14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I07 



in the Dakotas. Here are some of the largest, best-preserved, and 

 most impressive Indian townsites in the United States. They contain 

 much of the story of the development of Arikara, Mandan, and other 

 upper Missouri cultures. Their relationship to pottery sites on the 

 tributaries to east and west is still obscure. Limited excavations be- 

 fore the war constitute only a sampling of the field; and in com- 

 parison with what needs to be done, they represent little more than 

 scattered match-flares of knowledge in a twilight of archeological 

 ignorance. Their ultimate destruction will efface forever a substan- 

 tial part of the basic material of human history in what has been aptly 

 called one of the four major archeological areas north of Mexico. 



The locating of sites, obviously, is only the first step. Actual ex- 

 cavation is needed to determine the identity of the people who left the 

 remains and the way they lived. The immediate problem therefore is 

 to evaluate the findings of the preliminary survey, and to determine 

 which sites should be further tested or extensively excavated. Since 

 we cannot hope to evacuate all of the sites that will be destroyed, it will 

 be necessary to limit our excavation program to a few truly represen- 

 tative or otherwise particularly promising locations, in the further 

 hope that time may also be available for sampling some of the less 

 promising. In short, we shall have to strike a balance between the 

 work that ought to be done, and the work for which there may be 

 time and means. 



It is expected that the excavation program will be coordinated with 

 the construction schedule of the Bureau of Reclamation and the Corps 

 of Engineers. Additional survey work is needed, however, before a 

 long-range archeological program of any preciseness can be set up. 

 It is contemplated, therefore, that the work for the calendar year 

 1947 will consist of the following survey operations : 



(a) Additional surveys in those reservoirs listed above as having 

 archeological and paleontological potentialities, with limited test-pitting 

 of a number of sites in each unit. 



(b) A thorough reconnaissance of the mainstem in South Dakota, 

 with emphasis on the topographic mapping of sites showing such 

 surface features as house pits, defensive works, middens, etc., and 

 on correlated test-pitting and collecting of sample remains. This 

 mainstem work cannot be long delayed ; the Fort Randall dam a few 

 miles above the Nebraska-South Dakota line, which will flood lOO 

 miles of the Missouri valley, is in construction status ; the Oahe dam 

 above Pierre, which will flood upward of 150 miles of the valley, is 

 only a little more remote. The detailed survey of both reservoirs will 



