48 A N C I E N T M N U M E N T S . 



investigation and observation has tended to confirm, that most, perhaps all, of the 

 earthworks not manifestly defensive in their character, were in some way 

 connected with the superstitious rites of the builders, — though in what precise 

 manner, it is, and perhaps ever will be, impossible satisfactorily to determine. 



The general character of these works has already been briefly indicated. They 

 are mostly regular in thdr structure, and occupy the broad and level river bot- 

 toms, seldom occurring upon the table lands or where the surface of the ground 

 is undulating or broken. They are usually square or circular in form ; sometimes 

 they are slightly elliptical. Occasionally we find them isolated, but more frequently 

 in groups. The greater number of the circles are of small size, with a nearly 

 uniform diameter of two hundred and fifty or three hundred feet, and invariably 

 have the ditch interior to the Avail. These have always a single gateway, opening 

 oftenest to the east, though by no means observing a fixed rule in that res{)ect. 

 It frequently happens that they have one or more small mounds, of the class 

 denominated sacrificial, within the walls. These small circles occasionally occur 

 within larger works of a different character. Apart from these, numerous little 

 circles, froia thirty to fifty feet in diameter, are observed in the vicinity of large 

 works. They consist of very slight embankments of earth, and have no entrances or 

 passage ways. It has been suggested that these are the remains of ancient lodges 

 or buildings. The accounts which we have of the traces left of the huts of the 

 Mandans and other Indians, at their deserted villages, render this supposition not 

 improbable. It sometimes happens that we find small circles embracing large 

 mounds : these can hardly be regarded as of the same character with that numer- 

 ous class already noticed. 



The larger circles are oftenest found in combination wath rectangular works, 

 connected with them directly, or by avenues. Some of these circles are of great 

 extent, embracing fifty or more acres. They seldom have a ditch ; but whenever it 

 occurs, it is interior to the wall. As in the case of the square or rectangular works 

 to w Inch they are attached, (and which, it is believed, never have ditches, exterior or 

 interior,) the walls are usually composed of earth taken up evenly from the surface, 

 or from large pits in the neighborhood. Evident care appears in all cases to have 

 been exercised, in procuring the material, to preserve the surface of the adjacent 

 plain smooth, and as far as possible unbroken. This fact is in itself almost conclu- 

 sive against the supposition of a defensive design, especially as we have abundant 

 evidence that the mound-builders understood perfectly the value of the external 

 fosse in their works of defence. The Malls of these works are, for the most part, 

 comparatively slight, varying from three to seven feet in height. Sometimes they 

 are quite imposing ; as in the case of the great circle at Newark, Licking county, 

 Ohio; where, at the entrance, the wall from the bottom of the ditch has a vertical 

 height of not far from thirty feet. The square or rectangular works, a1 tending 

 these large circles, are of various dimensions. It has been observed, however, 

 that certain groups are marked by a great uniformity oT size. Five or six of these 

 arc noticed in the succeeding pages ; they are exact squares, each measuring one 

 thousand and eighty feot side, — a coincidence which could not possibly be acci- 

 dental, and which must possess some significance. It certainly establishes the 



