26 ANCIENT MONUMENTS. 



the Miami river, which, at the period of the erection of the work, must have 

 discharged itself into the Ohio at a point much lower down than it now does. I 

 have never been able to discover the eastern wall of the enclosure ; but if its 

 direction from the citadel to the Ohio was such as it should have been, to embrace 

 the largest space with the least labor, there could not have been less than three 

 hundred acres enclosed."* 



PLATE TX. No. .3. 



ANCIENT WORK NEAR LEXINGTON, KENTUCKY. 



[ From the Rafinesque MSS. ] 



This work is situated at the junction of the Town and South forks of the Elk- 

 horn river, seven miles from the town of Lexington, Kentucky. Its character is 

 sufficiently explained by the engraving. It is entirely singular in having a stream, 

 of considerable size, running through it. The river has probably encroached upon 

 its original proportions. About one hundred yards to the eastward of this work 

 is a small, oblong enclosure, and a large, elliptical, truncated mound. Other 

 mounds and enclosures occur in the vicinity .f 



PLATE X. 



Clark's work ; north fork of paint creek.j; 



The work here presented is one of the largest and most interesting in the Scioto 

 valley. It has many of the characteristics of a work of defence, and is accordingly 

 classified as such, although differing in position and some other respects from the 

 entrenched hills just described. The minor works which it encloses, or which 

 are in combination with it, are manifestly of a diflferent character, probably religious 



* Transactions Historical Society of Oliio, vol. i. p. 225. 



f Tliis work is not placed in the connection which it was designed to occupy. Its position in the text 

 was determined by circumstances ; and its character will be better understood in the progress of this 

 chapter. 



I Tliis plan is from an original, minute survey by the authors. A plan and description of the same 

 work were published by Mr. Atwater in the " Archseologia Americana." It will be found to differ in 

 some important respects. 



