216 THE ESSEX NATURALIST.
One of these banks, about 100 yards (not actually measured)
north-east of Mound A, looks very like a pillow mound, and may
possibly be of that character,
I have approximately sketched in the footpaths, some of
which may be of considerable antiquity, and also drawn in lines
of latitude and longitude as nearly as I can determine them.
The outlines of the mounds as shown on the sketch map re-
present the ditches surrounding the mounds as these appear
to-day. As will be seen in the sequel, the mounds were originally
much narrower, but have, spread by weathering.
The excavations are represented by round dots, with the
outlines broken where the ditches were trenched.
The Rabbit-Holes. The mounds are honeycombed with
burrows, which render exact archaeological work peculiarly
difficult. The usual method followed by the rabbits is to go
down to about 4 to 5 feet below the top of the mounds, and then
run horizontal tunnels in the Bagshot Sand, so as to leave a thin
roof of that sand above them. They continue to use the same
burrows for long periods, and only at rare intervals undertake
the labour of starting new ones.
Most of the disused holes are filled with a finely stratified
silting, which is very distinctive and easily recognizable, but
some merely fall in and become filled with mixed soil, and these
are more difficult to realise.
The burrows contain a large amount of the debris of the
present-day tripper, and many quite new objects occur down to
the 4 to 5 foot level. I found no object of medieval date in
them.
The Method of Levelling.—The method of levelling down-
wards from an imaginary datum line situated in the air, above the
surface, always appeals to me as being peculiarly unsatisfactory.
At least I always find that information published upon this
plan needs to be converted to a different basis before it is useful.
Two methods are useful, namely, that based on the Ordnance
Datum, and that based on the contour of the natural surface,
to show the artificial work done in digging or building up.
For the sections of these mounds I set the sight line of
my instrument to the contour of the natural surface, and
plotted the artificial work as plus or minus with reference to
that line as a datum.