236 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 93 



at present or within historic times, and no evidence is seen by which one may 

 assign any considerable age to the occurrence. Although probability favors a 

 Recent time to the deposit, there is no certain evidence precluding a late Pleisto- 

 cene age. 



The Bison material is not sufficiently complete to permit specific recognition 

 as species of this form cannot be identified without good skull and horn material. 

 The size, shape, and position of horns play an important part in the diagnosis. 

 PVom the material at hand, the form could be Bison bison. 



The differences between the two mammalian faunas [I and 11-111] is not im- 

 portant. Both are very incomplete, representing only a small percentage of the 

 forms inhabiting the region at the two stages. 



Gerrit S. Miller, Jr., adds: 



I can see nothing in the Bison remains from either of the three levels to 

 distinguish them from recent species. The teeth of the Antilocapra seem un- 

 usually large [level I]. I hope more of them will be found if further work is 

 done at this locality. 



The few bird remains, identified by Dr. A. Wetmore, are of existing 

 species. 



CONSIDERATIONS OF RELATIVE AGE 



The main ramparts of Signal Hutlc were apparently isolated back 

 in the Pleistocene, and for many thousands of years the mesa has 

 probably been of about the same shape as at present, although scaling 

 and erosion have been continuous. The earth cap is geologically 

 recent and was formed primarily by wind action and secondarily by 

 human occupations. The crescentic form of Signal Butte, backed 

 by the main Wildcat escarpment (pi. 23, fig. i), causes winds com- 

 ing across the plains to be deflected either tipward or around the 

 two ends. This creates an eddy of comparative quiet over its top. 

 Observation and experiment have shown that regardless of direction 

 the winds tend to blow upward toward its center. This was un- 

 pleasantly apparent while screening operations were in progress. These 

 eddying conditions over a long period have caused the deposition 

 of the natural portion of the earth cap, the thin grass cover helping 

 to hold this material in ])lace. Presumably, this process would have 

 commenced as soon as suitable material was available for the winds 

 to carry. The formation of sand dunes in the valley of the Platte 

 River, to which Signal Butte lies adjacent, point to the time, subse- 

 quent to the isolation of the butte, when dust became available. (The 

 tnore detailed study on which this brief summary is based cannot be 

 presented here but will appear in the complete report.) The dunes 

 nearest to the butte are a patch, about 4 miles in area, 3 miles to the 

 southwest. They are close to the abrupt escarpment of the Wildcat 

 Range and have reached a point where prairie-sweeping winds are 



