64 BUREAU OF AMERICAN ETHNOLOGY [bull. 76 



which a mussel shell and a flint chip were found ; and the top of the 

 clay was quite uneven, appearing as if carried and thrown here, as 

 perhaps some of it was early in the occupancy of the cave, w^ith the 

 object of making a more even or level floor farther back. But this 

 admixture was only superficial ; below it, the material had all the 

 appearance of a running water deposit. 



A ledge extended along the east wall for 40 feet, with a width of 

 12 to 14 feet ; at the inner end it was about 4 feet below the general 

 level of the floor. At 8 feet below its top a second ledge projected 

 from it, sloping toward the center, slightly for 8 feet then more 

 rapidly for 10 feet farther, where it merged into the bedrock. Then 

 came level, nearly smooth rock for 18 feet, to the foot of the slope 

 of the west wall, 14 feet out from that side of the cave. This was 

 probably the original drainage channel. 



By the gradual erosion of new channels through the limestone 

 and the consequent abandonment of old ones, subterranean drainage 

 is continually altering its direction and force. In this Avay caverns 

 may be left entirely dry, with bare floors ; or may, especially if they 

 receive the drainage of sink holes, be partially or even entirely 

 filled with debris thus carried in. Like others, Miller's Cave has 

 undergone such changes. It was begun by clear water ; enlarged 

 by erosion and by breaking down of walls and roof ; presently clay, 

 sand, and gravel were carried in; finally the water no longer flowed 

 through the front, but found its way out in some other direction. 

 In time the deposits became sufficiently dry to afford a good site for 

 camps and for permanent occupation. There is no way of ascer- 

 taining the rate at which these changes took place ; it may have re- 

 quired many centuries to make an appreciable difference in appear- 

 ance; or, on the other hand, the transition from one stage to the 

 next may have been rapid. 



Along the foot of the ledge from the east wall the clay was only 

 a few inches deep; farther out on the ledge, and on the projection 

 extending from it, were layers of red sand. Occasionally a small 

 patch of it appeared along the western side. Probably it was washed 

 in among the last of the natural deposits. 



There was considerable chert gravel mixed with the clay, making 

 excavation as difficult and laborious as digging up an old, much- 

 traveled macadamized highway. 



The surface of the ashes sloped upward rather rapidl}^ for a 

 distance of 29 feet from the front. Kitchen refuse, found in them 

 from the start, contained manj^ mussel shells; bones, including those 

 of bear, deer, panther, turkey, and other large fowls, tortoise, turtle, 

 fish, and various small mammals and birds; potsherds; broken flints, 

 with the debris of chipping work ; mortars, pestles, hammers, and 

 mullers. Near the west wall, 14 feet from the mouth, imbedded in 



