36 PAPERS, ETC. 
usually following the form of the ground, and presenting 
no very marked features of any fixed system of castrame- 
tation. DBesides these military works and sepulchral 
barrows of various dates and constructions, the earth- 
works generally met with are either Iynchets, small 
enclosures, almost invariably to be traced in the imme- 
diate vieinity of British towns and villages :—British track- 
ways, usually running between two low banks :—Roman 
roads, stretehing in straight lines over hill and dale, with 
a disregard of obstacles worthy of their invineible con- 
structors ; and boundary lines, which are in most cases 
high banks, with a ditch on one side. There are two other 
kinds of eartlı-works, occasionally met with, which are often 
mistaken for military. One of these consists of a small 
square area, surrounded by a moat and rampart, which is 
frequently supposed to be a Roman camp; this, however, 
from its small size, and the absence of the four regular gates, 
can hardly be so, but is, more probably, the site of a 
Saxon mansion, the wooden buildings of which have 
entirely disappeared. The other is a large, irregular en- 
closure, having the mote on the outside, and is probably a 
very ancient park. Besides these, we often find, on the 
steep side of a down, a series of terraces evidently artificial. 
These are generally considered to be the traces of early 
agriculture, and, in some instances, this may no doubt be 
the case. But from their analogy to the platforms before 
mentioned, as of general occurrence in British fortifications, 
and from the strength of the situations in which they are 
often found, added to the fact of their being so frequently 
met with in those distriets which we have reason to believe 
were for centuries the battle-ground of the Belge and 
original British tribes, Iam led to think that some, if not 
the greater number of them, are military field-works, and 
