THOMAS.] 



OHIO. 491 



Tliere was a small luouiul near this which rontained charcoal and ashes, 

 but no bones or relics. As a means of correcting' the errors made by 

 Messrs. Squier .and Davis in the measurements of this work, the result 

 of tlie Bureau survey is given here. 



The lengths of the sides of the square, as shown by the survcj^-, are 

 8r>4 feet east and west and S.j2 north and south, being an average of 

 i53 feet greater than Messrs. Squier and Davis's measurements. The 

 work is, however, very nearly au exact S(iuare. 



According to these autliors the parallels running north to the circle 

 are ] 00 feet apart and 47.") feet long. According to the resurvey they are 

 68 feet apart, measuring to the middle line of each wall, and the aver- 

 age length is 6.34 feet (the eastern 047 and western 621). The distance 

 from the square to the break of the ravine is 427 feet for the eastern 

 side and 400 for the western, the width of the r:ivine 110 feet. 



THE GKADKD WAV.' 



• 



The excavation along which the walls extend is an old watercourse. 

 Beaver creek comes down from the hills almost directly east of this 

 work and turns south at the foot of the hill, gradually encroaching on 

 the terrace for a few hundred yards, when it is shurply deflected 

 toward the hill again. At tliis point is an old cut-off, formed when the 

 bed of the creek was much higher than at present, starting out toward 

 the west and curving until it has a direction almost north. It leaves 

 the upper terrace at the level of tlie second or next lower one and dis- 

 charges its waters into the slough wiiich extends along the foot of the 

 upper terrace, or perhaps into the river when that had its course along 

 here. This is a measure of its geologic age. It does not seem to 

 have been at anytime a regular channel for the creek, but only a place 

 through which a portion of its ,sui-plus wjiter was discharged in time 

 of freshets. 



There is a secondary terrace along the west side of this cut-olf for 

 about half its length from the northern end; on the east there is a 

 slight terrace for a few hundred feet along the last turn; the east wall, 

 at its southern end, is built along this terrace, but rises to the main 

 level at a short distance. In the other direction (toward the south) the 

 terrace soon disajipears. 



The west wall is built its entire length on the minor terrace. It is 

 much higher above the. ground on which it stands than the eastern 

 wall, though the absolute height of the latter is greater on account of 

 the greater elevation of its foundation. At each end of the east wall 

 there is a small mound. The south end of the west wall is higher and 

 wider for about 100 feet than the other portions, being heaviest at the 

 very end, resembling a donu»-shai)ed mound when looked at from the 

 level ground immediately south. Both walls have several turns or 

 angles to correspond to the irregularities of the banks. The slopes 



' Squier and Davis, Plate xxxi. 



