Epping Forest had no permanent internal divisions such as still exist in
Hatfield and Writtle Forests. The named "woods" such as Bury Wood and
Hawk Wood are merely ill-defined topographical names and are not discrete
woods; their boundaries are not marked on the ground and do not correspond
to differences of management.
THE EPPING AREA BEFORE THE FOREST
As all naturalists know, Lowland England used to be covered almost en-
tirely by the trees of the prehistoric forest or wildwood, which developed at the
end of the last glaciation (about 10,000 B. C.) and has left a record of its history
in pollen deposits. The wildwood was destroyed in order to create our present
farmland, a process begun in the Neolithic (c. 4,000 B.C.).
Most writers on Epping Forest state that it is derived from a remnant of
wildwood, a remnant which was much larger in past centuries and in the early
middle ages covered the whole of S.W. Essex. The belief in a larger Forest is
based partly on a confusion between the legal and the physical Forest, and
partly on a general belief that Norman England was still a very wooded land.
Iron Age and Roman (Fig. 6)
The oldest witnesses against this traditional view are Ambresbury Banks
and Loughton Camp, great defensive earthworks of Iron Age or early Romano-
British antiquity in what is now Epping Forest. Their function as hillforts
presupposes that their immediate surroundings and perhaps also their distant
views were not obstructed (as they are now) by trees which might cover an
enemy. As early as 2,000 years ago the area was not an undisturbed wilderness
but had sufficient population or communications to be worth defending.
In Roman times the main road from London to Dunmow passed between
Epping and Hainault Forests, and on it was an important settlement (at least a
large villa) at Woolston in Chigwell. Other settlements are known at Wanstead,
Leyton, and possibly Theydon Bois (59). There are minor finds from Havering
and Woodford. Although most of this evidence comes from early and in-
complete excavations, it shows that S.W. Essex had a quite dense Romano-
British population, with many settlements and presumably agriculture around
and between what were to be Epping and Hainault Forests.
Anglo-Saxon
The place-names indicate that this is not an area of particularly late
colonization. The many names in -ing, -ham, and -ton tell of settlements
formed, well back into the Anglo-Saxon period, in what may well have been the
surviving Roman agricultural countryside. There are only a few minor names in
23