The range of cabins along the north wall of Fort Floyd as they were excavated in 1954. They would have included the kitchen/residence, a stock room for trade goods, and a room for pelts and furs. SHSND AHP 32-MNI-23 The physical location and identification of the Columbia Fur Company’s Fort Floyd has A REVISED long been a mystery for scholars of the Upper Missouri River fur trade. It was believed the HISTORY OF original post at what is today Fort Union Trading Post National Historic Site was called Fort Floyd because the pioneer fur trade historian FORT FLOYD Hiram M. Chittenden had so identified it in 1902.1 But this was not the case; indeed, its predecessor was some 118 river miles down the Missouri River, and now new information allows By W. Raymond Wood and Michael M. Casler us to tell its story in more detail.2 Volume 80.4 3 Fort Floyd (32MN1) and its vicinity, showing local features near the post at the time of its investigation in 1954. SHSND AHP 32MN1 figure 2 Let’s begin that story in 1954, when artifacts, and made a sketch map of the excavation of Fort Berthold I, an Alan R. Woolworth and W. Raymond the post.5 American Fur Company post further Wood excavated a fur-trading post downriver, and they set up camp on the east bank of the White Earth Recognizing that the erection of the in and around a log cabin on the River in Mountrail County, North Garrison Dam and the resulting Lake abandoned ranch a few hundred yards Dakota, on the property of the former Sakakawea threatened to destroy the west of the post.6 At some time in the Stewart Ranch. The site had been site, the River Basin Surveys made past, cultivation had obscured what known for years prior to 1938, when Smith’s information available to remained of the fort’s surface features, Thaddeus C. Hecker of the State Woolworth and Wood, who arrived at although the plowing was shallow and Historical Society of North Dakota the site in July 1954 with a crew from only minimally disturbed what was (SHSND) located and mapped its the SHSND. They had just completed left of its subsurface remains. The site remaining visible features.3 George F. was fully excavated in the following Will and Hecker later published a short weeks, revealing the charred remains note on what they called Kipp’s Trading of its buildings and the palisade that Post Site, but it was not until 1951 that James Kipp had been sent to the enclosed them. It was obvious that its the site received further attention.4 A future Fort Floyd location in the buildings and parts of the palisade had Smithsonian Institution Missouri River fall of 1826 to build a post for the been destroyed by fire. Basin Surveys field crew under the direction of historical archaeologist express purpose of trading with James Kipp had been sent to this G. Hubert Smith obtained permission the Assiniboines, who were then location in the fall of 1826 to build to carry out test excavations there a post for the express purpose of that summer. He successfully located living north of the post. trading with the Assiniboines, who the palisade trench, recovered some were then living north of the post. 4 North Dakota History At the time of its construction, the nearest trading post was Fort Clark I (1824–30) at the Mandan village of Mih-tutta-hang-kusch, near the mouth of the Knife River, about eighty miles downstream.7 Built in an area free of any competition, Fort Floyd would have been seen as a profitable venture for the company, and it may also have been built in response to a request by the Assiniboines for a post in their territory.8 THE SITE EXCAVatED The excavations revealed the architecture of this post and, because it had remained in operation for only about four years, later rebuilding did not obscure its ground plan. The post was square in outline, ninety-six feet on a side, and enclosed by a palisade consisting of closely set cottonwood logs set in a shallow trench. Its south wall, built near the edge of a high terrace facing the Missouri River, contained a ten-foot-wide entry, probably once closed by twin gates. A single bastion on the northeast corner of the palisade was simply a five- foot extension of the palisade, for it lacked the superstructure found on the blockhouses of later posts. Ground plan of Fort Floyd, as it was revealed by its excavation by the State Historical Society of There was no comparable feature North Dakota in 1954. The bastion on the northeast corner was not mirrored by one on the southwest on the southwest corner, although corner. SHSND AHP 32MN1 figure 3 fired clay chinking found among its posts might represent some served as the kitchen, for large quantities elaboration of the palisade. of animal bone, broken dishes, cups, and bowls were in its ruins. It also boasted a clay- The charred remains of a range chinked stick chimney. Abundant kitchen of closely spaced buildings along refuse littered a long trench behind the cabin the north wall faced the entry. and next to the palisade wall, filled with Three of them were cabins built broken dishes, wood ash, and the bones of of peeled logs and chinked with Canada geese, mallards, grouse, bison, deer, clay. Their foundations rested beaver, elk, and muskrat. on log sills, outlining floors that were about sixteen feet wide by The central and eastern cabins lacked eighteen feet deep. Floor joists fireplaces, making it unlikely they were used supported heavy, hewn-plank as living quarters—although both had plank flooring that had been laid down floors, frequently built to keep goods away over earlier, prepared clay floors. from moisture. One of the cabins probably James Kipp in retirement, as photographed Thin fragments of window glass served as a warehouse for trade goods and on March 29, 1873, probably near his reveal that some of the cabins had the other for storing furs. A fourth structure, home in Missouri, age 85 years 14 days. windows. The westernmost cabin denoted by parallel charcoal stains in the Montana Historical Society Research was probably a residence that also earth, may also have had a floor, but it Center Photograph Archives, Helena Volume 80.4 5 had been carried away by the plow. Kipp used at both posts to store goods out of brought at least one cannon to the post, for sight until a warehouse could be built.9 fragments of a one-pounder were scattered Whatever their function, the pits at Fort about the enclosure. The iron used to make Floyd were filled with refuse after their the tube was excessively porous and had original purpose was fulfilled. been repaired using a wrought iron patch, indicating the reason for its sudden failure. A small rectangular pit in the southeast One of its fragments was near what would corner of the enclosure contained the Rifle balls. SHSND 12621.21, 22 have been the front door of one of the remains, in a nailed wooden box, of an cabins, where it may have served as infant about six months of age. About a doorstop. five thousand blue glass seed beads accompanied the child. Because scaffold Gun flints, rifle balls and shot, and bullet burial was the usual custom of the mold waste material were recovered that Assiniboine customers of the post, surely indicate the use of horse pistols or trade the infant was the child of one of the guns, rifles or fowling pieces, pocket pistols employees and his Native American wife.10 or rifles, and carbines or trade guns, though A pair of conjoined ceramic doll legs found the only actual gun part recovered was a elsewhere suggests the presence of another hammer from a flintlock rifle. The lead shot child, a young girl; a groove around the was for fowling pieces or smoothbore guns. doll’s upper thighs would have served to Because waterfowl and other game birds attach them to a cloth body. There were were abundant in the kitchen refuse, these other hints that Native American wives Dishes. SHSND 12621.100 guns were obviously put to good use. were living there. Perhaps these women were using the small, charcoal-filled pits in Scattered droplets of lead found near a the compound for smoke-tanning hides, fireplace on the east side of the compound for the Hidatsa technique of suspending suggest the men cast rifle balls there. Five hides over smoking corncobs does not large pits scattered about the compound seem to have been used by the traders had steep, nearly vertical walls and a themselves.11 In addition, chipped-stone narrow ridge of native soil across their short end scrapers, a serrated fleshing tool, axis; two of them were floored with wood and grooved mauls or hammers were boards, which would have kept their goods recovered either from the cabins or refuse off the earth and away from moisture. pits. Such tools usually are associated with William Hunt wrote that these pits were women’s work in dressing hides. These similar in size and form to a number of pits nearly invisible members of the fur trade exposed during the archaeological work community probably also cooked, prepared at Fort Union that dated to its early years, the hide clothing preferred by many of the perhaps in the 1830s. Perhaps they were traders, and lightened the burden of the Artifacts unearthed during the Doll legs.