December 10, 1886. 



SCIENCE. 



539 



broken througli by the pressiire of -w'ater accumu- 

 lated behind it, but in all these cases it is apparent 

 that the ravines have been formed since the wall was 

 built. At only one jDoint did we observe a break 

 made since Professor Locke's survey. This is 

 through the long, curved stretch directly east from 

 where the so-called ' two large mounds ' are repre- 

 sented on the plat. 



If these ravines were defended, as is quite prob- 

 able, it must have been by some other means than a 

 wall of earth, which could not have withstood the 

 pressure through a single rainy season. 



Although the wall is built chiefly of earth (com- 

 posed largely in most places of clay) gathered from 

 the adjacent surface, and from the interior ditch 

 where it exists, it is partially underlaid at numerous 

 points with stones, which in some cases were laid up 

 loosely. This was noticed at the north-western 

 corner, where the wall has been cut through to make 

 way for the turnpike, and also at the extreme south- 

 eastern corner. At almost every point where a slight 

 cut has been made for a farm-road or other purpose, 

 stones were observed. 



generally crossed at the upper terminus by a wall of 

 the ordinary height, the ridge immediately outside 

 being cut down several feet so as to i^resent a steep 

 slope corresponding with the outside of the -wall. 

 This gives the appearance of a terrace on the hillside 

 a few feet below the wall. On the other hand, 

 where similar ridges form approaches to the south 

 portion, and also, at some places, to the north 

 portion, the defences are formed by raising the wall 

 considerably above the ordinary height. 



The isthmus, or point where the opposite walls ap- 

 proach nearest to each other, just north of the so- 

 called ' two large mounds,' is undefended, though on 

 the right or east side the ascent is by no means diffi- 

 cult : the declivity on the west forms a sufficient 

 defence without a wall. The plat at this point is 

 slightly erroneous, as the wall on the west side does 

 not extend quite so far north as represented. It is 

 possible that this extension was made theoretically, 

 on the su^Dposition that the wash which is apparent 

 here (shown in Atwater's figure) had carried away 

 the wall. That a small portion of the extreme end 

 was carried down, is true, but the ridge on which it 





WALL OF FORT ANCIENT. 



Mr. George Bidge, who lives near the two mounds 

 at the north-eastern corner, and who has for years 

 studied the fort, insists that the wall is to a consider- 

 able extent underlaid with stone. This fact is also 

 mentioned by Squier and Davis, who state that " they 

 are water-worn, and seem for the most part to have 

 been taken from the river." This is certainly an 

 error, as thej' are almost entirely of flat pieces of 

 limestone, showing no indications of having been 

 water-worn, such as could be obtained on the surface 

 or immediately below the brow of the hill. 



The two points at the isthmus, or neck, marked on 

 the plat ' two large mounds,^ are not ' mounds ' 

 properly so called, but the elevated terminations of 

 the walls on the sides, the opening here being an im- 

 portant gateway. The point at the extreme south- 

 eastern corner, marked on the plat ' mound,^ is only 

 an elevated portion of the wall thrown up to defend 

 an easy approach at this point. 



One of the most interesting facts observed, of 

 which mention has not heretofore been made, is the 

 different methods adopted of defending the more 

 easy approaches. On the north, these approaches, 

 which are usually narrow, ascending ridges, are 



runs never crossed the gap. Besides, in the original 

 plat, as given in the ' Portfolio,' the wall is represent- 

 ed as extending up to the so-called 'wash' (which is not 

 a ' wash,' but a small land-slide), and stopping there. 

 The wall never existed along the top at this point. 



The parallel walls starting out from the two mounds 

 near the north-eastern corner, represented in ' Sup- 

 plementary plan A,' Squier and Davis's figure, are 

 entirely obliterated except at the fence crossings, 

 where slight traces of them are visible. The in- 

 cluded mound at the east end is yet distinctly visible. 

 Mr. Bidge informed us that he has discovered, at a 

 depth of about eighteen inches, a pavement of stone 

 reaching from wall to wall, and from the mounds 

 eastward over a hundred yards. We had an oppor- 

 tunity of inspecting this at only one i^oint, and know 

 nothing further in regard to it than his statement, 

 which I believe to be trustworthy. 



Some of the problems presented by this work are 

 very difficult to solve, though others can be, in a 

 measure at least, satisfactorily determined without re- 

 sort to mere speculation. 



That it waf built and intended as a work of de- 

 fence is so apparent, that it is scarcely possible there 



