NO. 6 SMITHSONIAN EXPLORATIONS, 1 92O IO9 



surface of the hill above, over the ledge that forms the roof. This 

 accumulation spreads for 50 feet into the cave, but not so far on the 

 outside, because there it washes down the slope. From its surface to 

 the clay floor on which it rested the greatest depth was a few inches 

 over six feet. From top to bottom there was found in this cave the 

 ordinary debris of an Indian campfire. Buried at various depths 

 here and there, in the portion within the cave were 14 human skele- 

 tons, most of them so decayed that only a few fragments were remain- 

 ing. They are of persons of various ages ; some of the skulls were 

 low, small, and flat. 



A long period of occupancy is indicated by the conditions here as 

 at Miller's cave. While the amount of earth heaped in front of the 

 cave does not seem large, yet it all has come from a space not exceed- 

 ing 6,000 square feet in area and most of this is bare rock with humus 

 of decayed vegetation existing only in the crevices or on the few flat 

 surfaces. All the other water from the hill runs to the slopes and does 

 not reach the cave. A violent storm passed over the region soon 

 after the work was concluded, in which 12 inches of rain fell in 

 three days, yet not more than a wheelbarrow load of soil was washed 

 down over the roof to the pile already there. It is evident that cen- 

 turies would be required to build up the mass, throughout which these 

 traces of man's presence are scattered promiscuously. 



During all the period these caves were in use no improvement took 

 place in the fabrication of stone implements or pottery. Specimens 

 found nearest the top of the ashes or dirt could not be distinguished 

 from those of the same class from the rock or clay floors. The inhabi- 

 tants remained in the same plane of culture. 



The thousands of small mounds extending southward from the 

 upper swamp region of Missouri have long been a puzzle to arche- 

 ologists and until recently it has been supposed that in this state they 

 are confined to the southeastern portion : but in the course of 

 Air. Fowke's field-work they have been found to extend to the north 

 and west as well. Groups of them have been located in Oregon, 

 Dent, Phelps, Pulaski, Osage, and Morgan counties, the latter along 

 the Benton County line. Their purpose has not yet been determined, 

 although Thoburn's hypothesis that they are due to the Pawnees, 

 whose line of migration was through the area in which they occur, 

 and are the remains of earth-covered houses, seems the most tenable 

 yet advanced. This theory implies that when this tribe passed beyond 

 the region in which suitable timber for supporting the weight of the 

 earth, and also earth adapted to such use, could be procured, they were 

 compelled to substitute for them small poles overlaid with grass 



