172 BULLETIN 183, UNITED STATES NATIONAL MUSEUM 



gray in color and apparently contains mucli wood ash and fine particles 

 of bone. Of the potsherds, if they were saved, no trace has been found. 

 The use of logs and bark in roofing the chamber is somewhat reminis- 

 cent of the pole- and bark-covered burial pit in the Nicholls mound in 

 Trempealeau County, Wis. (McKern, 1931a, p. 209). 



The Polander Mound group, on the Mississippi River bluffs about a 

 mile above Lynxville, Crawford County, Wis. (fig. 20, 20), yielded 

 at least one example of what appears to have been a slab-walled vault. 

 According to Thomas (1894, p. 71), mound No. 4 "measured 26 feet 

 in diameter and 3 feet high. In the center was a kind of vault formed 

 by a circular stone wall 6 feet in diameter from outside to outside, and 

 4 feet inside, built in a pit dug in the original surface to the depth of 

 a foot or 18 inches. In this vault or grave was a skeleton very well 

 preserved, doubled up and lying on the right side. . . . The vault 

 was covered very carefully with flat limestones like those of which the 

 wall was built. No implement, ornaments, or relics of any kind were 

 found." 



Mound 12 of the same group is said to have contained a like 

 structure surrounding an extended skeleton. Possibly the walled 

 tomb in Allamakee County, Iowa, a few miles to the northwest, 

 represents another. Thomas (1894, p. 107 and fig. 50) describes this 

 as a walled circular vault in which the upper courses of stones were 

 gradually drawn in and finally capped by a single rock. In ad- 

 dition to a seated skeleton it contained "a small earthen vase of the 

 usual globular form." This vessel I have not been able to trace. 

 Several shell-tempered plain and incised (Oneota) sherds from Fish's 

 mounds, now in the national collections, are probably from earth 

 mounds near the vault. 



Two parallel dry masonry walls, each 3 feet high, 8 feet long, and 

 12 feet apart, are reported bjr Thomas (1894, p. 48) in a mound about 

 6 miles southeast of Prairie du Chien, in Crawford County, Wis. 

 Between these under a layer of "mortar" was a gi'oup of extended 

 skeletons. Except in the manner of erecting the walls, if we may 

 trust the published sketch, there is nothing to justify use of the term 

 vault in this case. 



The evidence adduced in the preceding pages has been summarized 

 in table 9. A brief examination of this table will suffice to show the 

 inequalities of the data, and to emphasize their deficiencies where 

 the observations of untrained early day excavators or more recent 

 relic collectors are involved. To what extent the areas in the dis- 

 tributional map (fig, 20) showing no vault mounds represent lacunae 

 in our information rather than true absences is conjectural. This 

 question, obviously, is one that future field studies can probably clear 

 up. Meanwhile, bearing in mind the often dubious and incomplete 

 nature of the record, we may still attempt certain generalizations. 



