ARCHEOLOGICAL INVESTIGATIONS AT THE 

 SITE OF FORT STEVENSON (32ML1), GAR- 

 RISON RESERVOIR, NORTH DAKOTA 



By G. Hubert Smith 



FOREWORD 



The Garrison Dam and Reservoir, a Corps of Engineers project, on 

 the Missouri River in west-central North Dakota, has inundated the 

 immediate valley of that river from just below the city of Garrison 

 nearly to the Montana State line. Within the now flooded area were 

 formerly located a large part of the Fort Berthold Indian Reserva- 

 tion, as well as the Fort Berthold Agency town of Elbowoods; the 

 town of Sanish; and the communities of Nishu, Independence, and 

 Shell Creek. There also were the remains of several important 19th- 

 century military and trading posts, many Indian village sites of the 

 past several centuries, and some significant geological outcroppings 

 and fossil localities. Because of this the Garrison Reservoir area 

 was one of major concern in the Inter- Agency Archeological and 

 Paleontological Salvage Program. 



The Missouri Basin Project of the Smithsonian Institution be- 

 gan an archeological reconnaissance of the area in 1946. This con- 

 sisted of a brief visit by Paul L. Cooper and J. Joseph Bauxar, staff 

 archeologists. An intensive archeological reconnaissance of a part 

 of the area was undertaken in 1947 by Marvin F. Kivett, archeologist 

 of the Missouri Basin Project staff (now director, Nebraska State 

 Historical Society Museum). Seventy archeological sites were lo- 

 cated. Subsequent field reconnaissance in 1950 and 1951, principally 

 by George Metcalf of the Missouri Basin Project staff (now of the 

 U.S. National Museum staff) , located 84 additional archeological sites 

 in the portion of the reservoir's area not previously surveyed. The 

 majority of these sites were of Indian origin, but several were of 

 White settlers, or, as they have been called, historic sites. Paleonto- 

 logical parties examined the area for fossil materials with good results 

 in 1950, 1951, and 1952. 



163 



