Phase II and Phase III Archaeological Database and Inventory
18AN1000 Larkington
Diagnostic Artifact Data
Features
Summary
The Larkington Site (18AN1000) is a multicomponent prehistoric and historic archeological site opposite Gelebe Bay from Edgewater in Anne Arundel County. The site is situated within a small drainage divide between Glebe Bay and Brewer Creek, both of which are embayments of the South River. The hillocky site was covered in woodland and underbrush at the time of its discovery and excavation. It has since been developed into a residential lot with a single-family home now occupying the site. Colemantown sandy loam covers the entire site.
The site lies within the larger Larkington patent of one John Brewer. Larkington was a 300-acre tract surveyed for Ellis Brown (arrived 1649, died by 1667), who conveyed his interest to John Larkin, who, in turn, assigned the land to John Brewer on September 2nd, 1663. The metes and bounds description of the tract places it on the north end of the peninsula defined by Glebe Bay (Burgess, or Shipping Creek in the 18th century) to the west, the South River on the north, and Pythers Creek on the east.
Larkington’s paper trail for the last third of the 17th century and all of the 18th century is poor, but I appears the land remained in Brewer hands. In his 1809 Last Will & Testament, William Brewer bequeathed to his eldest son, Joseph Newton Brewer, all of his apparel, furnishings, stock, slaves, and the 300-acre Larkington tract. This will reveals that an entail initiated by William Brewer’s great-grandfather (probably John Brewer ca. 1690) prohibited division of the estate and required passage from one generation to the next by way of the eldest surviving son. Joseph N. Brewer did not sustain the entail, possibly because he lacked sufficient wealth to set up his younger son, possibly because its observance flew in the face of the realities of the mid 19th century agricultural realities. His will, written three weeks before his death in January of 1841, divided Larkington between his eldest son, Joseph N. Brewer II, and Nathaniel N. Brewer.
We know from William Brewer’s will of 1809 that he owned slaves, horse, cattle, sheep and hogs, as well as household furnishings, plantation tools and utensils. This suggests that his plantation would have included slave quarters, stables, corn cribs, barns and sheep folds, and a house, possibly with a detached kitchen. The 1776 tax assessment enumerated William Brewer’s household as consisting of seven family members and eight slaves. The 1798 federal direct tax similarly listed 8 slaves, along with 3 buildings valued at more than $300 each.
Data from mid-19th century Agricultural Censuses reveal that the two Brewer farms (Joseph’s and Nathaniel’s) were practicing cooperative farming and equal division of proceeds. The Brewers practiced mixed farming, raising tobacco, wheat, maize, and rye with lesser quantities of potatoes (probably for household consumption). They had 12 sheep among them in 1850, yielding 36 pounds of wool, but none in 1860. They raised swine and dairy cattle, likely to supply household needs, and they kept oxen, undoubtedly for plowing. In 1850, the Brewer households produced 140 pounds of butter between them, an amount well short of the 200 pounds or more that each household likely consumed during the course of the year. Characteristic of households living in the Chesapeake region, the Brewers did not make cheese. The values of the two farms jumped from $3,000 each in 1850, to $6,000 and $5,000 in 1860, then tumbled to $3,500 and $4,400 in 1870. That pattern occurred throughout Maryland, prices rising precipitously in the 1850s as a result of international market developments, then crashing in the 1late 1860s, leaving many farmers with loans that exceeded the values of their holdings.
In 1882, Nathaniel N. Brewer II and Joseph A. Robinson (a nephew) executed simultaneous deeds that divided the lands of the late Joseph N. Brewer (I). By reference to an 1878 map of the area and the limited descriptions described in these deeds, it appears that Joseph N. Brewer’s moiety of the Larkington farm lay to the northwest of Nicholas N. Brewer’s, and that it was situated to the northwest of land owned by Edward Collinson. The 1882 division of Joseph N. Brewer’s land between his brother and their sister’s son (Joseph A. Robinson), conveyed the house of Joseph N. Brewer (I) and 100 acres to Robinson and 50 acres to Nicholas N. Brewer. By 1892, Nicholas N. Brewer conveyed 40.75 acres, including Long Point and a house, to his daughter, Mary E. Collinson (wife of Dr. John Collinson). This parcel would appear to be situated across the road from 18AN1000. This would make the site part of the land conveyed to Joseph Robinson and, likely, the house site of both Joseph N. Brewers (father and son), as well as that of their father and grandfather, William.
Joseph A. Robinson appeared in the Brewer household in the 1850 Federal Census, age 15, and with Sarah Brewer (Joseph N. Brewer’s widow) in the 1870 Census, aged 35. The 1870 household also included 17-year-old Caroline Randall (a black domestic servant) and her infant son Augustus (described as Mulatto). The farm was devalued between 1870 and 1880 from $4,400 to $1,500. This corresponds with the timeframe when 50 acres of the farm went to Nicholas N. Brewer (see above). Numbers of livestock did no differ markedly from the 1850 and 1860 values though, and the harvest of 250 bushels of maize from 20 acres also approximated past yields
In the 1880 Census, Joseph appeared as a 43-year-old single male. He was listed as a farmer and his neighbor (a 41-year-old African American named William Jennings) was an oysterman. Many of the neighboring households were listed as black. Robinson appears to have been listed as black as well, but that part of the entry is blotted and the census marshal might have recognized his mistake. All previous and subsequent censuses described Robinson and the Brewers as white. Robinson appears to have reduced his level of effort in field crops, or lost the entirety of his wheat and rye crops in 1880. His harvest of 850 pounds of tobacco from two acres was a change from no tobacco in 1870, but a significant reduction over what had been raised on the farm in years past. The loss of 50 acres in may have included preferred lands for tobacco and grain production. Robinson also paid only $75 in wages the previous years, in contrast to $280 paid by his aunt the previous census year.
In 1900, Nicholas and Mary Brewer appeared in the population schedule of the census as farmers boarding an African American named Thomas and his daughter, Bessie, both of whom likely worked for the Brewers. Joseph A. Robinson appeared as head of household with James and Louisa Baxton, their four sons (aged 8-17), and granddaughter: all listed as “colored”. James was listed as a 40-year-old farmer renting his house and farm, and three of the four sons appeared as farm hands. At 65-years-old, Robinson was unmarried, literate, owned his house and farm, but did not farm. The census entry listed his occupation as that of “carpenter”. He appears to have boarded an African American family that farmed the land, although on what terms remain uncertain. In 1910, Robinson again appeared as head of household, aged 73. His occupations were listed as “Farms” and “House____”. The latter is illegible, but appears to be “Painter” or “Framer”. There was no entry for where he should have appeared in the agricultural schedules, but there was one for Thomas Benson or Baxtor in the following entry. He was listed as a lodger in Robinson’s house along with his wife Lettie (illegible) and son Colonel (aged 3). Thomas and Lettie was described as black, illiterate, and 57 and 50 years old, respectively. Their son was described as mulatto. Thomas was also listed as a tonger, harvesting oysters, and Lettie was listed as a farm laborer “working out” (hired by others). These boarders may well have been related to those who lodged with Robinson in 1900, but they were not the same people.
A division of Robinson’s land occurred in 1943, the plat of which survives. The 1943 plat notes a family cemetery on the land of Edward Collinson, but does not indicate the presence of any of what should have been several dwellings and outbuildings on the neck and tip of the peninsula. Thomas and Margaret Kenney purchased the 27.8 acre Lot 1 (on which the site is situated) from the Collinsons in 1951. Margaret Kenney conveyed the site location to Glebe Bay General Partnership in 1993.
The site was first identified by the Anne Arundel County Archaeologist in 1994 following a walkover survey initiated by the need of a local developer to obtain a wetlands permit from the Maryland Department of Environment and the US Army Corps of Engineers prior to construction of new housing stock. In accordance with Section 106 of the National Historic Preservation Act of 1966 (as amended), a project requiring a federal permit must consider adverse impact to potentially significant cultural resources by its undertaking. The Anne Arundel County Archaeologist identified the site when he encountered a looter’s backdirt pile, from which he collected tin-glazed earthenware, refined red earthenware, scratch blue white salt-glazed stoneware, blue and gray salt-glazed stoneware, wrought nails, annular ware, transfer-printed whiteware, pearlware, and American brown salt-glazed stoneware (these materials are not included in the tally of artifacts provided above). In response to the pedestrian survey and the recordation of the site, the Maryland Historical Trust recommended to the Army Corps of Engineers that Site 18AN1000 be evaluated in order to determine its archeological significance and eligibility for listing on the NRHP.
Phase II evaluation of the site was carried out in 2000 when it became clear that the lot on which the site is situated was going to be constructed upon. Again, the developer’s need to obtain a wetlands permit triggered the need to evaluate the site. The Area of Potential Affect (APE) within Lot 60R which contained the Larkington site, was initially examined by a pedestrian survey and shovel testing. This information was then used in the placement of test units.
Pedestrian survey was carried out following the manual clearing of understory and thicket prior to the onset of fieldwork, particularly in the western portion of the site. The goal was to identify any exposed cultural features and artifact scatters visible through the remaining vegetation cover. Those that were identified were plotted on a site map and flagged in the field for later examination.
Shovel tests were laid out at 7.62 m (25 foot) intervals on a grid aligned to the local roads (which happened to approximate magnetic north). In addition, several judgmentally-placed shovel test pits (STPs) were excavated off the grid system in order to test strategic site locations. STPs were excavated manually using spade shovels and trowels. Each shovel test was excavated to a depth of 7.62 cm (3 inches) into culturally-sterile subsoil. The standard diameter of a shovel test measured 30.5 cm (1 ft). The soils were excavated stratigraphically in order to maintain vertical control over the material recovery. Soils removed from each unit were dry-screened in the field using hardware cloth. Materials recovered during the screening process were bagged according to provenience information, i.e., by shovel test and stratigraphic designations. A soil profile was also prepared for each STP on standardized forms, using standard methods and nomenclature. A total of 65 shovel tests were excavated across the site.
Four masonry features were identified at the Larkington site between the initial site visit and shovel testing. Prior to evaluating the site with test unit excavation, the horizontal limits of each of these features were established. This involved the clearing of remaining surface vegetation as well as the removal of overburden associated with prior site disturbance. Full horizontal exposure of structures was considered important in the determination of test unit placement. Artifacts were surface collected from the soils removed in association with exposing the horizontal limits of the masonry features. There was a need to be selective during this artifact collection because of the ubiquity of exposed materials present. Thus, only large, diagnostic artifacts were retained.
A total of 12 test units was excavated at the Larkington site. The placement of these units was based on the location of cultural features and dense artifact deposits. Units were oriented toward cultural resources rather than being adapted into the shovel test grid system. Following excavation, the units were mapped using a surveyor’s transit. Eleven of the test units were 91 X 91 cm (3 X 3 ft) in size, while the twelfth was reduced to 61 X 61 cm (2 X 2 ft) in order to place it within the confines of an identified feature. A datum for each unit was established in the southwest corner of each unit. Test units were excavated manually using flat shovels and trowels. The soils were removed stratigraphically in order to maintain vertical control over the artifacts recovered. Excavated soils were dry-screened in the field using hardware mesh and standard report forms were used to record data pertaining to test unit strata. When cultural features were encountered, plan views and sectional drawings, as well as photographs were prepared. Non-structural features were first exposed horizontally, and then excavated separately from the surrounding matrix. These features were bisected, sketched, and photographed. In addition, soil samples were retained for water-screening through window mesh. Unit excavation was generally halted once a 7.62 cm (3 in) depth was penetrated into culturally sterile subsoil. A hand-drawn soil profile was prepared for at least one wall of each test unit excavated and soils were described on standardized forms in the same manner as the STPs. Soil profiles were also documented through photography.
Five cultural features were identified prior to test unit excavation. They included Feature 1) the stone foundation of a dwelling containing two rooms, an exterior brick pier, and a hearth base, Feature 2) a looted cellar hole of an extinct frame house, Feature 3) a stone foundation for a possible springhouse or dairy, Feature 4) the stone foundation of an outbuilding, and Feature 5) a dry-laid brick well with a modern top addition. Seven test units were excavated in the eastern half of the Larkington site adjacent to Features 1, 4, and 5. Feature 7, a builder’s trench, was identified adjacent to the Feature 5 well. Two units were situated within artifact rich areas (based on shovel testing) in the same general vicinity as Features 1 and 4 (i.e. probable yard areas). Feature 6 was identified in one of these units, which was interpreted to be a roughly 2.5 m diameter refuse pit. The remaining five test units were excavated in the northwest corner of the Larkington site where Features 2 and 3 were located.
Between surface collection and excavation, 9,543 artifacts were recovered from the Larkington site during the Phase II excavations. Of these, 11 were prehistoric, including 1 rhyolite Perkiomen point, 1 expended chert angular core with evidence of utilization, 7 primary flakes, and 2 secondary waste flakes. The Perkiomen point is a Terminal Late Archaic broadspear, roughly dated between 1700-900 BC, a type commonly made from rhyolite. The historic assemblage consisted of 24 activity items (2 toy pieces, 2 possible utility knife blades, a pitchfork, a lead net sinker, and 18 pieces of lamp glass), 1,518 architectural artifacts (17 wrought nails, 19 wire nails, 281 cut nails, 1 whole brick, 778 brick fragments, 206 mortar fragments, 120 window glass, 92 iron sandstone fragments, and 4 slate pieces), 25 clothing objects (18 buttons, 1 buckle, 1 pin, 1 metal boot eyelet, 1 glass tuxedo fastener, and 3 shoe fragments), 1312 ceramic sherds (5 tin-glazed earthenware, 2 Jackfield, 5 North Devon, 47 unrefined earthenware, 5 refined earthenware, 41 creamware, 28 pearlware, 189 whiteware, 284 ironstone, 47 non-opaque porcelain, 9 opaque/hotel porcelain, 10 Rockingham, 24 yellowware, 18 white salt-glazed stoneware, 3 Rhenish stoneware, 4 domestic blue and gray stoneware), 991 bottle glass fragments (493 colorless, 226 aqua, 93 brown, 75 olive, 49 manganese-tinted, 19 green, 7 blue, 8 amber, 3 milk), 1 glass platter piece, 40 other vessel glass pieces, 5,655 faunal remains (4649 oyster shell, 1 conch shell fragment, bones and teeth including pig, sheep/goat, cow, bison, possibly horse, raccoon, muskrat, beaver, eastern fox or gray squirrel, turtle, and fish scales crania, vertebra, and rib bones), 36 floral objects, 11 other miscellaneous kitchen items (6 tin can fragments, 1 metal canning lid, 4 metal canning jar lids), 2 personal items (a 1917 Lincoln penny and a black comb fragment with an 1851 date embossed), 7 tobacco-related artifacts (kaolin pipe fragments), 1 arms object (a shotgun shell percussion cap), and 109 miscellaneous objects (8 snail shells, 1 barnacle, unidentified hardware, and plastic).
Based on documentary research and archeological fieldwork, the Larkington site was determined to be composed of two domestic complexes and occupied during two distinct time periods. The earlier occupation may have begun as early as the mid 17th century, but started no later than the mid 18th century, and ended before 1860. The occupants were affiliated with the Brewer family plantation and may have been overseers and/or tenant farmers. The later occupation began shortly after 1860 and lasted until 1916. Joseph A. Robinson, a member of the Brewer family and a small-scale farmer, was the occupant during this period. An examination of the ceramic artifacts and faunal remains suggested that the occupants of the Larkington site were of a modest socio-economic status.
Phase III data recovery was carried out at the site in the spring of 2005, again, in conjunction with the ongoing residential development of the area. The Phase III investigation consisted of the excavation of seventeen 1.524 X 1.524 m (5 X 5 ft) test units, and three smaller test trenches designed to expose portions of brick walls. Test unit and test trench locations were chosen based on the location of features identified during Phase II work and their predicted/projected extents. Soils from Phase III test units were screened through hardware cloth, but the test trenches were not screened.
A total of 11,334 artifacts were collected during the Phase III study including both prehistoric and historic artifacts. The prehistoric assemblage included at least 106 objects: 55 flaked stone objects (3 quartz, 8 quartzite, 8 chert, and 36 rhyolite), 6 pieces of fire-cracked rock, 8 other lithics, 14 Popes Creek sherds, 21 Mockley sherds, and 2 Rappahannock sherds. The prehistoric ceramic finds suggest occupation during the Middle Woodland and early Late Woodland periods. The meager artifact evidence suggests that these were small, single component occupations that focused on oyster harvesting and processing. The different prehistoric components have been smeared by plowing, to the extent that intact midden deposits are few and severely truncated, the associated materials spread to the point where individual component cannot be discerned. While most of the occupation is concentrated in the northeast portion of the site, prehistoric artifacts are spread out in 15 of the 29 screened units (both Phase II and III).
The Phase III work revisited all of the features encountered during the Phase II testing and was able to sort out the chronology and evolution of the farm landscape and associated features. Feature 1 (the stone foundation of the two-room house identified during Phase II) and an associated sheet midden were dated to the 18th century occupation of William Brewer. Feature 2, the looted cellar hole, was dated to the period of both William and Joseph N. Brewer (I). Features 1, 2, and 6 (trash pit) were associated with both Joseph N. Brewers (father and son), as well as Joseph A. Robinson. And Features 1, 4, and 6 with the occupation of Joseph A. Robinson.
The historic assemblage consisted of 11,228 objects. The pre-1800 ceramic assemblage included 2 Astbury, 3 Staffordshire, 13 tin-glazed earthenware, 7 slipped redware, 16 white salt-glaze stoneware, 3 British Brown stoneware, 3 Nottingham stoneware, 4 Rhenish gray stoneware, and 4 Westerwald stoneware sherds. At least 84 pearlware sherds were also recovered. Beyond these ceramic artifacts, the remaining 11,089 artifacts are cataloged in the tables above simply as miscellaneous historic objects. The full Phase III site report does not appear to distinguish between prehistoric and historic items, Phase II and Phase III artifacts, glass and ceramic vessels, etc. in discussion and, thus it is quite difficult to arrive at accurate counts for various artifact types.
Evidence of an early historic occupation, specifically from sometime in the 18th century, is evident from the artifacts recovered during both phases of investigation. There is no indication of a 17th century occupation: the few stonewares classified as Rhenish may, in fact, be 18th century Westerwald sherds, and all of the tobacco pipestem bore diameters (both Phase II and Phase III) are 4/64ths or 5/64th inches (1720-1800). Shovel testing demonstrated that these materials were deposited across the lot, and particularly along the north lot line, but concentrated in the northwest quadrant. Twenty-four excavation units (Phase II and III) produced highly fragmented, non-mending ceramics that probably pre-date 1800. The highest concentration of pre-1800 materials came from the area of Feature 6. Pre-1840 material occurred in its highest concentration in a relatively undisturbed context within Feature 1 (the two-room stone foundation structure). Evidence of plowing was observed, which was likely responsible for the widespread distribution and fragmentation of the prehistoric and early historic artifacts.
The recovery of more than 70 pieces (weighing nearly 20 ounces) of what appears to be burned daub from units in the northeast corner of the site suggests the possibility of a nearby earthfast dwelling that was the focus of an early historic houselot. Feature 6 may have been a borrow pit, or a series of intersecting borrow pits, whence sandy loam and sandy clay were quarried for use in a wattle and daub chimney. Postholes and molds from this hypothesized structure have not been found.
The bulk of the archeological evidence appears to be the products of two households, both small by local standards: that headed by Joseph N. Brewer from as early as 1811 until his death in 1841 (and by his widow, Sarah, after that), and that headed by Joseph A. Robinson from ca. 1880 until his death in 1911. It is entirely possible that Joseph Robinson lived alone for many of those years between 1880 and 1916. Most of the deposits, therefore, can be linked to one of two households representing successive generations of the same family: the Brewers from 1811 to 1880 and Joseph Robinson from 1880 to 1916.
Reasonably intact deposits were identified only for the latter occupation. They revealed a simple pattern of ceramic (undecorated white dinner services) and glass tableware acquisition, and possible persistence of family heirloom tablewares. Ceramic storage vessels were few, glassware fulfilling at least part of that need. Purchased packaged goods were few: no food tins were recovered, and most of the bottles appeared to have contained non-alcoholic beverages and common over-the-counter rostrums for colds and headaches. Despite the numerous ceramic and glass fragments recovered from the soils in and around Feature 4, Joseph Robinson appears not to have been prodigal in the use of his wealth. To the extent that he expressed his identity, it may have been through the historical connections of his dwelling and the farm, and possible through the use of some heirloom kitchen furnishings. His ancestor William Brewer, by contrast, built a brick house, probably maintained a wine cellar, and entailed his estate (revealed through archival research), all expressions of gentility and wealth.
Combined Phase II and III investigations have produced a large assemblage of material that may be used to further test models of consumer behavior. Test excavations have established a chronology for the features. Additional fieldwork at Larkington (18AN1000) is unlikely to wrest significant additional data on household consumer behavior patterns. No further investigations were recommended, and a new single-family residence has been constructed at the site.
|
