 8 (Figures 5 and 6). Of great interest is the finding that Units D and P contain identical glass buttons both Level 7 yet the units do not neighbor each other; Unit D is next to the back porch and Unit P is closer to the center of the yard. With a diameter of about 12 mm, each button has a wire eye style shank and resembles a black jewel, with its face cut in triangular facets. This connection between the units supports other evidence that they were related as to how they were used (at least for the period represented by Level 7); this can be seen in the similar patterns artifacts across the levels (see Figures 5 and 6). These two buttons could have been from a single garment of clothing; since buttons tend to match the style of the clothes they are attached to, the ornately decorative buttons probably belonged on fashionable outfit of the time. That the buttons are found on the same level contributes valuable information that ties the stratigraphy of each unit together at Level 7. During that time period, areas covered by the two units must have been part of a single landscape; this helps us envision the back lot as a place where activities were carried out across the space, rather than seeing discrete activities in specific areas. Prosser mold adornments Unit YC only had one ceramic button and one bead in Level 1 (Figure 4); this button was Prosser molded. Units D and P collectively had ceramic (Prosser) buttons in almost all levels, (n=21, Figures 5 and 6). Their prevalence throughout the levels is a testament to the continued success and popularity of the Prosser button, which was invented in It would be expected that more specific dates could be given that are 153

157 based on events of improvements in machinery, techniques, and materials, yet there are no references currently available for dating. As with all buttons, dimensions of thickness, diameter, and weight were recorded until more is known about dating buttons (Sprague 2002:123). Metallic personal adornments Metal personal adornments consisted of three buckles, three partial or whole safety pins, three clothing eyes, three shank style buttons, a collar button, and two badges: one star-shaped and one of a generic badge shape. The buckles included a brass D-shape double frame that could have been a shoestrap buckle, a square double-frame that looks like a clothing buckle, and a square single-frame of unknown use in a crumbling, deteriorated state. The steel safety pins were found in Levels 6 and 7, and were fragmented: an intact one was hyperextended, one was missing its head, and one was in two pieces. Their close proximity (all in Subunit 3) may indicate that they were deposited together. In Unit YC there were four metal buttons: three relied on a shank to be sewn to the garment and were composed of copper, brass, and tin/enamel. The fourth, a collar button, was made out of pewter. The only intact shank whose style could be identified was on the tin/enamel button; it was thought to be a shank through back plate after consulting various references that had accompanying diagrams (Aultman and Grillo 2003; Hughes and Lester 1993). Unfortunately, unlike most of the other shanks, there is no date associated with the shank through back plate style, so a date for its provenience cannot be determined. Unit P only contained one copper or brass button to compare, but 154

158 since the shank is missing and the front is corroded, it is difficult to extract any data. Unit D, on the other hand, had nine metal buttons (Table 1). The shanks were either missing or unidentifiable, and five buttons had a design of some sort on their faces, but the corroded state of the buttons made them difficult to identify. One button that stood out was from Level 4; a copper button was inlaid with a red glass oval setting. These buttons contribute to the previously mentioned trend of using showy buttons as an outlet for one s personal expression. Work-clothes buttons One brass button from Level 4 in Unit YC is stamped with Burlington Overall on two separate lines and inscribed by a rectangular border. At 17 mm it is considered a medium-sized button, which is likely from a pair of work overalls manufactured by the Burlington Overall Company between 1900 and 1935 (ebay 2009; Funk & Junk Clothing & Collectibles 2009; Hughes and Lester 1993). Unlike most clothing in the early 20 th century, work clothing was advertised to consumers with the goal of establishing brand loyalty (Psota 2002: 111). This branding, which is now hard to escape in the present day, was a unique concept in the consumer world at that time, and the strategic placement of the brand on the buttons themselves resulted in self-identity and associations being constructed as the buyer chose their clothes. Since work-clothing was specifically tailored for working men, the companies would proudly proclaim if the clothing was union made (Psota 2002:114). This is the case for the Burlington Overall Company: one of its advertisements had the words union made directly under the company name (e-bay 2009). Another strategy used to gain a following was to choose names and trademarks 155

159 that mimicked well-established companies or familiar features of the American landscape, such as railroad routes (Psota 2002:119). This is precisely what Burlington Overall Company did; their design was similar to the one found on the buttons for the Burlington Route Chicago, Burlington, and Quincy Railroad (Hughes and Lester 1993:683). Another button from work clothes was found in Unit D, Level 6, dated from between 1889 and It had a diameter of 18 mm, similar to other one, and was imprinted with the word CARHARTT S in a heart outline. Although an array of Carhartt button images were examined, I failed to find a match for the design of this button. A more popular association that consumers made with the Carhartt company brand was the road car or rebus trademark, registered in According to button researchers, the Hamilton Carhartt & Company firm started in 1889 in Detroit, and was one of the biggest names in the work-clothes business at that time (Hughes and Lester 1993:685); in fact, it is one of if not the only few work-clothing companies still in existence from an extensive study conducted (Psota 2002:115). It continues to have the high reputation that it did since its inception, and is considered one of the more expensive, durable, and high-quality brands of work clothing (Psota 2002: 120). These buttons indicate several things about the Yates house inhabitants: the buttons came from clothes of someone that did labor-intensive work; they required durable clothing and were willing to spend extra money to get a top-quality product. As a consumer, they were probably aware of different work-clothes companies through advertisements and company associations; given these influences, they made decisions about which of the 156

160 many brands to buy. These buttons suggest that durability and quality were highly valued, as these brands were among the higher tier at the time. Badges Two metal badges from Unit YC both had legible inscriptions on them, which were helpful in finding an approximate date for their manufacture, as well as providing some social context of the inhabitants of the house. The first badge was unearthed in Level 6, Subunit 1. It was very thin and it appears to be missing its fastener, perhaps a safety pin, at the back. Made of brass, it was in good condition; the metal was not corroded. The phrases MELVIN PURVIS JUNIOR G-MAN CORPS appeared in four rows, decorated with a star; this was a child s toy badge. An image of this badge can be found in a manual for the Junior G-Man Club (see Appendix). Started in 1936, this club featured famed Federal Bureau of Investigation agent, Melvin Purvis, who left the Bureau in 1935 in pursuit of show business. Children could mail in two Post Toasties cereal box tops to receive enrollment in the Junior G-Man Corps, a manual of instructions, a badge, and a catalog of other prizes. The badge excavated was intended for boys, as girls received one that read Girls Division and came on a smaller winged version (Bailey n.d.). The ownership of a Junior G-Man badge represents participation in a craze of 1930s popular culture among American youth: many Junior G-Man clubs were organized by advertisers that were looking for marketing promotions (Theoharis et al. 1999:278). Therefore, it is likely that a young boy living at the Yates house sent in his Post Toasties cereal boxtops in order to get his Junior G-Man badge. This act not only indicates participation in keeping up with the popular trends in the entertainment 157

161 industry, but also the ability to access mass-produced commercial goods: in this case, Post Toasties cereal. And thus, the presence of the badge helps us think not only about issues of adornment, but also diet choices as related to consumption patterns. Moreover, this act suggests that Freedman s Town residents were participating in widespread consumer patterns: they, the target audience of the brand name cereal marketers. The second badge was located in Level 8, Subunit 2. It is a star-shaped badge that is also missing a fastener on the back, and some of the legs on the star are bent inward. The main body of the star is engraved in all capital letters: BAND OF MERCY, GLORY TO GOD, and PEACE ON EARTH GOOD WILL TO ALL are found around an image of a shining sun. Each of the five legs of the star have a different stamped word that, altogether, read KINDNESS JUSTICE MERCY TO ALL. Founded July 28, 1882, the Band of Mercy organization offered free membership, once a pledge was signed that promised good will toward animals: I will try to be kind to all harmless living creatures, and try to protect them from cruel usage (Our Dumb Animals 1889:75). A newspaper publication boasts of its wide variety of readers: old and young rich and poor white and black Protestant, Roman Catholic, and Jew (Our Dumb Animals 1889:43). That the periodical explicitly states African-Americans as readers leads to the possibility of the affiliate organization, Band of Mercy, as a catalyst for congregation and socialization within the immediate community in Freedmen s Town; this young town developed identity by utilizing organizations such as Band of Mercy to facilitate familiarization with fellow citizens, and the accompanying ritual acts (comparable to going to a local church on Sundays) probably contributed in some way indirectly to their growth as a community. 158

162 Members could obtain an imitation gold badge such as the one found in Unit YC by either forming a Band of Mercy with at least thirty members of adults or children (and leading as President of the Band), or purchasing the badge for eight cents. The newspaper states that any intelligent boy or girl fourteen years old can form a Band with no cost, suggesting that the wearer of the excavated badge might have been a young person. However, a college supplement found in the December 1888 issue of Our Dumb Animals also provides the possibility of members being young adults. In an address before the Congress of Educators in 1885, the president of Band of Mercy, Geo. T. Angell, announced that the organization has now from the Gulf of St. Lawrence to the Gulf of Mexico, and from the Atlantic to the Pacific coast, over six thousand two hundred branches with probably over half a million members (Our Dumb Animals 1889). The owner of the badge at the Yates house was thus likely involved in social movements much bigger than the borders of their neighborhood, city, or even state. The Bands of Mercy met at regular weekly or monthly times to socialize and act as a charitable, humane member of society. This ritual of meeting and wearing the star badge legitimizes a particular identity as socially-conscious and connected, both to themselves and others. Garment A garment of clothing, a woman s slip with lace trim, was recovered in Level 5, Subunit 2. It was fully intact, although in far from perfect condition it had two long, unbroken tears (as if cut with scissors) from the underarm area down, next to the side seams. Many roots were carefully pulled or cut out of the lace, but the lace itself wasn t torn. The fabric was uniformly stained a dirty off-white, probably from being in the 159

163 ground, along with several rust-colored stains, probably from its close proximity to some corroded metal artifacts in the unit. The cloth material had a smoother-than-average feel almost like sateen or silk, but coarser, perhaps from weathering in the elements. The lace trim was used as an indicator for a terminus post quem date; in the Sears and Roebuck catalog, the lace trim does not appear on slips until the Spring 1930 issue where it is called the princess slip. Even then, it is not until the Spring 1933 issue that more than one style of lace slip is offered on the two to three pages with slip advertisements. By the Spring 1934 issue, there was a large variety of lacy slips, and it seemed that the lacy slip was becoming very popular (Sears and Roebuck ). Since then, lace on a slip has been a familiar feature that has never gone out of style. The catalogs offer slips in a variety of materials, including: rayon taffeta, rayon satin (with or without silk), sateen, or rayon or cotton flat crepe. Based on the appearance of each of these materials, the lace slip is likely made out of rayon taffeta. The sleek texture and slight sheen of taffeta were observed in the artifact. Taffeta was considered a high-end purchase in the 1930s; depending on the time when this slip was actually worn, its wearer could have been aiming to contribute to their self-identity with the donning of this garment. The remaining factors for selecting a particular item for one s personal agenda, then, are left: construction of social, gender, and class identity. Social identity is constructed from the wearer s intent to elaborate a plain garment, some would say, frivolously. The act of decorating something naturally calls attention to it, and in this time period, it could have been an especially sensitive thing to draw attention to. Gender identity is constructed from the donning of the slip, as it aimed to make the wearer feel more feminine. Back then and even now today, women s lacy, frilly undergarments contrasted from men s 160

164 and children s plain undergarments and suggested more femininity with its contrasting delicate details. Last, class identity was constructed from the purchase of the slip, as the material it was made from ranged in price from inexpensive and practical to high-end and luxurious. The slip could confirm class identity to its owner, as they practiced the ritual of buying the types of consumer goods that catered to their economic ability. Concluding remarks The study of the three units at Yates house aimed to glean information from the recovered artifacts, shedding light on various aspects of the inhabitants: as consumers of material culture, as participants in historical trends, and thus as purposeful creators of their own identities. Albeit a small sample size, the adornments signify conscientious choices made over time by the people residing in Yates house. The comparison of historical accounts of trends in material consumption with the artifacts that actually surfaced from the unit helps archaeologists to understand the introduction, accessibility, and popularity of adornments the historical accounts are substantiated, and the choices made by the consumer are given more context and richness to the inferences that can be drawn about the self-prescribed identity. The shift in materials recovered, from ocean to freshwater shell and from shell to vegetable ivory to plastic, reflects the current events of the time period and supports the historical accounts of such trends for the button industry. Similarly, the plastic buttons found that correspond to the Art Deco movement reflect the popular styles of the time, as well as enabling the consumer the freedom to express themselves and strongly convey their self-identity with the bold fashion choices available. 161

165 The phenomenon of branding surfaces in the unit as an obvious tool of identity, and show what elements are of value to the consumer. For example, a button imprinted with tiny words from Carhartt work clothing shows the consumer s concern for durability, quality, and brand-name value in their clothing. The imprinted metal buttons came from work clothing that, for the first time, blatantly allowed identity to be directly inscribed to material culture and interpreted by others. It is interesting to observe this trend of branding expand to what it is today, with nearly every good used in modern society to have association with a brand name. The small imprinted button accents from the past have been magnified to where certain clothing symbols are recognized immediately and associated with a certain degree of quality and sometimes even luxury, with the wearer s intentions of self-identity very apparent. The small artifacts found in the discards of society can serve as an honest, unscathed account of the consumer s intentions of use and identity. Free from any slant as are historical documents, they remain what they were the moment they left the consumer s hands: tiny trinkets of truth cast out that illustrate who the consumers were and how they chose to identify themselves. References Cited Albert, Lillian Smith and Kent, Kathryn 1949 The Complete Button Book. Garden City: Doubleday & Company, Inc. Aultman, Jennifer, and Grillo, Kate 2003 Digital Archaeological Archive of Comparative Slavery (DAACS) Cataloging Manual: Buttons. Electronic document, accessed April 6, Bailey, Reno, ed. n.d. The Saga of Melvin Purvis. Electronic document. 162

166 accessed May 1, Beeth, Howard 1992 Introduction. In Black Dixie: Afro-Texan history and culture in Houston. Howard Beeth and Cary D. Wintz, eds. College Station: Texas A&M University Press. Boyle, Robert D Invisible Houston: the Black experience in boom and bust. College Station: Texas A&M University Press. Claasen, Cheryl 1994 Washboards, pigtoes, and muckets: Historic musseling in the Mississippi watershed. Historical Archaeology 28(2): Cook, Lauren J., Yamin, Rebecca, and McCarthy, John P Shopping as Meaningful Action: Toward a Redefinition of Consumption in Historical Archaeology. Historical Archaeology 30(4): Díaz-Andreu, Margarita, and Lucy, Sam 2005 Introduction. In The Archaeology of Identity: Approaches to gender, age, status, ethnicity, and religion. Pp New York: Routledge. e-bay 2009 Vintage workwear: Vintage advertisment, vintage clocks, vintage store displays. Electronic image, W0QQ_armrsZ1, accessed April 1, Funk & Junk Vintage Clothing & Collectibles 2009 Funk & Junk Collectibles and Vintage Clothing ecrater Store. Electronic image, accessed May 1, Grillo, Kate, Aultman, Jennifer, and Bon-Harper, Nick 2003 Digital Archaeological Archive of Comparative Slavery (DAACS) Cataloging Manual: Buckles. Electronic document, accessed April 28, Heath, Barbara J Buttons, Beads, and Buckles: Contextualizing Adornment Within the Bounds of Slavery. In Historical Archaeology, Identity Formation, and the Interpretation of Ethnicity. Maria Franklin and Garrett Fesler, eds. Pp Williamsburg: Colonial Williamsburg Foundation. 163

167 Heath, Barbara J., and Bennett Amber 2000 The little Spots allow d them : The Archaeological Study of African- American Yards. Historical Archaeology 34(2): Hughes, Elizabeth, and Lester, Marion 1993 The Big Book of Buttons. 2 nd edition. Sedgwick: New Leaf Publishers. Intermountain Antiquities Computer System (IMACS) 1990 Intermountain Antiquities Computer System (IMACS) User s Guide, Instructions, and Computer Codes for Use with The IMACS Site Form: Buttons. Prepared by the University of Utah, Bureau of Land Management, and the U.S. Forest Service of Salt Lake City, Utah. Electronic document, accessed April 23, Lee, Lori 2008 Consumerism, Social Relations, and Slavery at Late Antebellum Poplar Forest ( ). Paper presented at the Annual Meeting of the Council for Northeast Historical Archaeology in St. Mary s City, Maryland, October 25. LeeDecker, Charles H Historical Dimensions of Consumer Research. Historical Archaeology 25(2): Mullins, Paul R Race and Affluence: An Archaeology of African America and Consumer Culture. Charles E. Orser, Jr., ed. New York: Klewer Academic / Plenum Publishers Race and the Genteel Consumer: Class and African-American Consumption, Historical Archaeology 33(1): Our Dumb Animals 1889 Extracts from address of Mr. Angell before the International Congress of Educators, at New Orleans, Louisiana, Feb. 26, Volume 21, Number 7. December College Supplement to Our Dumb Animals Cambridge: Massachusetts Society for the Prevention of Cruelty to Animals. Psota, Sunshine 2002 Boss of the Road: Early 20 th Century Consumer Selections of Work Clothing from Alabama Gates Work Camp, Owens Valley, California. Historical Archaeology 36(4): Sears and Roebuck Sears and Roebuck Catalogue. Microfilm on file at the Fondren Library, Rice University, Houston. 164

168 Scott, Emmett J The Red Book of Houston: A Compendium of Social, Professional, Religious, Educational and Industrial Interests of Houston s Colored Population. Houston: Sotex Publishing Company. Singleton, Theresa A., ed Introduction. In I, Too, Am America : Archaeological Studies of African American Life. Charlottesville: University Press of Virginia. SoRelle, James M The Emergence of Black Business in Houston, Texas: A Study of Race and Ideology, In Black Dixie: Afro-Texan history and culture in Houston. Howard Beeth and Cary D. Wintz, eds. College Station: Texas A&M University Press. Sprague, Roderick 2002 China or Prosser Button Identification and Dating. Historical Archaeology 36(2): Stine, Linda France, Cabak, Melanie A., and Groover, Mark D Blue Beads as African-American Cultural Symbols. Historical Archaeology 30(3): Theoharis, Athan G., Poveda, Tony G., Rosenfeld, Susan, and Powers, Richard Gid, eds The FBI: A Comprehensive Reference Guide. Phoenix: Oryx Press. White, Carolyn L American Artifacts of Personal Adornment, : A Guide to Identification and Interpretation. New York: Rowman & Littlefield Publishers, Inc. Wintz, Cary D Blacks in Houston. Houston: The Development of an International City. Fred R. von der Mehden, ed. Houston: Houston Center for the Humanities. Yentsch, Anne Elizabeth 1994 A Chesapeake Family and Their Slaves, A Study in Historical Archaeology. New York: Cambridge University Press. 165

169 Unit YC Unit D Unit P Total Ceramic Glass Metal Rubber Ecology Total Table 1: Total counts for each button material: Units YC, D, and P. Figure 1: Adornment material breakdown: Unit YC. Each material class is shown in proportion to the whole sample of 42 artifacts. 166

170 Figure 2: Adornment type breakdown: Unit YC. Each type of adornment is shown in proportion to the whole sample of 42 artifacts. Figure 3: Frequency of adornment types across time: Unit YC. The count for each adornment type is shown in proportion to the whole sample within each bar line. The whole bar line represents the total number of artifacts found in each level. 167

171 Figure 4: Frequency of adornment materials across time: Unit YC. The count for each material class is shown in proportion to the whole sample within each bar line. The whole bar line represents the total number of artifacts found in each level. Figure 5: Frequency of button materials across time: Unit D. The count for each material class is shown in proportion to the whole sample within each bar line. The whole bar line represents the total number of artifacts found in each level. 168

172 Figure 6: Frequency of button materials across time: Unit P. The count for each material class is shown in proportion to the whole sample within each bar line. The whole bar line represents the total number of artifacts found in each level. Figure 7: Shell button submaterial usage across time: Unit YC. The counts for freshwater and ocean shells are shown in proportion to each other within each bar line. The whole bar line represents the total number of artifacts found in each level. 169

173 Figure 8: Shell button submaterial usage across time: Unit D. The counts for freshwater and ocean shells are shown in proportion to each other within each bar line. The whole bar line represents the total number of artifacts found in each level. Figure 9: Shell button submaterial usage across time: Unit P. The counts for freshwater and ocean shells are shown in proportion to each other within each bar line. The whole bar line represents the total number of artifacts found in each level. 170

174 Figure 10: Frequency of button materials across time: Unit YC. The count for each material class is shown in proportion to the whole sample within each bar line. The whole bar line represents the total number of artifacts found in each level. Figure 11: Metallic buttons implemented in button analysis. Top row from left: copper with shank: Unit YC, 5-2; pewter collar button: Unit YC, 6-2; cast oval copper with decorative inset: Unit D, 4-3. Middle row: cast iron: Unit D, 6-4; cast iron/steel: Unit D, 7-2; cast enamel/tin: Unit YC, 5-2; molded iron/steel: Unit D, 7-1. Bottom row: cast copper with bird-like design: Unit D, 4-2; cast copper with curlicue design: Unit D, 6-2; cast copper with raised surface: Unit D,

175 Figure 12: Decorative buttons implemented in button analysis. Top row from left: spherical faceted black glass with wire eye: Unit D, 7-2; spherical faceted black glass with wire eye: Unit P, 7-1; spherical faceted clear glass with wire eye: Unit D, 8-4; white plastic with rope motif: Unit P, 4-2. Middle row: square faceted red plastic with drilled eye: Unit P, 5-2; square faceted red plastic with drilled eye: Unit P, 4-4; faceted red plastic with drilled eye: Unit D, 2-1; parallelogram-shaped faceted black plastic: Unit P, 6-4. Bottom row: black glass with fan/shell motif: Unit D, 7-2; oval black glass with flower motif: Unit D, 5-4; cream plastic: Unit YC, 2-2; cream plastic: Unit D,

176 Figure 13: Detailed metal badge. Gold-colored bronze with black inscription: MELVIN PURVIS/JUNIOR G-MAN CORPS : Unit YC,

177 Figure 14: Significant metallic findings. Clockwise from top left: cast/stamped brass button with inscription Burlington Overall in square outline: Unit YC, 4-3; star-shaped brass/copper badge with inscription: BAND OF MERCY/GLORY TO GOD/ KINDNESS JUSTICE MERCY TO ALL/PEACE ON EARTH/GOOD WILL TO ALL/MSPCA : Unit YC, 8-2; gold-colored bronze badge with inscription: MELVIN PURVIS/JUNIOR G-MAN CORPS : Unit YC, 4-3; cast/stamped brass button with inscription CARHARTT S in heart outline: Unit D, 6-1. Figure 15: Detailed Burlington Overall brass button. Unit YC,

178 Figure 16: Beads for Unit YC. Top row from left: five purple glass beads fused in a row: Unit YC, 6-2; green glass seed beads fused in rocky matrix: Unit YC, 6-2. Bottom row: white Prosser molded bead: Unit YC, 5-4; light blue glass bead fragment: Unit YC, 7-2; modern orange spherical plastic bead: Unit YC,

179 Figure 17: Prosser-molded ceramic buttons implemented in button analysis. First row from left: Unit P, 4-1; Unit P, 5-2; Unit P, 7-3; Unit P, 7-3. Second row: Unit P, 8-3; Unit P, 9; Unit YC, 1; Unit YC, 8-1. Third row: Unit D, 2-4; Unit D, 3-1; Unit D, 3-2; Unit D, 6-2. Fourth row: Unit D, 6-4; Unit D, 6-4; Unit D, 7-4; Unit D, all levels (wall fall). 176

180 Figure 18: Ecology buttons implemented in button analysis. Top row from left: cut/carved bone: Unit P, 2; bone: Unit P, 3-4; Unit P, 3-1; blue chalky burnt shell: Unit YC, 7-1. Bottom row: cut/carved vegetable ivory: Unit P, 2; cut/carved vegetable ivory: Unit P, 3-4; cut/carved vegetable ivory: Unit D,

181 Figure 19: Freshwater and ocean water shell buttons and buckle. First row is ocean water shell material, all other rows are fresh water. First row from left: Unit YC, 6-1; Unit YC, 7-2; Unit YC, 7-2; Unit YC, 9-2; Unit YC, 10-2; Unit P, 7-4. Second row: all Unit YC, 4-2. Third row: Unit YC, 5-2; Unit YC, 5-2; Unit YC, 5-2; Unit YC, 5-2; Unit YC, 4-1. Fourth row: Unit P, 3-1; Unit P, 3-1; Unit D, 3-2; Unit D, 3-4 (indeterminate material); Unit D, 6-2; Unit D,

182 Figure 20: Advertisement sign for Burlington Overalls (image from e-bay, 2009). (attached PDF) Figure 21: Junior G-Man Corps Prize Catalog included in Post Toasties Cornflakes boxes. (attached PDF) Figure 22: Microfilm image captures of Sears Roebuck and Company catalogs. The following dates show similarities and differences as well as emerging trends about women s slips through time. Page 1: Spring Page 2: Spring Page 3: blank. Page 4: Spring Page 5: Fall Page 6: Spring Page 7: Fall Page 8: Fall Page 9: Page 10: Fall Page 11: Spring Page 12: Fall

183 180

184 Miscellaneous Artifacts and Building Materials Nadine Boudreaux Introduction Unit YC yielded a good number of miscellaneous artifacts, ranging from candy wrappers to roofing slate to baseballs. Most of the artifacts tend to be modern and concentrated in the upper levels with the most common materials being wrappers, threads, plastics, and lithics. The other artifacts include coins, marbles, light bulbs, a pen, typeface, baseball, pull tabs, wires, and blades. After Level 5, the number and type of artifacts changes dramatically. While in Levels 1 through 4 there is a wide variety of modern materials, Levels 6 through 12 yielded various lithics, an eggshell, one piece of hard plastic, a metal wire, and two unidentifiable artifacts (possibly plastic). Building material, however, does not dramatically drop off until Level 8. Consisting of brick and concrete, it remains a fairly constant presence throughout the first seven levels. Wrappers A variety of wrappers were excavated in the upper levels of the unit (Levels 1 through 2) except for one wrapper in Level 5. All are either plastic or aluminum (except for one paper wrapper in Level 2) and most seem to be candy wrappers as evidenced by the bright colors, cartoonish drawings, and occasional informative writing. One fragment contained the letters CHOC which probably expanded into CHOCOLATE, one wrapper asked What happened to the wind? with the answer to the probable joke unfortunately lost, and another holding the familiar candy company s name BRACH S. And one wrapper ironically admonishes Don t be a litter bug. Keep America Beautiful. One aluminum wrapper was found in all three upper levels, green with small silver spots. While probably not belonging to just one wrapper, it attests to the relative popularity of the unknown item. Threads Twenty-two threads were excavated in the first four levels of the unit. A common thread was a thin, brown, tightly twined fiber sometimes still tightly held together and 181

185 sometimes more loosely frayed. Other threads include brown, red, and purple fibers, many frayed and loosely held together, and one white thread braided like a shoelace knotted together with a similar red and fraying thread. Hard Plastic The first five levels of the unit contain quite a number of hard plastic pieces of varied colors, sizes, and widths; only one fragment was found in Level 11. However, all but one of the plastic fragments from Level 5 were black rubber-like pieces, not the brightly-colored plastic of the upper levels (see Figure 1). This reveals a particular concentration of the black rubbery material in Level 5 (plus an isolated fragment in Level 2), and places most of the colorful, possibly more modern plastic in the upper levels (mostly 1 through 3 with one each in Levels 4, 5, and 11). Most fragments of plastic are unidentifiable as to their original form; a few boast highly defined ridges, and a few are complete objects whose function can still not be determined. Perhaps the most interesting artifact in this category was found in Level 3 and is what looks like a plastic game piece from the Ideal Toy Company (DPJohnson ), founded in 1907 as a major producer of dolls (Wikipedia, Ideal Toy Company 2009). It seems that it wasn t until the early 60 s that most of Ideal s board games were produced. The artifact is white and square with lipped edges and two protruding knobs on two opposite edges. On one side is their logo in raised letters, Ideal, and the other side shows an arrow. It is very likely a board game piece. Lithics About thirty-seven rocks were saved from the excavation, 17 of which could not be identified. Of the ones that could be identified, 5 were quartz, 1 was flint/chert, and 14 were slate. The flint/chert could have been used in tool-making as evidenced by the way it is cut. Of the slate fragments, two are unmodified and the rest were probably used as roofing material. The lithics are spread throughout the first nine levels with no particularly revealing concentration except for the roofing slate which only appears in Levels 7 through 9 (see Figure 2). Perhaps a nearby roof was being worked on during the 182

186 time period of those levels, although the small amount of slate does not attest to a substantial amount of activity. Random Objects A bottle cap was found in level two with the brand Olde English 800 a malt liquor manufactured by the Miller Brewing Company. The company itself has been around since the mid-1800s, but Olde English 800 began production in 1964 (SABMiller 2009). Three coins were also found: a 1971 penny in Level 2, a 1970 penny in Level 3, and a buffalo nickel in Level 5 (Figure 3). A mother-of-pearl handle was excavated in Level 4 (Figure 4), and a ceramic disk in Level 1 used for unknown purposes (Figure 5). A floating fish knife (as engraved on the knife itself) was found in Level 1 and bears a striking resemblance to a knife made by the Sherwin-Williams Company being sold online (WorthPoint Corporation, Floating Fish Knife From Sherman 2009). It also resembles this knife being sold online (WorthPoint Corporation, VINTAGE Japan figural fish fishing filet floating KNIFE 2009), but no information is given on its provenance or date of manufacture. Another particularly interesting blade is a straight shaving razor found in Level 5 (Figure 6). The object includes the metal blade and a plastic sheath hinged to one end and painted dark gray. The metal has rusted, rending the blade impossible to reconstruct. A baseball was also found in Level 4 (Figure 7) highly deteriorated with most of the cork inside exposed. The leather exterior had degraded and pieces were found in the vicinity and in the above layer (Level 3) of the same subunit. Three small light bulbs were found in Levels 2, 3, and 5, and three marbles were found in Levels 3 and 5. The two marbles excavated from Level 5 were found in the same subunit (Subunit 2) one complete and one only partial (Figure 8). Three pull tabs were found in Levels 2 and 3, two of them in the same area (Level 3, Subunit 3). The pull tab for aluminum cans was invented in 1962 and continued in use until the stay tab replaced it in 1975 (Wikipedia, Beverage Can 2009). Six fragments of siding were found scattered evenly in the first four levels of the unit, which is not enough to make any conclusive statement about where it might have come from (if it fell from a nearby structure or was thrown out), but does indicate that it was, at the very least, used nearby. A carbonized rod, 0.2 inches in diameter and 183

187 probably used in a battery, was found in Level 2. But perhaps the most interesting find was a blank metal typeface from Level 4 (Figure 9). It may be related to the occupation of the house by Rutherford B. Yates, as he was a printer. He moved into the house in 1912, and started the Yates Printing Company in 1922 which lasted until 1978 (Markeroni 2009). Building Materials Concrete and brick fragments were all collected and weighed throughout the excavation of Unit YC, and some were kept for analysis. The highest concentration of building material occurs in Level 4, and dramatically drops off after Level 7 (Figure 10). Four whole and almost-whole bricks were recovered from the unit in Levels 3, 4, 7 and 8. They appear to be hand-made, and the largest one is inscribed with two letters: either A, V, or Y oriented longways and R oriented widthwise. Another brick (and several of the fragments) have highly defined ridges on one side. References The Good Games of the Ideal Toy Corporation. Electronic document, accessed April 28, Markeroni 2009 #11691 Rutherford B. Yates House. Electronic document, mber=screwtape&which=singlesnarf, accessed April 28, SABMiller 2009 SABMiller Brands. Electronic document, accessed April 28, Wikipedia, The Free Encyclopedia 2009 Ideal Toy Company. Electronic document, accessed April 28, Wikipedia, The Free Encyclopedia 2009 Beverage Can. Electronic document, accessed April 28, WorthPoint Corporation 184

188 2009 Worthopedia Price Guide. Electronic document, accessed April 28, WorthPoint Corporation 2009 Worthopedia Price Guide. Electronic document, accessed April 28,

189 Figure 1 Figure 2 186

190 Figure 3 Figure 4 187

191 Figure 5 Figure 6 188

192 Figure 7 Figure 8 189

193 Figure 9 Figure

194 Appendix: Catalogue of Miscellaneous Artifacts Lot Lvl Subunit Material Submaterial Art. Class Qty. Wt (g) Comment I CL White circular disk V LI A white piece of plastic adhered to a black piece of G&RR* IV RA <0.1 rubber Bottom left corner of an aluminum wrapper; white with green writing giving the weight on one side and ingredients on G&RR VII OF <0.1 the other G&RR VII OF <0.1 Aluminum wrapper with pink and blue print G&RR VII OF <0.1 Clear piece of plastic wrapper G&RR VII OF <0.1 Pink fragment G&RR VII OF <0.1 White piece of plastic grocery bag G&RR VII OF <0.1 White (rubbed clear in places) pieces of aluminum wrapper. Pink printing in places NA III MA Engraved: Stainless steel Japan / Floating Fish Knife NA III MF <0.1 Possibly a sheath for a small razor NA IV RD <0.1 Twist cap 191

195 NA IV RD <0.1 two green fragments, and two white fragments NA IV RD <0.1 Green plastic and gray plastic NA IV RD <0.1 Blue plastic with ridges NA IV RD < NA V LB Two pieces of quartz NA V LI Indeterminate lithic NA VI EJ <0.1 Small seed NA VII OE <0.1 Brown and red fraying fibers NA VII OF <0.1 Siding tile (possibly asbestos) NA VII OF <0.1 Probable candy wrapper; green with small silver spots NA VII OF <0.1 Silver aluminum III MB Penny; year is III MI Twist-off bottlecap, says "Olde English Brand '800'" IV RD <0.1 Four hard plastic fragments: one dark green with ridges, one pink, one light green, one white long and thin IV RD <0.1 Cream-colored thin piece of plastic IV RD <0.1 Light brown VI EB A modified piece of leather, folded over and sewed together; etched triangular design VII OE <0.1 One brown, tightly twined, thin; one brown unravelling 192

196 VII OF Siding tile (possibly asbestos) VII OF <0.1 Probable candy wrapper; green with small silver spots VII OF <0.1 Probable chocolate wrapper; can read "CHOC" on the fragment VII OF <0.1 Probable candy wrapper; white with small blue spots VII OF <0.1 Probable candy wrapper; white with colorful drawings and letters VII OF <0.1 Strip of black wrapper plastic VII OF <0.1 Probable candy wrapper; brown with printing; on one side says: "What happened to the wind?" and "Answers ut " VII OF <0.1 Yellow plastic, ribbed ribbon like one used for decorating presents IV RA <0.1 Rubber band IV RB Possible car part IV RD <0.1 Decorated small fragmet of cream-colored plastic IV RD <0.1 Black IV RD <0.1 White/clear small plastic ring IV RD <0.1 Two pieces of vibrant blue plastic, one with ridges on one side VII OF Siding tile (possibly asbestos) 193

197 VII OF <0.1 Probable candy wrappers; two green pieces with small silver spots, two pieces of plain alumium, one piece from CHOC wrapper VII OF <0.1 Probable candy wrapper; green with small silver spots VII OF <0.1 Says "Don't be a litter bug. Keep America Beautiful" VII OF <0.1 Miscellaneous wrappers IV RD Miscellaneous pieces of hard plastic VII OF <0.1 One of the pieces says "BRACH'S" III MF <0.1 Probable light bulb piece; twisted wire around a painted red unidendifiable piece on one end, and a metal shard on the other III MF <0.1 Pull tab from Coke can V LJ Carbonized rod VII OE <0.1 Purple very thin (almost wiry) twined, crinkled thread VII OF <0.1 Probable candy wrapper; green with small silver spots VII OF <0.1 Probable candy wrapper; green with small silver spots VII OF < NA IV RD <0.1 Two identical pieces of white plastic joined together and 194

198 curved at the end NA VII OF <0.1 Two pieces of red plastic wrapper II GK Green, purple, and blue marble fragment IV RD <0.1 Most of a label that says, "Quality SERCO Remanufactured Engines" IV RD <0.1 Green plastic fragments and a cream-colored VI EF <0.1 Shell painted white and red VII OE <0.1 Brown, tightly twined, thin VII OE <0.1 Brown, tightly twined, thin VII OF <0.1 Small piece of wrapper - white with a little pink II GA Glass light bulb III MH <0.1 Circular thin piece of metal IV RD <0.1 Very thin pieces of clear plastic with red paint - most of which has flaked off VI EF <0.1 Shell with green paint VII OE <0.1 Brown, twined fiber, crinkled VII OF Siding tile (possibly asbestos) VII OF <0.1 Crunched piece of silver aluminum I CL Oval shape with backing and two slits III MB Penny; year is

199 III MF <0.1 Pull tab III MF <0.1 Pull tab IV RD <0.1 Long, rectangular strip of hard plastic IV RD <0.1 Small, white, looks like a flattened plastic screw IV RD <0.1 Green hard plastic fragment IV RD <0.1 Black hard plastic fragment IV RD <0.1 Tan thin plastic fragment IV RD Plastic game piece; white, square with protruding "knobs"; says "Ideal" on one side and has an arrow on the other side IV RD <0.1 Green IV RD <0.1 Light brown V LI Light pink lithic with leaf etched into one side V LI Indeterminate lithics VI EB <0.1 Leather pieces from the baseball VI EF Shell with red paint VII OE <0.1 Brown, tightly twined, thin VII OF Siding tile (possibly asbestos) VII OF <0.1 Probably candy wrapper; pink VII OF <0.1 Probable candy wrapper; green with small silver spots Crunched piece of silver VII OF <0.1 aluminum 196

200 Yellow pieces of a plastic wrapper VII OF < VII OF <0.1 Green wrapper Yellow pieces of a plastic wrapper; one piece says "Kraft" VII OF < VII OF <0.1 Small red paint chip VII OF <0.1 Small red paint chip IV RD Clear hard plastic Piece of flint/chert that could have been used in tool-making V LA V LB One piece of quartz A blue and a yellow tightly twined, stiff VII OE < VII OE <0.1 Brown, twined fiber, crinkled VII OE <0.1 Brown, tightly twined, thin VII OE <0.1 Brown, tightly twined, thin VII OF Siding tile (possibly asbestos) VII OF <0.1 Miscellaneous wrappers VII OF <0.1 Small green paint chip III MI Blank typeface VI EF Mother-of-pearl handle VII OE <0.1 Brown, twined fiber, crinkled Leather pieces from the VI EB baseball VII OE <0.1 Brown, twined fiber, crinkled VII OE <0.1 White shoelace-like rope (very frayed) tied into a knot intertwined with red "rope" (braided much like the white 197

201 piece) VII OG Baseball - small pieces remain of the leather outside, the rest is the green rubber/cork inside VII OF <0.1 Small red paint chips III MB? Buffalo nickel IV RD Black VII OG What looks like a cylinder of cement with some metal covering one end and wrapped in thin plastic II GJ Yellow marble II GK White and orange marble fragment III MI Pen IV RD <0.1 Black tiered circular plastic IV RD One slightly-curved triangleshaped; one with a lipped edge; both black IV RD Black IV RD Black V LB pieces of quartz V LH <0.1 Probably roofing slate V LI Unidentifiable lithics VII OG Indeterminate material in three layers, one of the layers being a green composite that was found 198

202 on some shells and the baseball fragments II GA Small light bulb III MI Insulated wire V LI V LI Indeterminate lithics VII OF <0.1 Aluminum wrapper III MA Shaving razor (with plastic sheath) III MF <0.1 The metal portion of a pencil eraser IV RD Black IV RD Black VII OF Some green composite covering some rock IV RD Brown cylinder with screw ridges inside IV RD Plastic body with two metal screws at the top V LH Piece of slate V LI <0.1 Indeterminate lithic VI EJ Looks like wood IV RB <0.1 Looks like vulcanized rubber V LH Piece of slate V LH Probably roofing slate V LH <0.1 Probably roofing slate V LI Indeterminate lithic 199

203 III MI Metal wire V LH Probably roofing slate V LI Large indeterminate lithic V LH Probably roofing slate V LH <0.1 Probably roofing slate V LH Probably roofing slate IV RD <0.1 Dark green plastic fragment ** 2 VI EA <0.1 White *G&RR: Grass and Root Removal **Shovel Testpit Catalogue of Building Materials Lot Lvl Subunit Material Submaterial Art. Class Qty. Wt (g) G&RR* I CH NA VII OA I CH I CH < VII OA VII OA VII OA I CH I CH VII OA VII OA Comment Red concrete (probably from being between two bricks) 200

204 VII OA I CH VII OA I CH I CH VII OA VII OA VII OA I CH I CH I CH VII OA I CH I CH Large brick with ridges on one side Concrete that has been painted red VII OA VII OG Brick attatched to concrete I CH I CH I CH I CH < VII OA I CH VII OA I CH < I CH Large brick, not machine made 201

205 VII OA VII OA < VII OA I CH VII OA I CH I CH VII OA VII OA VII OA VII OA I CH < I CH I CH I CH < I CH VII OA VII OA I CH VII OA I CH > I CH I CH The large fragment has a piece of black rubber or tar adhered to one side Large brick, not machine made, inscribed with "A," "V," or "Y" oriented longways and "R" oriented widthwise 202

206 I CH VII OA I CH I CH VII OA I CH Brick with stucco on two sides? 9 2 VII OA I CH VII OA < I CH I CH wallfall I CH *G&RR: Grass and Root Removal. 203

207 204

208 Soil Chemistry Analysis at 1512 Wilson Street: An Examination of Chemical Distributions in a Historic Houselot Sarah S. Nouri ABSTRACT Geoarchaeology is a rapidly expanding field, and geoarchaeological methods and analyses are being more commonly applied at archaeological sites. One specific type of geoarchaeological analysis is soil chemistry analysis, and it is frequently used to understand the spatial organization of sites and to determine the locations of specific activity areas based on element concentrations. In this study of a historic houselot in Freedman s Town, Houston, Texas, soil chemistry analysis via ICP-AES (Inductively Coupled Plasma-Atomic Emission Spectroscopy) was used to analyze the concentrations of phosphorus, potassium, calcium, magnesium, sodium, zinc, iron, copper, manganese, sulfur, and boron. By plotting the concentrations on a map of the houselot using GIS, element distributions were found to correspond to expected activity areas. High element concentrations, specifically of calcium and phosphorus, were correlated with a possible driveway and trash pile. Therefore, activity areas were identified solely based on soil chemistry data, indicating that soil chemistry analysis via ICP-AES can be a rapid and inexpensive way not only to identify activity areas but also to determine which areas of a site would be most profitable to excavate. INTRODUCTION Geoarchaeological Methods Geoarchaeological analyses have been very commonly applied to archaeological sites in the past two decades (Goldberg and Macphail 2006:1). Geoarchaeology focuses mainly on sediments, which are deposited on the earth s surface at low temperatures and pressures, and not soils, which are formed by factors such as parent material, climate, time, relief, and biota. Soils are formed in situ, while sediments have been removed from a source and deposited at a site (Goldberg and Macphail 2006: 46). This type of research is very important in archaeology, because, as French (2003: 35) wrote in Geoarchaeology 205

209 in Action: to study artifacts without regard to their sedimentary matrix, their spatial arrangement, subsequent movement and alterations induced by post-depositional processes is to study only a fraction of the archaeological record. Types of geoarchaeological analyses and their applications vary and include shallow geophysical methods (resistivity, paleomagnetism, seismology), microscopic methods, mineralogy, thin section analysis, and soil chemistry analysis. One common soil chemistry analysis is that of phosphate, which was developed by Olaf Arrhenius in the 1920s when he discovered that elevated phosphate concentrations corresponded to prehistoric human occupation in Sweden (Goldberg and Macphail 2006: 344; Middleton 2004). Many human activities have noticeable impacts on sediment chemistry. For example, agriculture has an extensive impact on soil, while such activities as cooking and waste disposal have intensive impacts by depositing either liquid or solid residues on the soil, which can be incorporated into the sediment (Middleton 2004: 49). These residues are usually preserved in the soil (Middleton 2004). Phosphorus is a good indicator of human activity because it is a ubiquitous and very complex element, found in both organic and inorganic systems. In other words, it can be deposited in sediments by a myriad of human activities (Middleton 2004). Some areas in which high phosphorus levels have been found are agricultural soils, cemeteries, stock areas and drove-ways, camps, occupation mounds, and middens (Goldberg and Macphail 2006: 347). Despite the usefulness of phosphate analysis in geoarchaeological methods, many archaeologists have started to conduct multi-element tests, referred to collectively as soil chemistry, because the ubiquity of phosphorus suggests that though it is effective in indicating activity areas, it is not useful in differentiating between types of activities, especially when an area is used for several different purposes (Middleton 2004; Linderholm and Lundberg 1994). Therefore archaeologists have started to turn towards other elements in an attempt to elucidate the effects of various activities on soil (Middleton 2004). Though common patterns have been found between certain elements and specific activities, there are still differences among studies. A reason for this is that at many sites there is not only a spatial but also temporal organization of activities; therefore, different elements can overlap at different sites, resulting in the assignment of one chemical to various activities (Middleton 2004: 55). Nevertheless, soil chemistry has 206

210 been highly successful in site prospection, the determination of site boundaries, and the spatial organization of activities (Middleton and Price 1996: 673; Middleton 2004). Many recent, multi-element soil analyses have been used to identify activity sites at archaeological sites, to which I turn now. Case Studies In a prominent study, Middleton and Price (1996) used ICP-AES to determine activity areas in Oaxaca, Mexico and British Columbia, Canada. They used ICP-AES because it can simultaneously measure several elements and is also relatively inexpensive. The study sampled three areas: (1) a modern earthen-floored house, (2) archaeological samples from houses occupied from one to two thousand years ago, and (3) archaeological samples from uncovered surfaces. The sample from the modern house served as a reference profile to which they compared the data from the archaeological samples in order to determine the significance of the chemical signatures. Middleton and Price (1996: 675) did not use a strong acid for the extraction because they believed that a strong treatment could release elements that are bound in mineral lattices, which would interfere with the concentration of the mobile elements found in anthropogenic data. They found that separate areas could easily be distinguished on the basis of chemical composition, and that certain chemical elements corresponded to specific human activities. For example, middens, excrement, and organic debris are high in phosphorus, while wood ash contains high levels of potassium and magnesium. Their statistical analyses showed correlations between elements with close atomic weight and other chemical properties (Middleton and Price 1996: 675). For example, calcium and strontium were very closely correlated, possibly because they are both alkali earth metals, and therefore have the same number of valence shell electrons. Aluminum and magnesium, which were also strongly correlated, are very close in atomic weight, as are iron and manganese. In addition, they determined that soils below standing structures were protected from natural weathering processes, which reduced the rate at which element residues were removed (Middleton and Price 1996: 681). In a similar study, Linderholm and Lundberg (1994) used ICP-AES for soil chemistry analysis on a Late Bronze Age site in Sweden. The goals of this study were to 207

211 determine the relevance of elements besides phosphorus in determining activity areas, and to use a multivariate statistical analysis of elements to classify related soil samples. They found that a multivariate statistical analysis was very useful in differentiating between areas of human activity and areas that were not affected by humans. They also determined that other elements, specifically calcium, iron, copper, and zinc were more reliable in terms of spatial organization data because of phosphorus s ability to leach out of the soil (Linderholm and Lundberg 1994: 311). Unlike Middleton and Price, Wilson et al. (2008) claimed that it is necessary to use a strong acid rather than the mild one, because the anthropogenic signal may be retained in resistant soil fractions (Wilson et al. 2008: 414). Since this study, most researchers have used strong acids for soil analyses (e.g., Hjülstrom and Isaksson 2009). Nevertheless, Wilson et al. (2008: 418) also found that multi-element analysis alone could indicate areas of different function. They took samples from the hearth, kitchen, byre, midden, garden, arable fields, grazed-out fields, and off-site references of six small, abandoned farms in Scotland, England, and Wales. Since they already knew the activities occurring in each area, the goal of the research was to match element concentrations with the specific activity areas (Wilson et al. 2008). Several elements in this study were successfully associated with specific artifacts or human activities: phosphorus concentrations were higher in middens and organic matter, while zinc, copper, strontium, and calcium were associated with charcoal, bone, and mineral grains. Wilson et al. did warn that such processes such as general leaching or the deposition of minerals in the subsoil, podzolization (a process in which iron and aluminum leach through to the subsoil), and gleying (which occurs when the soil is completely saturated in water and reduces iron due to the restriction of oxygen), can affect anthropogenic sediments (Wilson et al. 2008: 413). Another study that used soil chemistry to identify human activity was that of Hjülstrom and Isaksson (2009) in an Iron Age village in Lejre, Denmark. They combined both element and lipid analyses for samples collected from a multifunctional dwelling house, a smithy, and a pit, and found similar results to Middleton and Price and Wilson et al. regarding how elements relate to specific activities. They found that charcoal aided in the retention of several elements, and that potassium and magnesium could be used to 208

212 identify hearths. Though Hjülstrom and Isaksson (2009: 176) used reference sites in their study, they mentioned that it is also possible to focus on spatial patterns rather than absolute values. Knudson et al. (2004) also conducted a soil chemistry analysis using ICP-AES. The goal of the research was to use element analysis to elucidate activity areas in fish camps in the Yukon-Kuskokwim Delta in the Arctic (Knudson et al. 2004: 443). Samples were taken from two modern fish camps, in both of which fishermen processed fish. One of the sites, however, had been occupied for thirty years, while the other had only been occupied for one year (Knudson et al. 2004). Activity areas at the sites included tents, smokehouses, hearths, drying racks, and processing areas. Their results were very similar to other studies, in that the camps showed distinct anthropogenic signatures based on the element analysis (Knudson et al. 2004: 452). However, unlike other studies, they did not find elevated phosphorus, potassium, or magnesium levels at the hearths. In addition, the authors suggest that the cold climate resulted in elevated levels of all of the elements; cold climates prevent weathering while the permafrost ensures a permanently moist soil (Knudson et al. 2004: 449). The interesting result of this research is that even the site that had only been occupied for one year showed different element concentrations than the control samples. Several other studies have found similar patterns in element concentrations and distributions using ICP-AES, and have also been able to distinguish between activity areas and uninhabited spaces (Entwistle et al. 2000; Huston et al. 2007). Wood ash, human waste, and decaying plant and animal materials have been found to affect soil element concentrations (Shahack-Gross et al. 2004; Knudson et al. 2004; Entwistle et al. 2000; Hutson et al. 2007). Most of the studies discussed have focused on sites that were occupied at least one thousand years ago. One interesting study by Sullivan and Kealhofer (2004) takes place at a site from colonial Virginia, and therefore applies soil chemistry techniques to more recently deposited sediments (from the seventeenth century AD). The purpose of the study was to learn more about the domestic economy of the late seventeenth century, and to determine whether the local elites relied on a mixed subsistence economy or simply on the tobacco industry (Sullivan and Kealhofer 2004: 1559). They used both phytolith analysis and soil chemistry, but unlike other studies, they did not use reference 209

213 samples. Because of the transformed landscape, there was no area from which a reference sample could be taken; therefore, control samples, taken from undisturbed areas far from the study site, were used for relative comparisons (Sullivan and Kealhofer 2004: 1662). Thirty-three samples were taken from features between buildings and fenced garden areas (Sullivan and Kealhofer 2004: 1662). High phosphorus concentrations were correlated with kitchen-related activities and possibly livestock (Sullivan and Kealhofer 2004: 1671). A significant conclusion that they drew is that calcium carbonate is an important component of ash and oyster shells, which were used for both paving and consumption (Sullivan and Kealhofer 2004). This is similar to a result from a study conducted by Entwistle et al. (2000) in the Isle of Skye stating that more calcium was found wherever there was shell sand, bone, and fish refuse. Like Sullivan and Kealhofer s research, the current study focuses on an area of relatively recent occupation in Freedman s Town, in the Fourth Ward in Houston, Texas. Project Background: Freedman s Town Freedman s Town, which is the only freedman s town in the country that is still occupied by descendants of its original founders, was listed as a historically significant historic district on the National Register of Historic Places in 1984 (McDavid et al. 2008: 37). The Fourth Ward has been described as a historic residential inner city community, lightly mixed with commercial, religious, and public facilities and as the mother ward for African Americans (Feit and Jones 2007: 13). Thousands of African Americans moved to the area at the end of the Civil War in 1865, and the racism of the Jim Crow era made Freedman s Town like a city within a city (Feit and Jones 2007: 1). One prominent founder of Freedman s Town was John Henry Jack Yates. Jack Yates was born in Gloucester County, Virginia but moved to Texas when his wife s slaves masters moved (Yates 1985). After emancipation, he and his family moved to Houston, where Yates became an ordained minister. He also began purchasing land in the Fourth Ward on Andrews Street in 1869 in the hopes of creating a community for African Americans, and built his own house at 1318 Andrews Street (Yates 1985; Bryant 1994). Eventually, the Fourth Ward became a very densely-populated, African American area. Jack Yates also helped found a school, Houston College (later Texas Southern 210

214 University) in 1885, and built Bethel Church, which is still standing, on the corner of Andrews and Wilson Street (Yates 1985; Feit and Jones 2007). Jack Yates had a significant influence on Freedman s town, both in founding it and in creating a true sense of community by, for example, hosting community-wide Juneteenth celebrations (Yates 1985). Residents of Freedman s Town described it as a nurturing neighborhood (Feit and Jones 2007: 15). Jack Yates family remained in Freedman s town even after his death in His son Rutherford B.H. Yates, who owned the Yates Printing and Lithography Company of Houston, had built a house at 1314 Andrews Street (Yates 1985; Bryant 1994). This residence is a Texas Historic Landmark, part of the Rutherford B. H. Yates Homestead Culture Park, and the site of the Rutherford B.H. Yates Museum, Inc. (McDavid and Bruner 2004). Since the registry of Freedman s Town with the NRHP, archaeologists of the Yates Community Archaeology Project (YCAP) have been conducting research at several sites in the district, including the Rutherford B. H. Yates Museum (McDavid et al. 2008). The City of Houston has wanted to use the land, which is very close to downtown, for housing developments (Feit and Jones 2007). Residents of Freedman s Town have been fighting for years to prevent gentrification and preserve the community, and the archaeologists at YCAP have made their project community-based in the hopes of aiding residents in the preservation of Freedman s Town. Apart from archaeological work, documentary research, oral histories, and interviews have also been undertaken (McDavid et al. 2008). The research that is conducted in Freedman s Town by YCAP is central to the issue of community development, and is used by various groups, such as preservationists, residents, politicians, and developers (McDavid et al. 2008: 37). The archaeologists have extensively studied the few remaining brick streets of Freedman s Town, and found a different (herringbone) pattern of brick at intersections (McDavid et al. 2008: 44). A similar herringbone pattern can be found at Bethel Church (McDavid et al. 2008: 46). The archaeologists attributed this pattern to the crossroads symbol, which is significant in African and African American cultures (McDavid et al. 2008: 46). Some other goals and questions of the Yates Community Archaeology Project include: (1) using material remains to understand past issues of community identity and applying 211

215 those to contemporary situations, (2) better understanding how the residents of Freedman s Town dealt with social inequality and the Jim Crow laws, (3) learning about the lives of individuals and families, and (4) comparing data from Freedman s Town, an urban area, to data from rural African American communities at historic plantations in Texas (McDavid and Bruner 2004: 10). The Rice University archaeological field school course, ANTH 362, has been held at the Rutherford B.H. Yates Museum since Excavations in the spring of 2008 took place over four weekends under the supervision of Professor Susan McIntosh of the Rice University Anthropology department, and Dr. Carol McDavid and Dr. David Bruner of YCAP. One 2 m x 2 m unit, Unit P, was excavated in the backyard of the Rutherford B.H. Yates Museum (Rice University 2008: 2). Recovered artifacts included ceramics, glass, metal (mostly nails), faunal remains, and building materials, but no features were found. Several artifacts were used as chronological markers, and it was determined that the earliest occupational phase dated to the late 1800s until the 1910s (Rice University 2008: 106). The presence of certain artifacts of personal adornment suggested that the Yates family was of a middle class economic status (Rice University 2008: 108). The site of the unit was chosen to be near the perimeter of the building because the archaeologists at YCAP were hoping to identify a path leading from the house to a previously uncovered trash burning pile in the backyard (Rice University 2008: 110). However, neither a path nor evidence of the trash pile was found in Unit P (Rice University 2008: 111). Research Goals and Expectations The soil analysis in this study took place at the site of one of the historic houses in Freedman s Town. The goals of this research were to test whether soil chemistry analyses would be effective in areas that have only been occupied for approximately one century, and to develop a rapid and relatively inexpensive way to test sites at Freedman s Town for specific activity areas. By developing such a method, the archaeologists at YCAP would be able to identify activity areas and therefore determine which areas would be most productive to excavate. Expected activity areas were trash piles, shell and cement driveways, and domestic features like cisterns on the houselots; there is some historic evidence that these features might be present on site. Based on the studies that have been 212

216 discussed, higher concentrations of phosphorus were expected at all three features, especially the trash pile due to decaying materials. In addition, high calcium concentrations were expected at the driveway because of shell concentrations. From Middleton and Price s (1996) conclusion that structures protect soils from weathering and therefore prevent the removal of elemental residues, somewhat higher element concentrations were expected underneath the preexisting house. Otherwise, element concentrations were expected to be relatively even across the lot. Since the types of activity areas at the study site were already known from historic evidence, the reason for the soil chemistry analysis was to be able to match certain elements with specific activity areas so as to use this data to predict specific types of activity areas at future sites of research in Freedman s town (Wilson et al. 2008). METHODS Site Background The site is located at 1512 Wilson Street, in Freedman s Town in the Fourth Ward of Houston, Texas, and once contained a shotgun house, which was burnt down in March Mortgage papers and a building permit indicate that the structure was present in 1921, measured 885 square feet, and was valued at $1 190,00. The house was adjacent to another, very similar, shotgun house, which is still standing (Figure 1). Site Description The sampled lot was 25 m by 9 m. The lot is surrounded on the south and east sides by a padlocked, chain-link fence lined with barbed wire, separating it from Wilson Street (east) and Victor Street (south), and on the north and west sides by a wrought iron fence, separating it from other residences. Though the area is theoretically impenetrable, many pieces of trash, such as glass bottles and plastic wrappers, were found throughout the site. The site is in the Austroriparian Biotic Province within the Gulf Prairies and Marshes region, with a subtropical, humid climate (Feit and Jones 2007). The soil in the project area has been described as a Bernard-Urban land complex (Bg) overlying Beaumont clays from the Pleistocene era (Feit and Jones 2007; McDavid and Bruner 213

217 2004). The Bernard-Urban land complex can be approximately 64 inches thick in unaltered areas; however, in the urbanized area of Freedman s town, the soil is covered in either 4-18 inches of clayey fill or up to 10 feet of fill composed of random soils and building materials (McDavid and Bruner 2004). Determining Locations A permanent datum point (SPKSTP) on Wilson Street as well as surveying equipment was used to determine the exact points for the samples. The distance from the datum point to the chain-link fence (3.44 m) was first measured using a laser. Two spots were then marked in the plot along a straight line from the datum fence line (at m and m). The chain-link fence along the east side of the plot was assumed to run north to south, and to be perpendicular to our measured line. This gave us an error of a few centimeters per sample point. Using the measured points, the sampling points were laid out and marked with flags. There were 32 points total, with three rows of 6 points and two rows of 7. The points were offset in order to minimize the intervals between samples in every direction. This method also reduces the chance of producing false linearity. Each point was therefore approximately 2.83 m away from the nearest point (Figure 2). Taking Samples A soil dibble (23 cm long, 2 cm diameter) was used to extract soil cores. Approximately g were extracted per samples. In order to obtain that amount of soil, it was necessary to extract 3-5 cores because the dibble held no more than 75 g of soil. The extra cores were all extracted within a 10-cm radius of the original, flagged sample point; the stratigraphy was assumed to be the same for each core. For each sample, the length of the soil core was measured, the soil horizons were described, any present artifacts were noted, and Munsell tests were taken (Table 1). The humus layer of each sample was discarded, but the rest of the soil was stored in a tin and weighed on a digital scale (Table 1). 214

218 Testing Soil Chemistry: Procedures and Limitations Soil chemistry is often tested using methods such as thin layer chromatography, gas chromatography, mass spectroscopy, soil micromorphology, particle size analysis, and ICP-AES (inductively coupled plasma-atomic emission spectroscopy). Common heavy metals whose concentrations are measured are copper, lead, and zinc (Goldberg and Macphail 2006: 342). Once element concentrations are determined, they are often mapped using GIS to show the distribution of patterns (Goldberg and Macphail 2006: 344). Generally, a total phosphate method, such as the use of strong (2N) nitric acid or hydrochloric acid, is necessary to extract phosphorus from both organic and inorganic matter, and to prevent interference from calcium carbonate and oxidation (Goldberg and Macphail 2006: 347). There are several potential sources of error in phosphate analysis. One problem is that animal activity can also increase phosphate concentrations; therefore, it cannot always be assumed based on only phosphate levels that there was human occupation of a site (Knudson et al. 2004). Another is that phosphorus can move down through the soil over the span of several years, thereby giving a false soil phosphate concentration (Goldberg and Macphail 2006: 347). One important consideration in conducting soil chemistry analysis is to have reference or control samples, which have not been affected by human influence (Hjülstrom and Isaksson 2009: 176). When reference samples are used, statistical analyses can be performed to determine which element concentrations exceed the natural distribution (Hjülstrom and Isaksson 2009: 176). However, some studies do not use reference samples and instead focus on relative values and general patterns (Hjülstrom and Isaksson 2009: 176). Soil Analysis The soil samples were oven-dried at 53ºC for 39 hours in the laboratory of Dr. Evan Siemann, of the Ecology and Evolutionary Biology Department at Rice University. Upon removal from the oven, the samples were weighed, transferred to Ziploc bags, and sent to the Texas A&M University Soil, Water, and Forage Testing Laboratory, where they underwent a nitric acid digest and were analyzed via Inductively Coupled Plasma 215

219 Atomic Emission Spectroscopy (ICP-AES). Concentrations in parts per million were determined for phosphorus, potassium, calcium, magnesium, sodium, zinc, iron, copper, manganese, sulfur, and boron. The concentrations for each chemical element were mapped on top of Sanborn Insurance maps of the lot using ESRI s ArcGIS version 9 (Figures 3-13). An ESRI (dbase) Shapefile was created to record the spatial location of each sample point, and the Shapefile was then associated with the concentrations of the elements by linking the database file of the spatial locations to the database file of the concentrations. RESULTS Soil Samples and Artifacts The weights before and after drying, the Munsell color tests, soil descriptions, and the artifacts in each sample are presented in Table 1. In general, the soil found in the front yard area consisted mostly of construction fill or mixed soil, the soil from the area of the preexisting house was heavy clay, and the soil from the back yard was loam. Artifacts are rather evenly distributed across the lot, though the lowest concentration is found along the path between the two houses (samples 28-31). A high frequency of charcoal occurs along the front of the house, in the north side of the front yard (samples 2, 12, and 14). There is a high concentration of brick and building material fragments in the back of the house and in the north side of the backyard (samples 4, 5, 8, 9, 10, and 18). Also, there are a lot of shell fragments in the southern section of the backyard (samples 20 and 32). A lot of rust from ferrous metal was found at sample 17. The soil from sample 27 contained a high amount of gravel, and that from sample 1 seemed highly disturbed, collapsing whenever cores were taken. At some of the sample points, full (20-cm) cores could not be taken because of impenetrable objects: large bricks were located at sample points 17 and 23, tree roots were located at sample point 21, and shell was present at sample point 32. Mapped Data The concentrations in parts per million of the elements are summarized in Table 2, and the mapped out concentrations of the elements are presented in Figures 3 through 216

220 13. Phosphorus concentrations are highest in the backyard area, especially at sample points 20, 31, and 32. Potassium concentrations are also high throughout the backyard; however, the highest concentrations are located in the space between the two houses (sample points 28, 29, and 30). Like phosphorus, the highest concentrations of calcium are at sample points 20, 31, and 32 in the backyard, but also at points 1, 13, 14, and 26 at the north end of the front yard. The distribution of magnesium is very similar to calcium: it is high in the same part of the backyard, as well as in the front yard and in the space between the northern side of the house and the wrought-iron fence (sample points 1-5 and 13). Sodium and zinc are distributed mostly in the southern part of the backyard; the concentration of zinc is highest at point 32. Iron, copper, and manganese concentrations are highest throughout the backyard area. However, there is a high amount of iron at point 17 and relatively high levels of copper underneath the house. Sulfur concentrations are highest in the southern part of the backyard and between the two houses. The concentration of boron is very low in the entire plot, except for points 31 and 32 (it is highest at 32). DISCUSSION Soil Samples and Artifacts The types of soil at the houselot are typical of Freedman s Town: both clayey and construction fill were present at the site. Heavy clays are often located underneath houses because topographically that is the lowest area of the lot; therefore, water accumulation can cause the settlement of finer sediments, such as clay. The loam in the backyard may have also been a fill, since loam is the best type of soil for gardening. Most of the soils were 10YR 3/2 (very dark grayish brown), but some samples were determined to be 2.5Y 3/2 (very dark grayish brown), which has a slightly greener hue, probably due to high rust concentrations. The low concentration of artifacts between the two houses is probably due to the fact that the area was used solely as a pathway. This is also indicated by the gravel that was found in one of the samples from the pathway area. The large bricks located at samples 17 and 23, underneath the preexisting house, may be remnants of some type of foundation or piers for the house, especially due to their proximity to the western and 217

221 southern walls. The high concentration of shell fragments at samples 20 and 32 may indicate the presence of a driveway or paving, as seen in Sullivan and Kealhofer s research (2004). Therefore, from the artifacts alone, it can only be inferred that the space between the houses was used as a pathway and that there was a driveway in the backyard. Mapped Data The distributions of the elements were correlated with the expected activity areas, such as the driveway and the trash pile, both of which were thought to be located in the backyard. Phosphorus concentrations were highest at samples 20, 31, and 32, which, along with the topographic data, suggest the possible presence of a trash pile. Calcium concentrations were also highest in that area. This may be due to either the shell artifacts or to animal bone refuse in a potential trash pile (Sullivan and Kealhofer 2004; Entwhistle et al. 2000). Magnesium distributions mirrored those of calcium very closely, possibly because of the similarities in their chemical properties. Both magnesium and calcium are alkali earth metals with the same number of valence electrons, indicating a strong correlation (Middleton and Price 1996: 675). The concentrations of the rest of the elements sodium, zinc, iron, copper, manganese, sulfur tended to be highest in the backyard area, specifically at the sites of the possible trash pile and driveway. The close correlation between iron and manganese distributions can also be explained by their chemical properties: they are very close in atomic weight (Middleton and Price 1996: 675). The elevated concentration of iron at sample 17 is probably due to the high amount of rust and ferrous metal artifacts found in that sample. The concentrations of sulfur and potassium were relatively high along the pathway between the two houses, while the concentrations of boron were extremely low except at samples 31 and 32. Though these results do not exactly parallel those from other studies, it is possible that in this case sulfur and potassium correspond to heavily trafficked areas or gravel, which was found between the two houses, while boron, like some of the other elements, is correlated with shell. In general, the highest concentrations were located around samples 20, 31, and 32 in the backyard, where there was likely a trash pile next to a driveway. The area 218

222 underneath the preexisting house did, indeed, have higher concentrations of elements, especially copper and phosphorus, indicating a lesser degree of weathering (Middleton and Price 1996). In sum, the data obtained from the soil chemistry analysis is likely effective in determining activity areas. If excavations bear out the location of the potential trash pile, as well as areas relatively devoid of activities, then the soil chemistry will be shown to be a relatively rapid way of locating buried deposits with out intensive excavations. Limitations There are several potential sources of error in this experiment, both in the sample collection and in the analysis. Multiple samples had to be taken for each point; therefore, some of the samples were taken up to ten centimeters away from the designated point. Since all of the samples for each point were combined and analyzed together, this may have skewed the data. However, since samples were several meters apart from each other, ten centimeters may not be very significant. Another potential problem from collecting the soil samples is contamination. The dibble was not washed between sample collections, so a small amount of soil from one sample may have been mixed in with soil from another sample. In addition, though the tin cans used to store the soil samples were washed prior to use, there were some remnants of soil from a previous experiment left. These also may have contaminated the soil samples. Nevertheless, the remnants of soil on both the dibble and the tin cans were so slight that it is unlikely for them to have significantly affected element concentrations. The nitric acid used to prepare the soil samples for ICP-AES analysis was a strong acid at a relatively high concentration. Though it is generally agreed that strong acids are necessary to release anthropogenic signals from soil, some archaeologists have argued that the use of a strong acid can also release certain elements from mineral lattices, which would interfere with the anthropogenic data (Middleton and Price 1996; Wilson et al. 2008; Hjülstrom and Isaksson 2009). As previously mentioned, one problem with phosphate analysis is that phosphorus tends to leach through the soil, giving a false soil phosphate concentration (Goldberg and Macphail 2006: 347; Linderholm and Lundberg 1994). Also, the ubiquity of 219

223 phosphorus in both organic and inorganic systems makes it hard to differentiate between specific activity areas (Middleton 2004; Linderholm and Lundberg 1994). This was evident in this experiment, since phosphorus concentrations were high at every expected activity area and also showed a generally high range of concentrations across the houselot. However, analyzing the concentrations of other elements allowed for greater clarification of the activity areas. Nevertheless, specific elements were not necessarily paired with exact activity areas. Most of the elements had higher concentrations near the trash pile and the driveway, and can therefore be inferred to correspond to organic debris or shell. However, some of the other distributions cannot be as easily explained. For example, the high calcium and magnesium concentrations in the north side of the front yard were not directly correlated with any artifacts or activity areas. In addition, the high sulfur and potassium concentrations along the pathway between the two houses were unexpected. There are several reasons that some elements could not be matched with certain activities. One is that no excavations were undertaken at this site; therefore, activity areas that were not necessarily expected were not discovered. In other words, it is possible that some of the element distributions correspond to activity areas that remain unknown. Another reason is that some of the results found in this experiment are not discussed in other studies. Boron, for instance, has not yet been analyzed in soil chemistry analyses. A third reason is the issue of temporal organization. Temporal organization can be just as important as the spatial organization of a site (Middleton 2004: 55). It is possible that an area that was once used for a certain activity was reused later for a different activity. Since the entire soil cores (approximately 20 cm) except for the humus layers were analyzed together, temporal organization was not taken into account; the soil was not separately analyzed based on stratigraphic layers. This may have caused an overlap of certain elements. For example, there was a significant amount of overlap for elements in the trash pile and driveway areas, suggesting that perhaps that southern area of the backyard (samples 20, 31, and 32) was used for both purposes at different times. Finally, neither reference nor control samples were used in this experiment, because there was no undisturbed area in Freedman s Town from which such samples could be obtained. Therefore, statistical analyses were not performed to determine which 220

224 concentration values were significantly different from those of undisturbed soils. Element concentrations were looked at in relative terms of spatial organization. For this reason, very slight changes in element concentrations were not considered to be significant. CONCLUSION The main goals of this study were to test whether soil chemistry analyses could be useful at sites with relatively short and recent occupation periods, and to use soil chemistry analyses to identify possible activity areas. The analyses were significant in that chemical distributions varied across the houselot; therefore, evidence for human activity can be found in the soil even during a short occupation period. Soil element concentrations did correspond with the expected activity areas: high levels of most of the elements were located in a certain area of the backyard, expected to be the location of a driveway and trash pile. Though historic data provided evidence for these activity areas, excavations were not undertaken to determine whether the driveway and trash pile were actually located in those areas. However, if the area were excavated to reveal a driveway and a trash pile, the element concentrations could be matched directly with those specific activity areas. This would then allow the use of soil chemistry analyses in Freedman s Town as an inexpensive and effective way to locate activity areas before excavating. For example, Unit P, which was excavated in 2008, was thought to contain a pathway and a trash burning pile. However, after the excavation was complete, it was determined that no evidence for either of those activity areas was found. Using soil chemistry analyses would have provided the same results in a much faster and less expensive manner, and perhaps the unit could have been excavated in a more significant area of the houselot. Therefore, this study shows that soil chemistry analysis can be a very useful tool for archaeologists working at sites in Freedman s Town. Moreover, it provides valuable information for the spatial organization of entire houselots by highlighting activity areas. ACKNOWLEDGMENTS I would like to thank the Yates Community Archaeology Project and the Rutherford B. H. Yates Museum for allowing this collaboration, and Robert Marcom for 221

225 his invaluable help during fieldwork and for the production of the figures for this paper. In addition, I would like to thank Dr. Jeffrey Fleisher for both funding this project and for his tremendous help in advising and reviewing this paper, and Dr. Evan Siemann for allowing the use of his laboratory. References Bryant, Thelma (ed.) 1994 Precious Memories: a Memoir by Olee Yates McCullough. Photocopy provided by author and located in archives of the Rutherford B.H. Yates Museum, Inc. Entwistle, Jane A., Peter W. Abrahams, R.A. Dodgshon 2000 The Geoarchaeological Significance and Spatial Variability of a Range of Physical and Chemical Soil Properties from a Former Habitation Site, Isle of Skye. Journal of Archaeological Science, 27, Feit, Rachel, Bradford M. Jones 2007 A Lotta People Have Histories Here History and Archaeology in Houston s Vanishing Freedmen s Town: Results of Field Investigations at the Gregory Lincoln/HSPVA 4 th Ward Property. The Houston Independent School District and The Texas Historical Commission, Permit #3837, Archaeology Report No French, Charles 2003 Geoarchaeology in action: Studies in Soil Micromorphology and Landscape Evolution. London: Routledge. Goldberg, Paul and Richard I. Macphail 2006 Practical and Theoretical Geoarchaeology. Oxford: Blackwell Science Ltd. Hjulström, Bjorn, S. Isaksson 2009 Identification of Activity Area Signatures in a Reconstructed Iron Age House by Combining Element and Lipid Analyses of Sediments. Journal of Archaeological Science, 36, Huston, Scott R., Travis W. Stanton, Aline Magnoni, Richard Terry, Jason Craner 2007 Beyond the Buildings: Formation Processes of Ancient Maya Houselots and Methods for the Study of Non-architectural Space. Journal of Anthropological Archaeology, 26,

226 Knudson, Kelly J., Lisa Frink, Brian W. Hoffman, T. Douglas Price 2004 Chemical Characterization of Arctic Soils: Activity Area Analysis in Contemporary Yup ik Fish Camps Using ICP-AES. Journal of Archaeological Science, 31, Linderholm, J., E. Lundberg 1994 Chemical Characterization of Various Archaeological Soil Samples Using Main and Trace Elements Determined by Inductively Coupled Plasma Atomic Emission Spectroscopy. Journal of Archaeological Science, 21, McDavid, Carol, David Bruner 2004 The Yates Community Archaeology Program (YCAP): Proposal, with Detailed Operating Plan, Phase One Investigations. Rutherford B.H. Yates Museum, Inc. McDavid, Carol, David Bruner, Robert Marcom 2008 Urban Archaeology and the Pressures of Gentrification: Claiming, Naming, and Negotiating Freedom in Freedmen s Town, Houston. Bulletin of the Texas Archaeological Society, 79, Middleton, W.D Identifying Chemical Activity Residues on Prehistoric House Floors: a Methodology and Rationale for Multi-elemental Characterization of a Mild Acid Extract of Anthropogenic Sediments. Archaeometry, 46(1), Middleton, William D., T. Douglas Price 1996 Identification of Activity Areas by Multi-element Characterization of Sediments from Modern and Archaeological House Floors Using Inductively Coupled Plasma-atomic Emission Spectroscopy. Journal of Archaeological Science, 23, Shahack-Gross, Ruth, Fiona Marshall, Kathleen Ryan, Steve Weiner 2004 Reconstruction of Spatial Organization in Abandoned Maasai Settlements: Implications for Site Structure in the Pastoral Neolithic of East Africa. Journal of Archaeological Science, 31, Sullivan, K.A., L. Kealhofer 2004 Identifying Activity Areas in Archaeological Soils from a Colonial Virginia House Lot Using Phytolith Analysis and Soil Chemistry. Journal of Archaeological Science, 31, Wilson, Clare A., Donald A. Davidson, Malcolm S. Cresser 2008 Multi-element Soil Analysis: an Assessment of its Potential as an Aid to Archaeological Interpretation. Journal of Archaeological Science, 35,

227 Yates, Rutherford B.H., Paul L. Yates 1985 The Life and Efforts of Jack Yates (O.Y. McCullough, Ed.). Houston: Texas Southern University Press. 224

228 Sample number Description Depth (cm) Munsell Weight 1 (g) Weight 2 (g) Artifacts 1 Sandy clay YR 3/ Medium Clay YR 3/ Charcoal 3 Clay loam YR 3/ Y 5/3 4 Clay YR 3/ Brick 5 Sandy clay YR 2/ Bone, glass, coal, ferrous metal, mortar 10YR 4/2 10YR 3/2 6 Silty sand YR 3/ Rust 7 Silty sand YR 3/ Brick 10YR 2/1 10YR 4/1 8 Silty sand YR 3/ Brick, coal, mortar 10YR 5/2 9 Heavy clay YR 2/ Clay YR 3/ Rust, mortar 11 Heavy clay YR 2/ Sandy clay YR 3/ Charcoal 13 Mixed YR 3/ Brick 10YR 4/3 14 Mixed YR 3/ Glass, charcoal, calcium carbonate 15 Heavy clay YR 2/ Glass, brick 16 Clay YR 2/ Brick, charcoal 17 Clay YR 3/ Rust 2.5Y 3/2 18 Clay YR 3/ Brick, mortar Sandy clay 19 Loam YR 3/ Brick 20 Loam YR 3/ Shell 21 Sandy loam YR 3/ Clay YR 3/ Rust, brick 2.5Y 3/2 23 Clay YR 3/ Y 3/2 24 Clay YR 3/ Clay YR 3/ Brick, mortar 26 Mixed YR 3/ Clay YR 3/ Gravel, glass, rust 28 Clay YR 3/ Brick 29 Clay YR 3/ Ferrous metal 30 Clay YR 3/ Clay YR 3/ Y 3/2 10YR 4/6 32 Silty sand YR 3/ Shell Table 1. A description of the soil samples, as well as the average depth of the extracted cores, Munsell values, weights, and artifacts found in the soil samples. Weight 1 = weight before drying; weight 2 = weight after drying. 225

229 Sample number P (ppm) K (ppm) Ca (ppm) Mg (ppm) Na (ppm) Zn (ppm) Fe (ppm) Cu (ppm) Mn (ppm) S (ppm) B (ppm) < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < Table 2. Concentrations of the chemical elements in parts per million (ppm). P = phosphorus, K = Potassium, Ca = Calcium, Mg = Magnesium, Na = Sodium, Zn = Zinc, Fe = Iron, Cu = Copper, Mn = Manganese, S = Sulfur, B = Boron. 226

230 Figure 1. Map of part of Freedman s Town from The yellow dialog box indicates the location of 1512 Wilson Street, with an almost identical house adjacent to it. 227

231 Figure 2. Locations of the sample points, marked by open circles, superimposed on a georeferenced Sanborn Insurance map. 228

232 Figure 3. Phosphorus (P) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 229

233 Figure 4. Potassium (K) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 230

234 Figure 5. Calcium (Ca) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 231

235 Figure 6. Magnesium (Mg) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 232

236 Figure 7. Sodium (Na) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 233

237 Figure 8. Zinc (Zn) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 234

238 Figure 9. Iron (Fe) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 235

239 Figure 10. Copper (Cu) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 236

240 Figure 11. Manganese (Mn) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 237

241 Figure 12. Sulfur (S) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 238

242 Figure 13. Boron (B) concentrations (ppm) for each sample point superimposed on a georeferenced Sanborn Insurance map. 239

243 Conclusion Jeffrey Fleisher The 2009 excavations in the back yard of the Rutherford B. H. Yates house have broaden the data set with which to discuss the archaeology of this historic house and its inhabitants. In general, the areas excavated, and the materials analyzed, by Rice University are still very small samples of a rich and complicated yard midden. However, in conjunction and comparison with data from the 2008 excavations, a number of issues can be addressed, and questions for future research can be posed. This conclusion will present a brief summary of the chronology represented in Unit YC, and its relation to the unit stratigraphy. Next, I will address one of the main questions that that these excavations hoped to answer, namely whether there was an historic outbuilding located in the back lot, and how the use of the yard space apparently changed through time. Finally, the I will discuss some of consumption patterns that were evidence in the major artifact classes, namely faunal remains, ceramics, glass, and adornments. Chronology and Stratigraphy: In keeping with excavation strategies used by the Yates Community Archaeology Project, excavations were carried out in 5 cm intervals. We attempted to excavate these arbitrary intervals within natural strata and were somewhat successful. However, the nature of the soils either very sandy, or very clayey made digging in natural strata sometimes difficult. However, as is clear from the profiles presented in Johnson s chapter above, groups of levels can be easily related to the natural strata recorded in the profiles. The following table (Table 1) correlates excavated levels to particular decades. Levels Corresponding Period 1-4 ~1960s present s-1950s s 9-12 late 1800s early 1900s Table 1: Summary of depositional periods. 240

244 Levels 1-4 contained a mixture of modern materials some of which would be considered contemporary trash including wrappers, string, hard plastic objects, bottle caps and pull tables. These levels also contained, however, a large amount of window pane glass, a fact that will discussed in greater detail below. Two coins a 1970 and 1971 penny, found in Levels 3 and 2, respectively provide a good sense that these levels date to very recent decades. The presence of much crushed oyster shell in these top levels also represents a continuation of a driveway that is still visible south of the trench, which runs along the east side of the Yates house. Robert Marcom suggested that there may have been a carport on the property close by the unit, and this is also confirmed by the presence of small car parts and other unidentifiable metal fragments in these upper levels. Levels 5 and 6 most likely date to the second quarter of the 20 th century, just after World War I. This date was derived from a number of patterns found in the assemblages of glass, ceramics, and nails. A number of well-dated objects from these levels also provide some confirmation of this scheme: a lace slip from the 1930s was found in Level 5, as well as a number of tightly dated bottles from the 1930s. Another indication comes from a shift in the shell used to make buttons, where there was a clear shift from ocean to freshwater shell buttons in Levels 5 to 7. The use of freshwater shell for buttons becomes more common during and after WWI, and thus the prevalence of freshwater buttons for the first time in Level 5 provides an indication of that period. Finally, the Junior G-man badge is well dated to the mid 1930s and was located in Level 6. Levels 7 and 8 most likely date to the first quarter of the 20 th century. The prevalence of square nails in these levels and lower is a good indicator of these decades. Two well-dated finds also suggest this dating: a pen clip found in Level 7 that was produced between , and a Band of Mercy metal badge found in Level 8 that was in common use at the turn of the century. Levels 9-12 are difficult to date accurately due to low numbers of artifacts found, but likely date primarily to the end of the 19 th and beginning of the 20 th century. The deposits in Unit YC differ somewhat from other excavations in the yard. For one, there was no obvious stratigraphic break as found in Unit P (McIntosh and Clark 2008), where Levels 6 and 6A were a sandy spread that was thought to relate to 241

245 construction infilling. In Unit P, these levels provided a clear break between deposits from the 1800s-1910s and those later than the 1920s. Unit YC exhibited a more continuous stratigraphy, and with less clear strata dating to the earlier occupation of the site. This may be due to the construction of an outbuilding in the vicinity of Unit YC, a topic to which I now turn. Outbuilding Unit YC was placed specifically to investigate area of a possible historic outbuilding. This building, represented on Sanborn Fire Maps from (see Figure 1), would have been a small structure at the back of the houselot. Although little Figure 1: Outbuilding in upper right corner. direct evidence of the structure was found, a number of individual finds and artifact patterns suggest that the excavations were either in the area of, or situated directly on it. 242