ARCHEOLOGY. 



43 



figure of irregular curvature, with one diameter more than a hundred 

 feet longer than the other. Instead of continuing around the head of the 



Fig. 3. 



ravine south of the entrance, the embankment stops more than 300 feet 

 from the corresponding point on the opposite side. The original forest 

 extends over this portion of the work, and the ground is perfectly level; 

 while the southern part is in a cultivated field, and the wall quite distinct. 

 So there can be no claim that any portion of it has ever been obliterated. 

 Yet in the original description of this plate, the authors claim to have 

 been at great pains to ensure the absolute accuracy of their work. In a 

 foot-note presumably referring to this figure, they give the field notes 

 and diagram of a purported survey in which they make chords of 300 

 feet — something no surveyor would think of doing — turning off 30° at 

 the end of each, with the angle on the embankment, thus completing 

 their figure with twelve chords. This would form a dodecagon with a 

 perimeter of 3600 feet, which they have inscribed within a "circle" hav- 

 ing, according to their figures, a diameter of 800 or a circumference of 

 2513 feet. 



It is not the province of this article to criticise; but when such mis- 

 takes are made the foundation of a science, it is well to present them in 

 their proper light. 



The minor errors due to hasty observation, or incorrect deductions, 

 are numberless. Mounds are almost invariably represented much steeper 

 than they are. Nearly all cuts of the large mound at Marietta show its 

 height to be equal to its breadth; sometimes it is shown as perpendicular 

 for fully ten feet up its sides. But the base diameter of a mound undis- 

 turbed by cultivation is never less than four times and from that to ten 



