86 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1963 



vated a portion of the Davis site (39C014), some 200 yards west of 

 the Red Horse site. There, a complex defensive system and a series 

 of long rectangular houses were partly uncovered. Apparently there 

 were at least two, and perhaps three, occupation periods represented, 

 but time did not permit sufficient excavation to recover the whole 

 story. The earliest occupation of the Davis site was several centuries 

 earlier than that at the Eed Horse site. Continuation of the work 

 was planned for the next season. 



In the Big Bend Reservoir area, three field parties were at work 

 at the beginning of the year and a fourth party was added during 

 July. One of the parties was directed by Dr. Warren W. Caldwell, 

 assisted by Richard E. Jensen and a crew of 11. They excavated at 

 two sites. The Langdeau site (39LM209) had been a village of long- 

 rectangular houses and 15 depressions were visible. Four of these 

 house remains were excavated, and three long trenches were dug in 

 an unsuccessful attempt to find a fortification system. The houses 

 were 30-40 feet wide with no small structural posts at the ends. En- 

 trances were to the south or southwest and floors w^ere compact and 

 stained with red ochre. Pottery found there is of the Anderson 

 and Foreman types, suggesting relationship to the early rectangular- 

 house period at the Dodd site near Pierre, but other artifacts were 

 extremely exotic, including copper, shell, bone, and stone tools and 

 ornaments. This crew's second excavation was at the Jiggs Thompson 

 site (39LM208), located 9 miles north of Lower Brule in the loop of 

 the Big Bend. This site had been a small village of 17 long- 

 rectangular houses situated on a high terrace finger that was separated 

 from the rest of the terrace by a moat 4.5 feet deep and 11 feet wide. 

 Two houses were excavated, the moat was sampled, and numerous 

 other test trenches were dug. The houses had been about 30 by 20 

 feet with entrances to the south. They did not have end posts, but 

 there were massive central support posts. Architecture and artifacts 

 suggest a close relationship to the Langdeau site; both are in the 

 Anderson-Foreman and Swanson traditions of early rectangular- 

 house culture. This party completed its w^ork on August 26 after 

 11 weeks in the field. 



The second Big Bend party was also directed by Dr. Caldwell, with 

 the assistance of Richard E. Carter. It consisted of a crew of nine. 

 Excavations were carried out at a two-component site (39LM2) 

 overlooking Medicine Creek some 8 miles northwest of Lower Brule. 

 This had been a village of small, rectangular houses with ramp 

 entrances to the south, minimal end support posts, and many cache 

 pits. The remains of the first occupation were overlain by those of a 

 village of square (or subrectangular) houses, 35 feet in diameter, 

 which had four central support posts of the kind usually fomid in 



