532 TRANS. ST. LOUIS ACAD. SCIENCE. 



prairie. This settlement had a breadth of about fifty rods, and 

 was traced for a third of a mile to a cultivated field on the south, 

 where it abruptly terminated. On learning that the field had been 

 under cultivation nearly forty years, I concluded it accounted for 

 the sudden limit to the settlement in that direction, as constant til- 

 lage for that length of time would completely obliterate all traces. 

 To satisfy myself of this fact, I followed the bank of the swamp by 

 a path deeply worn in the soil to a narrow belt of timber separating 

 the first from a second cultivated field, and found, as I anticipated, 

 depressions in the soil overgrown with timber. This discovery 

 extended the settlement to considerably over half a mile, greatly 

 augmenting the number of dwelling-places. 



The evidences of this ancient settlement consist of circular cavi- 

 ties in the earth, generally three feet in depth, with diameters 

 varying from eight to fourteen feet. They were originally some 

 twenty inches deeper than at present, the growth and decay of 

 vegetation from year to year having partially filled them. We 

 may suppose that the dwellings, being well covered, were com- 

 fortable quarters in cold weather, completely sheltering the 

 occupants from the low temperature of winter, while in summer 

 they could quickly be made as open and airy as desirable. The 

 covering was probably constructed of poles and bark, and were 

 of no considerable height, not equal to the ordinary wigwams of 

 the Indians. If it were possible, with the evidence I shall adduce, 

 to doubt the object sought to be attained by the inhabitants in 

 digging these cavities, the opinion of an eminent English antiqua- 

 rian may be quoted. He says, " we have undoubted proofs from 

 history and former existing remains, that the earliest habitations 

 were pits or slight excavations in the ground, covered and pro- 

 tected from the inclemency of the weather by boughs of trees or 

 sods of turf." * 



I examined twelve of these dwellings, and found in each a 

 hearth the substance of which was composed of a mixture of clay 

 and coarse grass, having the color and appearance of brick, 

 though in no instance the molded form. Many of the hearths 

 were located on the west side of the cavities near the surface, 

 others were discovered in the centre. These hearths average 



• Sir R. Hoare on the Antiquities of Wiltshire. 



