THOMAS.] 



OHIO. 



481 



error, as owing to long cultivation it is impossible to tell with certainty 

 where the center of the embankment was ; but by placing the instru- 

 ment as near the center as could be determined, about midway of the 

 length, and placing the rod the same way near each end, the error will 

 be very slight; not more than a few minutes. 



Great care was taken in measuring the sjnaller circle, as on it Squier 

 and Davis appear to base their claim to the mathematical accuracy of 

 the works constructed by the "Mound-builders." 



In order to prove this accuracy they present a supplementary plan 

 on the plate, showing their method of determining the curves, and in 

 a footnote on the following page give the lengths of the arcs and the 

 directions. It is somewhat disappointing to find, on carefully examin- 

 ing this showing, that it relates wholly to a hypothetical case, and not 

 to a circle actually surveyed. This is apparent from the following- 

 facts: The circumference of the circle, given as an illustration, is 3,600 

 feet and the diameter 1,146 feet, which agrees with no circle given in 

 their entire work. Had they given an illustration from an actual sur- 

 vey it would have been much more satisfactory. 



The courses and distances of Mr. Middleton's survey of Ihe smaller 

 circle of this group are inserted here, made after the plan Squier and 

 Davis suggest, but with shorter chords; also the plat made therefrom 

 (Fig. 321). The place of beginning is at station 1, the course south- 

 west, westward, and so on around. 



The smnll circle. 



12 ETH- 



-31 



